Annotation of imach/src/imach.c, revision 1.367
1.367 ! brouard 1: /* $Id: imach.c,v 1.366 2024/07/02 09:42:10 brouard Exp $
1.126 brouard 2: $State: Exp $
1.360 brouard 3: $Log: imach.c,v $
1.367 ! brouard 4: Revision 1.366 2024/07/02 09:42:10 brouard
! 5: Summary: trying clang on Linux
! 6:
1.366 brouard 7: Revision 1.365 2024/06/28 13:53:38 brouard
8: * imach.c (Module): fixing some bugs in gnuplot and quantitative variables, but not completely solved
9:
1.365 brouard 10: Revision 1.364 2024/06/28 12:27:05 brouard
11: * imach.c (Module): fixing some bugs in gnuplot and quantitative variables, but not completely solved
12:
1.364 brouard 13: Revision 1.363 2024/06/28 09:31:55 brouard
14: Summary: Adding log lines too
15:
1.363 brouard 16: Revision 1.362 2024/06/28 08:00:31 brouard
17: Summary: 0.99s6
18:
19: * imach.c (Module): s6 errors with age*age (harmless).
20:
1.362 brouard 21: Revision 1.361 2024/05/12 20:29:32 brouard
22: Summary: Version 0.99s5
23:
24: * src/imach.c Version 0.99s5 In fact, the covariance of total life
25: expectancy e.. with a partial life expectancy e.j is high,
26: therefore the complete matrix of variance covariance has to be
27: included in the formula of the standard error of the proportion of
28: total life expectancy spent in a specific state:
29: var(X/Y)=mu_x^2/mu_y^2*(sigma_x^2/mu_x^2 -2
30: sigma_xy/mu_x/mu_y+sigma^2/mu_y^2). Also an error with mle=-3
31: made the program core dump. It is fixed in this version.
32:
1.361 brouard 33: Revision 1.360 2024/04/30 10:59:22 brouard
34: Summary: Version 0.99s4 and estimation of std of e.j/e..
35:
1.360 brouard 36: Revision 1.359 2024/04/24 21:21:17 brouard
37: Summary: First IMaCh version using Brent Praxis software based on Buckhardt and Gegenfürtner C codes
38:
1.359 brouard 39: Revision 1.6 2024/04/24 21:10:29 brouard
40: Summary: First IMaCh version using Brent Praxis software based on Buckhardt and Gegenfürtner C codes
1.358 brouard 41:
1.359 brouard 42: Revision 1.5 2023/10/09 09:10:01 brouard
43: Summary: trying to reconsider
1.357 brouard 44:
1.359 brouard 45: Revision 1.4 2023/06/22 12:50:51 brouard
46: Summary: stil on going
1.357 brouard 47:
1.359 brouard 48: Revision 1.3 2023/06/22 11:28:07 brouard
49: *** empty log message ***
1.356 brouard 50:
1.359 brouard 51: Revision 1.2 2023/06/22 11:22:40 brouard
52: Summary: with svd but not working yet
1.355 brouard 53:
1.354 brouard 54: Revision 1.353 2023/05/08 18:48:22 brouard
55: *** empty log message ***
56:
1.353 brouard 57: Revision 1.352 2023/04/29 10:46:21 brouard
58: *** empty log message ***
59:
1.352 brouard 60: Revision 1.351 2023/04/29 10:43:47 brouard
61: Summary: 099r45
62:
1.351 brouard 63: Revision 1.350 2023/04/24 11:38:06 brouard
64: *** empty log message ***
65:
1.350 brouard 66: Revision 1.349 2023/01/31 09:19:37 brouard
67: Summary: Improvements in models with age*Vn*Vm
68:
1.348 brouard 69: Revision 1.347 2022/09/18 14:36:44 brouard
70: Summary: version 0.99r42
71:
1.347 brouard 72: Revision 1.346 2022/09/16 13:52:36 brouard
73: * src/imach.c (Module): 0.99r41 Was an error when product of timevarying and fixed. Using FixedV[of name] now. Thank you Feinuo
74:
1.346 brouard 75: Revision 1.345 2022/09/16 13:40:11 brouard
76: Summary: Version 0.99r41
77:
78: * imach.c (Module): 0.99r41 Was an error when product of timevarying and fixed. Using FixedV[of name] now. Thank you Feinuo
79:
1.345 brouard 80: Revision 1.344 2022/09/14 19:33:30 brouard
81: Summary: version 0.99r40
82:
83: * imach.c (Module): Fixing names of variables in T_ (thanks to Feinuo)
84:
1.344 brouard 85: Revision 1.343 2022/09/14 14:22:16 brouard
86: Summary: version 0.99r39
87:
88: * imach.c (Module): Version 0.99r39 with colored dummy covariates
89: (fixed or time varying), using new last columns of
90: ILK_parameter.txt file.
91:
1.343 brouard 92: Revision 1.342 2022/09/11 19:54:09 brouard
93: Summary: 0.99r38
94:
95: * imach.c (Module): Adding timevarying products of any kinds,
96: should work before shifting cotvar from ncovcol+nqv columns in
97: order to have a correspondance between the column of cotvar and
98: the id of column.
99: (Module): Some cleaning and adding covariates in ILK.txt
100:
1.342 brouard 101: Revision 1.341 2022/09/11 07:58:42 brouard
102: Summary: Version 0.99r38
103:
104: After adding change in cotvar.
105:
1.341 brouard 106: Revision 1.340 2022/09/11 07:53:11 brouard
107: Summary: Version imach 0.99r37
108:
109: * imach.c (Module): Adding timevarying products of any kinds,
110: should work before shifting cotvar from ncovcol+nqv columns in
111: order to have a correspondance between the column of cotvar and
112: the id of column.
113:
1.340 brouard 114: Revision 1.339 2022/09/09 17:55:22 brouard
115: Summary: version 0.99r37
116:
117: * imach.c (Module): Many improvements for fixing products of fixed
118: timevarying as well as fixed * fixed, and test with quantitative
119: covariate.
120:
1.339 brouard 121: Revision 1.338 2022/09/04 17:40:33 brouard
122: Summary: 0.99r36
123:
124: * imach.c (Module): Now the easy runs i.e. without result or
125: model=1+age only did not work. The defautl combination should be 1
126: and not 0 because everything hasn't been tranformed yet.
127:
1.338 brouard 128: Revision 1.337 2022/09/02 14:26:02 brouard
129: Summary: version 0.99r35
130:
131: * src/imach.c: Version 0.99r35 because it outputs same results with
132: 1+age+V1+V1*age for females and 1+age for females only
133: (education=1 noweight)
134:
1.337 brouard 135: Revision 1.336 2022/08/31 09:52:36 brouard
136: *** empty log message ***
137:
1.336 brouard 138: Revision 1.335 2022/08/31 08:23:16 brouard
139: Summary: improvements...
140:
1.335 brouard 141: Revision 1.334 2022/08/25 09:08:41 brouard
142: Summary: In progress for quantitative
143:
1.334 brouard 144: Revision 1.333 2022/08/21 09:10:30 brouard
145: * src/imach.c (Module): Version 0.99r33 A lot of changes in
146: reassigning covariates: my first idea was that people will always
147: use the first covariate V1 into the model but in fact they are
148: producing data with many covariates and can use an equation model
149: with some of the covariate; it means that in a model V2+V3 instead
150: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
151: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
152: the equation model is restricted to two variables only (V2, V3)
153: and the combination for V2 should be codtabm(k,1) instead of
154: (codtabm(k,2), and the code should be
155: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
156: made. All of these should be simplified once a day like we did in
157: hpxij() for example by using precov[nres] which is computed in
158: decoderesult for each nres of each resultline. Loop should be done
159: on the equation model globally by distinguishing only product with
160: age (which are changing with age) and no more on type of
161: covariates, single dummies, single covariates.
162:
1.333 brouard 163: Revision 1.332 2022/08/21 09:06:25 brouard
164: Summary: Version 0.99r33
165:
166: * src/imach.c (Module): Version 0.99r33 A lot of changes in
167: reassigning covariates: my first idea was that people will always
168: use the first covariate V1 into the model but in fact they are
169: producing data with many covariates and can use an equation model
170: with some of the covariate; it means that in a model V2+V3 instead
171: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
172: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
173: the equation model is restricted to two variables only (V2, V3)
174: and the combination for V2 should be codtabm(k,1) instead of
175: (codtabm(k,2), and the code should be
176: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
177: made. All of these should be simplified once a day like we did in
178: hpxij() for example by using precov[nres] which is computed in
179: decoderesult for each nres of each resultline. Loop should be done
180: on the equation model globally by distinguishing only product with
181: age (which are changing with age) and no more on type of
182: covariates, single dummies, single covariates.
183:
1.332 brouard 184: Revision 1.331 2022/08/07 05:40:09 brouard
185: *** empty log message ***
186:
1.331 brouard 187: Revision 1.330 2022/08/06 07:18:25 brouard
188: Summary: last 0.99r31
189:
190: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function
191:
1.330 brouard 192: Revision 1.329 2022/08/03 17:29:54 brouard
193: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
194:
1.329 brouard 195: Revision 1.328 2022/07/27 17:40:48 brouard
196: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
197:
1.328 brouard 198: Revision 1.327 2022/07/27 14:47:35 brouard
199: Summary: Still a problem for one-step probabilities in case of quantitative variables
200:
1.327 brouard 201: Revision 1.326 2022/07/26 17:33:55 brouard
202: Summary: some test with nres=1
203:
1.326 brouard 204: Revision 1.325 2022/07/25 14:27:23 brouard
205: Summary: r30
206:
207: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
208: coredumped, revealed by Feiuno, thank you.
209:
1.325 brouard 210: Revision 1.324 2022/07/23 17:44:26 brouard
211: *** empty log message ***
212:
1.324 brouard 213: Revision 1.323 2022/07/22 12:30:08 brouard
214: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
215:
1.323 brouard 216: Revision 1.322 2022/07/22 12:27:48 brouard
217: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
218:
1.322 brouard 219: Revision 1.321 2022/07/22 12:04:24 brouard
220: Summary: r28
221:
222: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
223:
1.321 brouard 224: Revision 1.320 2022/06/02 05:10:11 brouard
225: *** empty log message ***
226:
1.320 brouard 227: Revision 1.319 2022/06/02 04:45:11 brouard
228: * imach.c (Module): Adding the Wald tests from the log to the main
229: htm for better display of the maximum likelihood estimators.
230:
1.319 brouard 231: Revision 1.318 2022/05/24 08:10:59 brouard
232: * imach.c (Module): Some attempts to find a bug of wrong estimates
233: of confidencce intervals with product in the equation modelC
234:
1.318 brouard 235: Revision 1.317 2022/05/15 15:06:23 brouard
236: * imach.c (Module): Some minor improvements
237:
1.317 brouard 238: Revision 1.316 2022/05/11 15:11:31 brouard
239: Summary: r27
240:
1.316 brouard 241: Revision 1.315 2022/05/11 15:06:32 brouard
242: *** empty log message ***
243:
1.315 brouard 244: Revision 1.314 2022/04/13 17:43:09 brouard
245: * imach.c (Module): Adding link to text data files
246:
1.314 brouard 247: Revision 1.313 2022/04/11 15:57:42 brouard
248: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
249:
1.313 brouard 250: Revision 1.312 2022/04/05 21:24:39 brouard
251: *** empty log message ***
252:
1.312 brouard 253: Revision 1.311 2022/04/05 21:03:51 brouard
254: Summary: Fixed quantitative covariates
255:
256: Fixed covariates (dummy or quantitative)
257: with missing values have never been allowed but are ERRORS and
258: program quits. Standard deviations of fixed covariates were
259: wrongly computed. Mean and standard deviations of time varying
260: covariates are still not computed.
261:
1.311 brouard 262: Revision 1.310 2022/03/17 08:45:53 brouard
263: Summary: 99r25
264:
265: Improving detection of errors: result lines should be compatible with
266: the model.
267:
1.310 brouard 268: Revision 1.309 2021/05/20 12:39:14 brouard
269: Summary: Version 0.99r24
270:
1.309 brouard 271: Revision 1.308 2021/03/31 13:11:57 brouard
272: Summary: Version 0.99r23
273:
274:
275: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
276:
1.308 brouard 277: Revision 1.307 2021/03/08 18:11:32 brouard
278: Summary: 0.99r22 fixed bug on result:
279:
1.307 brouard 280: Revision 1.306 2021/02/20 15:44:02 brouard
281: Summary: Version 0.99r21
282:
283: * imach.c (Module): Fix bug on quitting after result lines!
284: (Module): Version 0.99r21
285:
1.306 brouard 286: Revision 1.305 2021/02/20 15:28:30 brouard
287: * imach.c (Module): Fix bug on quitting after result lines!
288:
1.305 brouard 289: Revision 1.304 2021/02/12 11:34:20 brouard
290: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
291:
1.304 brouard 292: Revision 1.303 2021/02/11 19:50:15 brouard
293: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
294:
1.303 brouard 295: Revision 1.302 2020/02/22 21:00:05 brouard
296: * (Module): imach.c Update mle=-3 (for computing Life expectancy
297: and life table from the data without any state)
298:
1.302 brouard 299: Revision 1.301 2019/06/04 13:51:20 brouard
300: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
301:
1.301 brouard 302: Revision 1.300 2019/05/22 19:09:45 brouard
303: Summary: version 0.99r19 of May 2019
304:
1.300 brouard 305: Revision 1.299 2019/05/22 18:37:08 brouard
306: Summary: Cleaned 0.99r19
307:
1.299 brouard 308: Revision 1.298 2019/05/22 18:19:56 brouard
309: *** empty log message ***
310:
1.298 brouard 311: Revision 1.297 2019/05/22 17:56:10 brouard
312: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
313:
1.297 brouard 314: Revision 1.296 2019/05/20 13:03:18 brouard
315: Summary: Projection syntax simplified
316:
317:
318: We can now start projections, forward or backward, from the mean date
319: of inteviews up to or down to a number of years of projection:
320: prevforecast=1 yearsfproj=15.3 mobil_average=0
321: or
322: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
323: or
324: prevbackcast=1 yearsbproj=12.3 mobil_average=1
325: or
326: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
327:
1.296 brouard 328: Revision 1.295 2019/05/18 09:52:50 brouard
329: Summary: doxygen tex bug
330:
1.295 brouard 331: Revision 1.294 2019/05/16 14:54:33 brouard
332: Summary: There was some wrong lines added
333:
1.294 brouard 334: Revision 1.293 2019/05/09 15:17:34 brouard
335: *** empty log message ***
336:
1.293 brouard 337: Revision 1.292 2019/05/09 14:17:20 brouard
338: Summary: Some updates
339:
1.292 brouard 340: Revision 1.291 2019/05/09 13:44:18 brouard
341: Summary: Before ncovmax
342:
1.291 brouard 343: Revision 1.290 2019/05/09 13:39:37 brouard
344: Summary: 0.99r18 unlimited number of individuals
345:
346: The number n which was limited to 20,000 cases is now unlimited, from firstobs to lastobs. If the number is too for the virtual memory, probably an error will occur.
347:
1.290 brouard 348: Revision 1.289 2018/12/13 09:16:26 brouard
349: Summary: Bug for young ages (<-30) will be in r17
350:
1.289 brouard 351: Revision 1.288 2018/05/02 20:58:27 brouard
352: Summary: Some bugs fixed
353:
1.288 brouard 354: Revision 1.287 2018/05/01 17:57:25 brouard
355: Summary: Bug fixed by providing frequencies only for non missing covariates
356:
1.287 brouard 357: Revision 1.286 2018/04/27 14:27:04 brouard
358: Summary: some minor bugs
359:
1.286 brouard 360: Revision 1.285 2018/04/21 21:02:16 brouard
361: Summary: Some bugs fixed, valgrind tested
362:
1.285 brouard 363: Revision 1.284 2018/04/20 05:22:13 brouard
364: Summary: Computing mean and stdeviation of fixed quantitative variables
365:
1.284 brouard 366: Revision 1.283 2018/04/19 14:49:16 brouard
367: Summary: Some minor bugs fixed
368:
1.283 brouard 369: Revision 1.282 2018/02/27 22:50:02 brouard
370: *** empty log message ***
371:
1.282 brouard 372: Revision 1.281 2018/02/27 19:25:23 brouard
373: Summary: Adding second argument for quitting
374:
1.281 brouard 375: Revision 1.280 2018/02/21 07:58:13 brouard
376: Summary: 0.99r15
377:
378: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
379:
1.280 brouard 380: Revision 1.279 2017/07/20 13:35:01 brouard
381: Summary: temporary working
382:
1.279 brouard 383: Revision 1.278 2017/07/19 14:09:02 brouard
384: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
385:
1.278 brouard 386: Revision 1.277 2017/07/17 08:53:49 brouard
387: Summary: BOM files can be read now
388:
1.277 brouard 389: Revision 1.276 2017/06/30 15:48:31 brouard
390: Summary: Graphs improvements
391:
1.276 brouard 392: Revision 1.275 2017/06/30 13:39:33 brouard
393: Summary: Saito's color
394:
1.275 brouard 395: Revision 1.274 2017/06/29 09:47:08 brouard
396: Summary: Version 0.99r14
397:
1.274 brouard 398: Revision 1.273 2017/06/27 11:06:02 brouard
399: Summary: More documentation on projections
400:
1.273 brouard 401: Revision 1.272 2017/06/27 10:22:40 brouard
402: Summary: Color of backprojection changed from 6 to 5(yellow)
403:
1.272 brouard 404: Revision 1.271 2017/06/27 10:17:50 brouard
405: Summary: Some bug with rint
406:
1.271 brouard 407: Revision 1.270 2017/05/24 05:45:29 brouard
408: *** empty log message ***
409:
1.270 brouard 410: Revision 1.269 2017/05/23 08:39:25 brouard
411: Summary: Code into subroutine, cleanings
412:
1.269 brouard 413: Revision 1.268 2017/05/18 20:09:32 brouard
414: Summary: backprojection and confidence intervals of backprevalence
415:
1.268 brouard 416: Revision 1.267 2017/05/13 10:25:05 brouard
417: Summary: temporary save for backprojection
418:
1.267 brouard 419: Revision 1.266 2017/05/13 07:26:12 brouard
420: Summary: Version 0.99r13 (improvements and bugs fixed)
421:
1.266 brouard 422: Revision 1.265 2017/04/26 16:22:11 brouard
423: Summary: imach 0.99r13 Some bugs fixed
424:
1.265 brouard 425: Revision 1.264 2017/04/26 06:01:29 brouard
426: Summary: Labels in graphs
427:
1.264 brouard 428: Revision 1.263 2017/04/24 15:23:15 brouard
429: Summary: to save
430:
1.263 brouard 431: Revision 1.262 2017/04/18 16:48:12 brouard
432: *** empty log message ***
433:
1.262 brouard 434: Revision 1.261 2017/04/05 10:14:09 brouard
435: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
436:
1.261 brouard 437: Revision 1.260 2017/04/04 17:46:59 brouard
438: Summary: Gnuplot indexations fixed (humm)
439:
1.260 brouard 440: Revision 1.259 2017/04/04 13:01:16 brouard
441: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
442:
1.259 brouard 443: Revision 1.258 2017/04/03 10:17:47 brouard
444: Summary: Version 0.99r12
445:
446: Some cleanings, conformed with updated documentation.
447:
1.258 brouard 448: Revision 1.257 2017/03/29 16:53:30 brouard
449: Summary: Temp
450:
1.257 brouard 451: Revision 1.256 2017/03/27 05:50:23 brouard
452: Summary: Temporary
453:
1.256 brouard 454: Revision 1.255 2017/03/08 16:02:28 brouard
455: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
456:
1.255 brouard 457: Revision 1.254 2017/03/08 07:13:00 brouard
458: Summary: Fixing data parameter line
459:
1.254 brouard 460: Revision 1.253 2016/12/15 11:59:41 brouard
461: Summary: 0.99 in progress
462:
1.253 brouard 463: Revision 1.252 2016/09/15 21:15:37 brouard
464: *** empty log message ***
465:
1.252 brouard 466: Revision 1.251 2016/09/15 15:01:13 brouard
467: Summary: not working
468:
1.251 brouard 469: Revision 1.250 2016/09/08 16:07:27 brouard
470: Summary: continue
471:
1.250 brouard 472: Revision 1.249 2016/09/07 17:14:18 brouard
473: Summary: Starting values from frequencies
474:
1.249 brouard 475: Revision 1.248 2016/09/07 14:10:18 brouard
476: *** empty log message ***
477:
1.248 brouard 478: Revision 1.247 2016/09/02 11:11:21 brouard
479: *** empty log message ***
480:
1.247 brouard 481: Revision 1.246 2016/09/02 08:49:22 brouard
482: *** empty log message ***
483:
1.246 brouard 484: Revision 1.245 2016/09/02 07:25:01 brouard
485: *** empty log message ***
486:
1.245 brouard 487: Revision 1.244 2016/09/02 07:17:34 brouard
488: *** empty log message ***
489:
1.244 brouard 490: Revision 1.243 2016/09/02 06:45:35 brouard
491: *** empty log message ***
492:
1.243 brouard 493: Revision 1.242 2016/08/30 15:01:20 brouard
494: Summary: Fixing a lots
495:
1.242 brouard 496: Revision 1.241 2016/08/29 17:17:25 brouard
497: Summary: gnuplot problem in Back projection to fix
498:
1.241 brouard 499: Revision 1.240 2016/08/29 07:53:18 brouard
500: Summary: Better
501:
1.240 brouard 502: Revision 1.239 2016/08/26 15:51:03 brouard
503: Summary: Improvement in Powell output in order to copy and paste
504:
505: Author:
506:
1.239 brouard 507: Revision 1.238 2016/08/26 14:23:35 brouard
508: Summary: Starting tests of 0.99
509:
1.238 brouard 510: Revision 1.237 2016/08/26 09:20:19 brouard
511: Summary: to valgrind
512:
1.237 brouard 513: Revision 1.236 2016/08/25 10:50:18 brouard
514: *** empty log message ***
515:
1.236 brouard 516: Revision 1.235 2016/08/25 06:59:23 brouard
517: *** empty log message ***
518:
1.235 brouard 519: Revision 1.234 2016/08/23 16:51:20 brouard
520: *** empty log message ***
521:
1.234 brouard 522: Revision 1.233 2016/08/23 07:40:50 brouard
523: Summary: not working
524:
1.233 brouard 525: Revision 1.232 2016/08/22 14:20:21 brouard
526: Summary: not working
527:
1.232 brouard 528: Revision 1.231 2016/08/22 07:17:15 brouard
529: Summary: not working
530:
1.231 brouard 531: Revision 1.230 2016/08/22 06:55:53 brouard
532: Summary: Not working
533:
1.230 brouard 534: Revision 1.229 2016/07/23 09:45:53 brouard
535: Summary: Completing for func too
536:
1.229 brouard 537: Revision 1.228 2016/07/22 17:45:30 brouard
538: Summary: Fixing some arrays, still debugging
539:
1.227 brouard 540: Revision 1.226 2016/07/12 18:42:34 brouard
541: Summary: temp
542:
1.226 brouard 543: Revision 1.225 2016/07/12 08:40:03 brouard
544: Summary: saving but not running
545:
1.225 brouard 546: Revision 1.224 2016/07/01 13:16:01 brouard
547: Summary: Fixes
548:
1.224 brouard 549: Revision 1.223 2016/02/19 09:23:35 brouard
550: Summary: temporary
551:
1.223 brouard 552: Revision 1.222 2016/02/17 08:14:50 brouard
553: Summary: Probably last 0.98 stable version 0.98r6
554:
1.222 brouard 555: Revision 1.221 2016/02/15 23:35:36 brouard
556: Summary: minor bug
557:
1.220 brouard 558: Revision 1.219 2016/02/15 00:48:12 brouard
559: *** empty log message ***
560:
1.219 brouard 561: Revision 1.218 2016/02/12 11:29:23 brouard
562: Summary: 0.99 Back projections
563:
1.218 brouard 564: Revision 1.217 2015/12/23 17:18:31 brouard
565: Summary: Experimental backcast
566:
1.217 brouard 567: Revision 1.216 2015/12/18 17:32:11 brouard
568: Summary: 0.98r4 Warning and status=-2
569:
570: Version 0.98r4 is now:
571: - displaying an error when status is -1, date of interview unknown and date of death known;
572: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
573: Older changes concerning s=-2, dating from 2005 have been supersed.
574:
1.216 brouard 575: Revision 1.215 2015/12/16 08:52:24 brouard
576: Summary: 0.98r4 working
577:
1.215 brouard 578: Revision 1.214 2015/12/16 06:57:54 brouard
579: Summary: temporary not working
580:
1.214 brouard 581: Revision 1.213 2015/12/11 18:22:17 brouard
582: Summary: 0.98r4
583:
1.213 brouard 584: Revision 1.212 2015/11/21 12:47:24 brouard
585: Summary: minor typo
586:
1.212 brouard 587: Revision 1.211 2015/11/21 12:41:11 brouard
588: Summary: 0.98r3 with some graph of projected cross-sectional
589:
590: Author: Nicolas Brouard
591:
1.211 brouard 592: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 593: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 594: Summary: Adding ftolpl parameter
595: Author: N Brouard
596:
597: We had difficulties to get smoothed confidence intervals. It was due
598: to the period prevalence which wasn't computed accurately. The inner
599: parameter ftolpl is now an outer parameter of the .imach parameter
600: file after estepm. If ftolpl is small 1.e-4 and estepm too,
601: computation are long.
602:
1.209 brouard 603: Revision 1.208 2015/11/17 14:31:57 brouard
604: Summary: temporary
605:
1.208 brouard 606: Revision 1.207 2015/10/27 17:36:57 brouard
607: *** empty log message ***
608:
1.207 brouard 609: Revision 1.206 2015/10/24 07:14:11 brouard
610: *** empty log message ***
611:
1.206 brouard 612: Revision 1.205 2015/10/23 15:50:53 brouard
613: Summary: 0.98r3 some clarification for graphs on likelihood contributions
614:
1.205 brouard 615: Revision 1.204 2015/10/01 16:20:26 brouard
616: Summary: Some new graphs of contribution to likelihood
617:
1.204 brouard 618: Revision 1.203 2015/09/30 17:45:14 brouard
619: Summary: looking at better estimation of the hessian
620:
621: Also a better criteria for convergence to the period prevalence And
622: therefore adding the number of years needed to converge. (The
623: prevalence in any alive state shold sum to one
624:
1.203 brouard 625: Revision 1.202 2015/09/22 19:45:16 brouard
626: Summary: Adding some overall graph on contribution to likelihood. Might change
627:
1.202 brouard 628: Revision 1.201 2015/09/15 17:34:58 brouard
629: Summary: 0.98r0
630:
631: - Some new graphs like suvival functions
632: - Some bugs fixed like model=1+age+V2.
633:
1.201 brouard 634: Revision 1.200 2015/09/09 16:53:55 brouard
635: Summary: Big bug thanks to Flavia
636:
637: Even model=1+age+V2. did not work anymore
638:
1.200 brouard 639: Revision 1.199 2015/09/07 14:09:23 brouard
640: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
641:
1.199 brouard 642: Revision 1.198 2015/09/03 07:14:39 brouard
643: Summary: 0.98q5 Flavia
644:
1.198 brouard 645: Revision 1.197 2015/09/01 18:24:39 brouard
646: *** empty log message ***
647:
1.197 brouard 648: Revision 1.196 2015/08/18 23:17:52 brouard
649: Summary: 0.98q5
650:
1.196 brouard 651: Revision 1.195 2015/08/18 16:28:39 brouard
652: Summary: Adding a hack for testing purpose
653:
654: After reading the title, ftol and model lines, if the comment line has
655: a q, starting with #q, the answer at the end of the run is quit. It
656: permits to run test files in batch with ctest. The former workaround was
657: $ echo q | imach foo.imach
658:
1.195 brouard 659: Revision 1.194 2015/08/18 13:32:00 brouard
660: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
661:
1.194 brouard 662: Revision 1.193 2015/08/04 07:17:42 brouard
663: Summary: 0.98q4
664:
1.193 brouard 665: Revision 1.192 2015/07/16 16:49:02 brouard
666: Summary: Fixing some outputs
667:
1.192 brouard 668: Revision 1.191 2015/07/14 10:00:33 brouard
669: Summary: Some fixes
670:
1.191 brouard 671: Revision 1.190 2015/05/05 08:51:13 brouard
672: Summary: Adding digits in output parameters (7 digits instead of 6)
673:
674: Fix 1+age+.
675:
1.190 brouard 676: Revision 1.189 2015/04/30 14:45:16 brouard
677: Summary: 0.98q2
678:
1.189 brouard 679: Revision 1.188 2015/04/30 08:27:53 brouard
680: *** empty log message ***
681:
1.188 brouard 682: Revision 1.187 2015/04/29 09:11:15 brouard
683: *** empty log message ***
684:
1.187 brouard 685: Revision 1.186 2015/04/23 12:01:52 brouard
686: Summary: V1*age is working now, version 0.98q1
687:
688: Some codes had been disabled in order to simplify and Vn*age was
689: working in the optimization phase, ie, giving correct MLE parameters,
690: but, as usual, outputs were not correct and program core dumped.
691:
1.186 brouard 692: Revision 1.185 2015/03/11 13:26:42 brouard
693: Summary: Inclusion of compile and links command line for Intel Compiler
694:
1.185 brouard 695: Revision 1.184 2015/03/11 11:52:39 brouard
696: Summary: Back from Windows 8. Intel Compiler
697:
1.184 brouard 698: Revision 1.183 2015/03/10 20:34:32 brouard
699: Summary: 0.98q0, trying with directest, mnbrak fixed
700:
701: We use directest instead of original Powell test; probably no
702: incidence on the results, but better justifications;
703: We fixed Numerical Recipes mnbrak routine which was wrong and gave
704: wrong results.
705:
1.183 brouard 706: Revision 1.182 2015/02/12 08:19:57 brouard
707: Summary: Trying to keep directest which seems simpler and more general
708: Author: Nicolas Brouard
709:
1.182 brouard 710: Revision 1.181 2015/02/11 23:22:24 brouard
711: Summary: Comments on Powell added
712:
713: Author:
714:
1.181 brouard 715: Revision 1.180 2015/02/11 17:33:45 brouard
716: Summary: Finishing move from main to function (hpijx and prevalence_limit)
717:
1.180 brouard 718: Revision 1.179 2015/01/04 09:57:06 brouard
719: Summary: back to OS/X
720:
1.179 brouard 721: Revision 1.178 2015/01/04 09:35:48 brouard
722: *** empty log message ***
723:
1.178 brouard 724: Revision 1.177 2015/01/03 18:40:56 brouard
725: Summary: Still testing ilc32 on OSX
726:
1.177 brouard 727: Revision 1.176 2015/01/03 16:45:04 brouard
728: *** empty log message ***
729:
1.176 brouard 730: Revision 1.175 2015/01/03 16:33:42 brouard
731: *** empty log message ***
732:
1.175 brouard 733: Revision 1.174 2015/01/03 16:15:49 brouard
734: Summary: Still in cross-compilation
735:
1.174 brouard 736: Revision 1.173 2015/01/03 12:06:26 brouard
737: Summary: trying to detect cross-compilation
738:
1.173 brouard 739: Revision 1.172 2014/12/27 12:07:47 brouard
740: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
741:
1.172 brouard 742: Revision 1.171 2014/12/23 13:26:59 brouard
743: Summary: Back from Visual C
744:
745: Still problem with utsname.h on Windows
746:
1.171 brouard 747: Revision 1.170 2014/12/23 11:17:12 brouard
748: Summary: Cleaning some \%% back to %%
749:
750: The escape was mandatory for a specific compiler (which one?), but too many warnings.
751:
1.170 brouard 752: Revision 1.169 2014/12/22 23:08:31 brouard
753: Summary: 0.98p
754:
755: Outputs some informations on compiler used, OS etc. Testing on different platforms.
756:
1.169 brouard 757: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 758: Summary: update
1.169 brouard 759:
1.168 brouard 760: Revision 1.167 2014/12/22 13:50:56 brouard
761: Summary: Testing uname and compiler version and if compiled 32 or 64
762:
763: Testing on Linux 64
764:
1.167 brouard 765: Revision 1.166 2014/12/22 11:40:47 brouard
766: *** empty log message ***
767:
1.166 brouard 768: Revision 1.165 2014/12/16 11:20:36 brouard
769: Summary: After compiling on Visual C
770:
771: * imach.c (Module): Merging 1.61 to 1.162
772:
1.165 brouard 773: Revision 1.164 2014/12/16 10:52:11 brouard
774: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
775:
776: * imach.c (Module): Merging 1.61 to 1.162
777:
1.164 brouard 778: Revision 1.163 2014/12/16 10:30:11 brouard
779: * imach.c (Module): Merging 1.61 to 1.162
780:
1.163 brouard 781: Revision 1.162 2014/09/25 11:43:39 brouard
782: Summary: temporary backup 0.99!
783:
1.162 brouard 784: Revision 1.1 2014/09/16 11:06:58 brouard
785: Summary: With some code (wrong) for nlopt
786:
787: Author:
788:
789: Revision 1.161 2014/09/15 20:41:41 brouard
790: Summary: Problem with macro SQR on Intel compiler
791:
1.161 brouard 792: Revision 1.160 2014/09/02 09:24:05 brouard
793: *** empty log message ***
794:
1.160 brouard 795: Revision 1.159 2014/09/01 10:34:10 brouard
796: Summary: WIN32
797: Author: Brouard
798:
1.159 brouard 799: Revision 1.158 2014/08/27 17:11:51 brouard
800: *** empty log message ***
801:
1.158 brouard 802: Revision 1.157 2014/08/27 16:26:55 brouard
803: Summary: Preparing windows Visual studio version
804: Author: Brouard
805:
806: In order to compile on Visual studio, time.h is now correct and time_t
807: and tm struct should be used. difftime should be used but sometimes I
808: just make the differences in raw time format (time(&now).
809: Trying to suppress #ifdef LINUX
810: Add xdg-open for __linux in order to open default browser.
811:
1.157 brouard 812: Revision 1.156 2014/08/25 20:10:10 brouard
813: *** empty log message ***
814:
1.156 brouard 815: Revision 1.155 2014/08/25 18:32:34 brouard
816: Summary: New compile, minor changes
817: Author: Brouard
818:
1.155 brouard 819: Revision 1.154 2014/06/20 17:32:08 brouard
820: Summary: Outputs now all graphs of convergence to period prevalence
821:
1.154 brouard 822: Revision 1.153 2014/06/20 16:45:46 brouard
823: Summary: If 3 live state, convergence to period prevalence on same graph
824: Author: Brouard
825:
1.153 brouard 826: Revision 1.152 2014/06/18 17:54:09 brouard
827: Summary: open browser, use gnuplot on same dir than imach if not found in the path
828:
1.152 brouard 829: Revision 1.151 2014/06/18 16:43:30 brouard
830: *** empty log message ***
831:
1.151 brouard 832: Revision 1.150 2014/06/18 16:42:35 brouard
833: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
834: Author: brouard
835:
1.150 brouard 836: Revision 1.149 2014/06/18 15:51:14 brouard
837: Summary: Some fixes in parameter files errors
838: Author: Nicolas Brouard
839:
1.149 brouard 840: Revision 1.148 2014/06/17 17:38:48 brouard
841: Summary: Nothing new
842: Author: Brouard
843:
844: Just a new packaging for OS/X version 0.98nS
845:
1.148 brouard 846: Revision 1.147 2014/06/16 10:33:11 brouard
847: *** empty log message ***
848:
1.147 brouard 849: Revision 1.146 2014/06/16 10:20:28 brouard
850: Summary: Merge
851: Author: Brouard
852:
853: Merge, before building revised version.
854:
1.146 brouard 855: Revision 1.145 2014/06/10 21:23:15 brouard
856: Summary: Debugging with valgrind
857: Author: Nicolas Brouard
858:
859: Lot of changes in order to output the results with some covariates
860: After the Edimburgh REVES conference 2014, it seems mandatory to
861: improve the code.
862: No more memory valgrind error but a lot has to be done in order to
863: continue the work of splitting the code into subroutines.
864: Also, decodemodel has been improved. Tricode is still not
865: optimal. nbcode should be improved. Documentation has been added in
866: the source code.
867:
1.144 brouard 868: Revision 1.143 2014/01/26 09:45:38 brouard
869: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
870:
871: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
872: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
873:
1.143 brouard 874: Revision 1.142 2014/01/26 03:57:36 brouard
875: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
876:
877: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
878:
1.142 brouard 879: Revision 1.141 2014/01/26 02:42:01 brouard
880: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
881:
1.141 brouard 882: Revision 1.140 2011/09/02 10:37:54 brouard
883: Summary: times.h is ok with mingw32 now.
884:
1.140 brouard 885: Revision 1.139 2010/06/14 07:50:17 brouard
886: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
887: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
888:
1.139 brouard 889: Revision 1.138 2010/04/30 18:19:40 brouard
890: *** empty log message ***
891:
1.138 brouard 892: Revision 1.137 2010/04/29 18:11:38 brouard
893: (Module): Checking covariates for more complex models
894: than V1+V2. A lot of change to be done. Unstable.
895:
1.137 brouard 896: Revision 1.136 2010/04/26 20:30:53 brouard
897: (Module): merging some libgsl code. Fixing computation
898: of likelione (using inter/intrapolation if mle = 0) in order to
899: get same likelihood as if mle=1.
900: Some cleaning of code and comments added.
901:
1.136 brouard 902: Revision 1.135 2009/10/29 15:33:14 brouard
903: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
904:
1.135 brouard 905: Revision 1.134 2009/10/29 13:18:53 brouard
906: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
907:
1.134 brouard 908: Revision 1.133 2009/07/06 10:21:25 brouard
909: just nforces
910:
1.133 brouard 911: Revision 1.132 2009/07/06 08:22:05 brouard
912: Many tings
913:
1.132 brouard 914: Revision 1.131 2009/06/20 16:22:47 brouard
915: Some dimensions resccaled
916:
1.131 brouard 917: Revision 1.130 2009/05/26 06:44:34 brouard
918: (Module): Max Covariate is now set to 20 instead of 8. A
919: lot of cleaning with variables initialized to 0. Trying to make
920: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
921:
1.130 brouard 922: Revision 1.129 2007/08/31 13:49:27 lievre
923: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
924:
1.129 lievre 925: Revision 1.128 2006/06/30 13:02:05 brouard
926: (Module): Clarifications on computing e.j
927:
1.128 brouard 928: Revision 1.127 2006/04/28 18:11:50 brouard
929: (Module): Yes the sum of survivors was wrong since
930: imach-114 because nhstepm was no more computed in the age
931: loop. Now we define nhstepma in the age loop.
932: (Module): In order to speed up (in case of numerous covariates) we
933: compute health expectancies (without variances) in a first step
934: and then all the health expectancies with variances or standard
935: deviation (needs data from the Hessian matrices) which slows the
936: computation.
937: In the future we should be able to stop the program is only health
938: expectancies and graph are needed without standard deviations.
939:
1.127 brouard 940: Revision 1.126 2006/04/28 17:23:28 brouard
941: (Module): Yes the sum of survivors was wrong since
942: imach-114 because nhstepm was no more computed in the age
943: loop. Now we define nhstepma in the age loop.
944: Version 0.98h
945:
1.126 brouard 946: Revision 1.125 2006/04/04 15:20:31 lievre
947: Errors in calculation of health expectancies. Age was not initialized.
948: Forecasting file added.
949:
950: Revision 1.124 2006/03/22 17:13:53 lievre
951: Parameters are printed with %lf instead of %f (more numbers after the comma).
952: The log-likelihood is printed in the log file
953:
954: Revision 1.123 2006/03/20 10:52:43 brouard
955: * imach.c (Module): <title> changed, corresponds to .htm file
956: name. <head> headers where missing.
957:
958: * imach.c (Module): Weights can have a decimal point as for
959: English (a comma might work with a correct LC_NUMERIC environment,
960: otherwise the weight is truncated).
961: Modification of warning when the covariates values are not 0 or
962: 1.
963: Version 0.98g
964:
965: Revision 1.122 2006/03/20 09:45:41 brouard
966: (Module): Weights can have a decimal point as for
967: English (a comma might work with a correct LC_NUMERIC environment,
968: otherwise the weight is truncated).
969: Modification of warning when the covariates values are not 0 or
970: 1.
971: Version 0.98g
972:
973: Revision 1.121 2006/03/16 17:45:01 lievre
974: * imach.c (Module): Comments concerning covariates added
975:
976: * imach.c (Module): refinements in the computation of lli if
977: status=-2 in order to have more reliable computation if stepm is
978: not 1 month. Version 0.98f
979:
980: Revision 1.120 2006/03/16 15:10:38 lievre
981: (Module): refinements in the computation of lli if
982: status=-2 in order to have more reliable computation if stepm is
983: not 1 month. Version 0.98f
984:
985: Revision 1.119 2006/03/15 17:42:26 brouard
986: (Module): Bug if status = -2, the loglikelihood was
987: computed as likelihood omitting the logarithm. Version O.98e
988:
989: Revision 1.118 2006/03/14 18:20:07 brouard
990: (Module): varevsij Comments added explaining the second
991: table of variances if popbased=1 .
992: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
993: (Module): Function pstamp added
994: (Module): Version 0.98d
995:
996: Revision 1.117 2006/03/14 17:16:22 brouard
997: (Module): varevsij Comments added explaining the second
998: table of variances if popbased=1 .
999: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
1000: (Module): Function pstamp added
1001: (Module): Version 0.98d
1002:
1003: Revision 1.116 2006/03/06 10:29:27 brouard
1004: (Module): Variance-covariance wrong links and
1005: varian-covariance of ej. is needed (Saito).
1006:
1007: Revision 1.115 2006/02/27 12:17:45 brouard
1008: (Module): One freematrix added in mlikeli! 0.98c
1009:
1010: Revision 1.114 2006/02/26 12:57:58 brouard
1011: (Module): Some improvements in processing parameter
1012: filename with strsep.
1013:
1014: Revision 1.113 2006/02/24 14:20:24 brouard
1015: (Module): Memory leaks checks with valgrind and:
1016: datafile was not closed, some imatrix were not freed and on matrix
1017: allocation too.
1018:
1019: Revision 1.112 2006/01/30 09:55:26 brouard
1020: (Module): Back to gnuplot.exe instead of wgnuplot.exe
1021:
1022: Revision 1.111 2006/01/25 20:38:18 brouard
1023: (Module): Lots of cleaning and bugs added (Gompertz)
1024: (Module): Comments can be added in data file. Missing date values
1025: can be a simple dot '.'.
1026:
1027: Revision 1.110 2006/01/25 00:51:50 brouard
1028: (Module): Lots of cleaning and bugs added (Gompertz)
1029:
1030: Revision 1.109 2006/01/24 19:37:15 brouard
1031: (Module): Comments (lines starting with a #) are allowed in data.
1032:
1033: Revision 1.108 2006/01/19 18:05:42 lievre
1034: Gnuplot problem appeared...
1035: To be fixed
1036:
1037: Revision 1.107 2006/01/19 16:20:37 brouard
1038: Test existence of gnuplot in imach path
1039:
1040: Revision 1.106 2006/01/19 13:24:36 brouard
1041: Some cleaning and links added in html output
1042:
1043: Revision 1.105 2006/01/05 20:23:19 lievre
1044: *** empty log message ***
1045:
1046: Revision 1.104 2005/09/30 16:11:43 lievre
1047: (Module): sump fixed, loop imx fixed, and simplifications.
1048: (Module): If the status is missing at the last wave but we know
1049: that the person is alive, then we can code his/her status as -2
1050: (instead of missing=-1 in earlier versions) and his/her
1051: contributions to the likelihood is 1 - Prob of dying from last
1052: health status (= 1-p13= p11+p12 in the easiest case of somebody in
1053: the healthy state at last known wave). Version is 0.98
1054:
1055: Revision 1.103 2005/09/30 15:54:49 lievre
1056: (Module): sump fixed, loop imx fixed, and simplifications.
1057:
1058: Revision 1.102 2004/09/15 17:31:30 brouard
1059: Add the possibility to read data file including tab characters.
1060:
1061: Revision 1.101 2004/09/15 10:38:38 brouard
1062: Fix on curr_time
1063:
1064: Revision 1.100 2004/07/12 18:29:06 brouard
1065: Add version for Mac OS X. Just define UNIX in Makefile
1066:
1067: Revision 1.99 2004/06/05 08:57:40 brouard
1068: *** empty log message ***
1069:
1070: Revision 1.98 2004/05/16 15:05:56 brouard
1071: New version 0.97 . First attempt to estimate force of mortality
1072: directly from the data i.e. without the need of knowing the health
1073: state at each age, but using a Gompertz model: log u =a + b*age .
1074: This is the basic analysis of mortality and should be done before any
1075: other analysis, in order to test if the mortality estimated from the
1076: cross-longitudinal survey is different from the mortality estimated
1077: from other sources like vital statistic data.
1078:
1079: The same imach parameter file can be used but the option for mle should be -3.
1080:
1.324 brouard 1081: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 1082: former routines in order to include the new code within the former code.
1083:
1084: The output is very simple: only an estimate of the intercept and of
1085: the slope with 95% confident intervals.
1086:
1087: Current limitations:
1088: A) Even if you enter covariates, i.e. with the
1089: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
1090: B) There is no computation of Life Expectancy nor Life Table.
1091:
1092: Revision 1.97 2004/02/20 13:25:42 lievre
1093: Version 0.96d. Population forecasting command line is (temporarily)
1094: suppressed.
1095:
1096: Revision 1.96 2003/07/15 15:38:55 brouard
1097: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
1098: rewritten within the same printf. Workaround: many printfs.
1099:
1100: Revision 1.95 2003/07/08 07:54:34 brouard
1101: * imach.c (Repository):
1102: (Repository): Using imachwizard code to output a more meaningful covariance
1103: matrix (cov(a12,c31) instead of numbers.
1104:
1105: Revision 1.94 2003/06/27 13:00:02 brouard
1106: Just cleaning
1107:
1108: Revision 1.93 2003/06/25 16:33:55 brouard
1109: (Module): On windows (cygwin) function asctime_r doesn't
1110: exist so I changed back to asctime which exists.
1111: (Module): Version 0.96b
1112:
1113: Revision 1.92 2003/06/25 16:30:45 brouard
1114: (Module): On windows (cygwin) function asctime_r doesn't
1115: exist so I changed back to asctime which exists.
1116:
1117: Revision 1.91 2003/06/25 15:30:29 brouard
1118: * imach.c (Repository): Duplicated warning errors corrected.
1119: (Repository): Elapsed time after each iteration is now output. It
1120: helps to forecast when convergence will be reached. Elapsed time
1121: is stamped in powell. We created a new html file for the graphs
1122: concerning matrix of covariance. It has extension -cov.htm.
1123:
1124: Revision 1.90 2003/06/24 12:34:15 brouard
1125: (Module): Some bugs corrected for windows. Also, when
1126: mle=-1 a template is output in file "or"mypar.txt with the design
1127: of the covariance matrix to be input.
1128:
1129: Revision 1.89 2003/06/24 12:30:52 brouard
1130: (Module): Some bugs corrected for windows. Also, when
1131: mle=-1 a template is output in file "or"mypar.txt with the design
1132: of the covariance matrix to be input.
1133:
1134: Revision 1.88 2003/06/23 17:54:56 brouard
1135: * imach.c (Repository): Create a sub-directory where all the secondary files are. Only imach, htm, gp and r(imach) are on the main directory. Correct time and other things.
1136:
1137: Revision 1.87 2003/06/18 12:26:01 brouard
1138: Version 0.96
1139:
1140: Revision 1.86 2003/06/17 20:04:08 brouard
1141: (Module): Change position of html and gnuplot routines and added
1142: routine fileappend.
1143:
1144: Revision 1.85 2003/06/17 13:12:43 brouard
1145: * imach.c (Repository): Check when date of death was earlier that
1146: current date of interview. It may happen when the death was just
1147: prior to the death. In this case, dh was negative and likelihood
1148: was wrong (infinity). We still send an "Error" but patch by
1149: assuming that the date of death was just one stepm after the
1150: interview.
1151: (Repository): Because some people have very long ID (first column)
1152: we changed int to long in num[] and we added a new lvector for
1153: memory allocation. But we also truncated to 8 characters (left
1154: truncation)
1155: (Repository): No more line truncation errors.
1156:
1157: Revision 1.84 2003/06/13 21:44:43 brouard
1158: * imach.c (Repository): Replace "freqsummary" at a correct
1159: place. It differs from routine "prevalence" which may be called
1160: many times. Probs is memory consuming and must be used with
1161: parcimony.
1162: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
1163:
1164: Revision 1.83 2003/06/10 13:39:11 lievre
1165: *** empty log message ***
1166:
1167: Revision 1.82 2003/06/05 15:57:20 brouard
1168: Add log in imach.c and fullversion number is now printed.
1169:
1170: */
1171: /*
1172: Interpolated Markov Chain
1173:
1174: Short summary of the programme:
1175:
1.227 brouard 1176: This program computes Healthy Life Expectancies or State-specific
1177: (if states aren't health statuses) Expectancies from
1178: cross-longitudinal data. Cross-longitudinal data consist in:
1179:
1180: -1- a first survey ("cross") where individuals from different ages
1181: are interviewed on their health status or degree of disability (in
1182: the case of a health survey which is our main interest)
1183:
1184: -2- at least a second wave of interviews ("longitudinal") which
1185: measure each change (if any) in individual health status. Health
1186: expectancies are computed from the time spent in each health state
1187: according to a model. More health states you consider, more time is
1188: necessary to reach the Maximum Likelihood of the parameters involved
1189: in the model. The simplest model is the multinomial logistic model
1190: where pij is the probability to be observed in state j at the second
1191: wave conditional to be observed in state i at the first
1192: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1193: etc , where 'age' is age and 'sex' is a covariate. If you want to
1194: have a more complex model than "constant and age", you should modify
1195: the program where the markup *Covariates have to be included here
1196: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1197: convergence.
1198:
1199: The advantage of this computer programme, compared to a simple
1200: multinomial logistic model, is clear when the delay between waves is not
1201: identical for each individual. Also, if a individual missed an
1202: intermediate interview, the information is lost, but taken into
1203: account using an interpolation or extrapolation.
1204:
1205: hPijx is the probability to be observed in state i at age x+h
1206: conditional to the observed state i at age x. The delay 'h' can be
1207: split into an exact number (nh*stepm) of unobserved intermediate
1208: states. This elementary transition (by month, quarter,
1209: semester or year) is modelled as a multinomial logistic. The hPx
1210: matrix is simply the matrix product of nh*stepm elementary matrices
1211: and the contribution of each individual to the likelihood is simply
1212: hPijx.
1213:
1214: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1215: of the life expectancies. It also computes the period (stable) prevalence.
1216:
1217: Back prevalence and projections:
1.227 brouard 1218:
1219: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1220: double agemaxpar, double ftolpl, int *ncvyearp, double
1221: dateprev1,double dateprev2, int firstpass, int lastpass, int
1222: mobilavproj)
1223:
1224: Computes the back prevalence limit for any combination of
1225: covariate values k at any age between ageminpar and agemaxpar and
1226: returns it in **bprlim. In the loops,
1227:
1228: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1229: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1230:
1231: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1232: Computes for any combination of covariates k and any age between bage and fage
1233: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1234: oldm=oldms;savm=savms;
1.227 brouard 1235:
1.267 brouard 1236: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1237: Computes the transition matrix starting at age 'age' over
1238: 'nhstepm*hstepm*stepm' months (i.e. until
1239: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1240: nhstepm*hstepm matrices.
1241:
1242: Returns p3mat[i][j][h] after calling
1243: p3mat[i][j][h]=matprod2(newm,
1244: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1245: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1246: oldm);
1.226 brouard 1247:
1248: Important routines
1249:
1250: - func (or funcone), computes logit (pij) distinguishing
1251: o fixed variables (single or product dummies or quantitative);
1252: o varying variables by:
1253: (1) wave (single, product dummies, quantitative),
1254: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1255: % fixed dummy (treated) or quantitative (not done because time-consuming);
1256: % varying dummy (not done) or quantitative (not done);
1257: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1258: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1259: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.364 brouard 1260: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, eliminating 1 1 if
1.226 brouard 1261: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1262:
1.226 brouard 1263:
1264:
1.324 brouard 1265: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1266: Institut national d'études démographiques, Paris.
1.126 brouard 1267: This software have been partly granted by Euro-REVES, a concerted action
1268: from the European Union.
1269: It is copyrighted identically to a GNU software product, ie programme and
1270: software can be distributed freely for non commercial use. Latest version
1271: can be accessed at http://euroreves.ined.fr/imach .
1272:
1273: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1274: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1275:
1276: **********************************************************************/
1277: /*
1278: main
1279: read parameterfile
1280: read datafile
1281: concatwav
1282: freqsummary
1283: if (mle >= 1)
1284: mlikeli
1285: print results files
1286: if mle==1
1287: computes hessian
1288: read end of parameter file: agemin, agemax, bage, fage, estepm
1289: begin-prev-date,...
1290: open gnuplot file
1291: open html file
1.145 brouard 1292: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1293: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1294: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1295: freexexit2 possible for memory heap.
1296:
1297: h Pij x | pij_nom ficrestpij
1298: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1299: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1300: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1301:
1302: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1303: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1304: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1305: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1306: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1307:
1.126 brouard 1308: forecasting if prevfcast==1 prevforecast call prevalence()
1309: health expectancies
1310: Variance-covariance of DFLE
1311: prevalence()
1312: movingaverage()
1313: varevsij()
1314: if popbased==1 varevsij(,popbased)
1315: total life expectancies
1316: Variance of period (stable) prevalence
1317: end
1318: */
1319:
1.187 brouard 1320: /* #define DEBUG */
1321: /* #define DEBUGBRENT */
1.203 brouard 1322: /* #define DEBUGLINMIN */
1323: /* #define DEBUGHESS */
1324: #define DEBUGHESSIJ
1.224 brouard 1325: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1326: #define POWELL /* Instead of NLOPT */
1.224 brouard 1327: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1328: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1329: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1330: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.359 brouard 1331: /* #define POWELLORIGINCONJUGATE /\* Don't use conjugate but biggest decrease if valuable *\/ */
1332: /* #define NOTMINFIT */
1.126 brouard 1333:
1334: #include <math.h>
1335: #include <stdio.h>
1336: #include <stdlib.h>
1337: #include <string.h>
1.226 brouard 1338: #include <ctype.h>
1.159 brouard 1339:
1340: #ifdef _WIN32
1341: #include <io.h>
1.172 brouard 1342: #include <windows.h>
1343: #include <tchar.h>
1.159 brouard 1344: #else
1.126 brouard 1345: #include <unistd.h>
1.159 brouard 1346: #endif
1.126 brouard 1347:
1348: #include <limits.h>
1349: #include <sys/types.h>
1.171 brouard 1350:
1351: #if defined(__GNUC__)
1352: #include <sys/utsname.h> /* Doesn't work on Windows */
1353: #endif
1354:
1.126 brouard 1355: #include <sys/stat.h>
1356: #include <errno.h>
1.159 brouard 1357: /* extern int errno; */
1.126 brouard 1358:
1.157 brouard 1359: /* #ifdef LINUX */
1360: /* #include <time.h> */
1361: /* #include "timeval.h" */
1362: /* #else */
1363: /* #include <sys/time.h> */
1364: /* #endif */
1365:
1.126 brouard 1366: #include <time.h>
1367:
1.136 brouard 1368: #ifdef GSL
1369: #include <gsl/gsl_errno.h>
1370: #include <gsl/gsl_multimin.h>
1371: #endif
1372:
1.167 brouard 1373:
1.162 brouard 1374: #ifdef NLOPT
1375: #include <nlopt.h>
1376: typedef struct {
1377: double (* function)(double [] );
1378: } myfunc_data ;
1379: #endif
1380:
1.126 brouard 1381: /* #include <libintl.h> */
1382: /* #define _(String) gettext (String) */
1383:
1.349 brouard 1384: #define MAXLINE 16384 /* Was 256 and 1024 and 2048. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1385:
1386: #define GNUPLOTPROGRAM "gnuplot"
1.343 brouard 1387: #define GNUPLOTVERSION 5.1
1388: double gnuplotversion=GNUPLOTVERSION;
1.126 brouard 1389: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1390: #define FILENAMELENGTH 256
1.126 brouard 1391:
1392: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1393: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1394:
1.349 brouard 1395: #define MAXPARM 216 /**< Maximum number of parameters for the optimization was 128 */
1.144 brouard 1396: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1397:
1398: #define NINTERVMAX 8
1.144 brouard 1399: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1400: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1401: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1402: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1403: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1404: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1405: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1406: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1407: /* #define AGESUP 130 */
1.288 brouard 1408: /* #define AGESUP 150 */
1409: #define AGESUP 200
1.268 brouard 1410: #define AGEINF 0
1.218 brouard 1411: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1412: #define AGEBASE 40
1.194 brouard 1413: #define AGEOVERFLOW 1.e20
1.164 brouard 1414: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1415: #ifdef _WIN32
1416: #define DIRSEPARATOR '\\'
1417: #define CHARSEPARATOR "\\"
1418: #define ODIRSEPARATOR '/'
1419: #else
1.126 brouard 1420: #define DIRSEPARATOR '/'
1421: #define CHARSEPARATOR "/"
1422: #define ODIRSEPARATOR '\\'
1423: #endif
1424:
1.367 ! brouard 1425: /* $Id: imach.c,v 1.366 2024/07/02 09:42:10 brouard Exp $ */
1.126 brouard 1426: /* $State: Exp $ */
1.196 brouard 1427: #include "version.h"
1428: char version[]=__IMACH_VERSION__;
1.360 brouard 1429: char copyright[]="April 2024,INED-EUROREVES-Institut de longevite-Japan Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25293121), Intel Software 2015-2020, Nihon University 2021-202, INED 2000-2024";
1.367 ! brouard 1430: char fullversion[]="$Revision: 1.366 $ $Date: 2024/07/02 09:42:10 $";
1.126 brouard 1431: char strstart[80];
1432: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1433: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.342 brouard 1434: int debugILK=0; /* debugILK is set by a #d in a comment line */
1.187 brouard 1435: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 brouard 1436: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
1437: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
1.335 brouard 1438: int cptcovn=0; /**< cptcovn decodemodel: number of covariates k of the models excluding age*products =6 and age*age but including products */
1.330 brouard 1439: int cptcovt=0; /**< cptcovt: total number of covariates of the model (2) nbocc(+)+1 = 8 excepting constant and age and age*age */
1.335 brouard 1440: int cptcovs=0; /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
1441: int cptcovsnq=0; /**< cptcovsnq number of SIMPLE covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1442: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1.349 brouard 1443: int cptcovprodage=0; /**< Number of fixed covariates with age: V3*age or V2*V3*age 1 */
1444: int cptcovprodvage=0; /**< Number of varying covariates with age: V7*age or V7*V6*age */
1445: int cptcovdageprod=0; /**< Number of doubleproducts with age, since 0.99r44 only: age*Vn*Vm for gnuplot printing*/
1.145 brouard 1446: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.335 brouard 1447: int cptcoveff=0; /* Total number of single dummy covariates (fixed or time varying) to vary for printing results (2**cptcoveff combinations of dummies)(computed in tricode as cptcov) */
1.233 brouard 1448: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1449: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.339 brouard 1450: int ncovvt=0; /* Total number of effective (wave) varying covariates (dummy or quantitative or products [without age]) in the model */
1.349 brouard 1451: int ncovvta=0; /* +age*V6 + age*V7+ age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 Total number of expandend products [with age]) in the model */
1452: int ncovta=0; /*age*V3*V2 +age*V2+agev3+ageV4 +age*V6 + age*V7+ age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 Total number of expandend products [with age]) in the model */
1453: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (single or product, dummy or quantitative) in the model */
1454: int ncovva=0; /* +age*V6 + age*V7+ge*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 Total number of effective (wave and stepm) varying with age covariates (single or product, dummy or quantitative) in the model */
1.234 brouard 1455: int nsd=0; /**< Total number of single dummy variables (output) */
1456: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1457: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1458: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1459: int ntveff=0; /**< ntveff number of effective time varying variables */
1460: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1461: int cptcov=0; /* Working variable */
1.334 brouard 1462: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs+1 declared globally ;*/
1.290 brouard 1463: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1464: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1465: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1466: int nlstate=2; /* Number of live states */
1467: int ndeath=1; /* Number of dead states */
1.130 brouard 1468: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.339 brouard 1469: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable*/
1470: int ncovcolt=0; /* ncovcolt=ncovcol+nqv+ntv+nqtv; total of covariates in the data, not in the model equation*/
1.126 brouard 1471: int popbased=0;
1472:
1473: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1474: int maxwav=0; /* Maxim number of waves */
1475: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1476: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1.359 brouard 1477: int gipmx = 0;
1478: double gsw = 0; /* Global variables on the number of contributions
1.126 brouard 1479: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1480: int mle=1, weightopt=0;
1.126 brouard 1481: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1482: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1483: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1484: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1485: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1486: int selected(int kvar); /* Is covariate kvar selected for printing results */
1487:
1.130 brouard 1488: double jmean=1; /* Mean space between 2 waves */
1.366 brouard 1489: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b); /* test */
1490: /* double **matprod2(); *//* test */
1.126 brouard 1491: double **oldm, **newm, **savm; /* Working pointers to matrices */
1492: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1493: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1494:
1.136 brouard 1495: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1496: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1497: FILE *ficlog, *ficrespow;
1.130 brouard 1498: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1499: double fretone; /* Only one call to likelihood */
1.130 brouard 1500: long ipmx=0; /* Number of contributions */
1.126 brouard 1501: double sw; /* Sum of weights */
1502: char filerespow[FILENAMELENGTH];
1503: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1504: FILE *ficresilk;
1505: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1506: FILE *ficresprobmorprev;
1507: FILE *fichtm, *fichtmcov; /* Html File */
1508: FILE *ficreseij;
1509: char filerese[FILENAMELENGTH];
1510: FILE *ficresstdeij;
1511: char fileresstde[FILENAMELENGTH];
1512: FILE *ficrescveij;
1513: char filerescve[FILENAMELENGTH];
1514: FILE *ficresvij;
1515: char fileresv[FILENAMELENGTH];
1.269 brouard 1516:
1.126 brouard 1517: char title[MAXLINE];
1.234 brouard 1518: char model[MAXLINE]; /**< The model line */
1.217 brouard 1519: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1520: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1521: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1522: char command[FILENAMELENGTH];
1523: int outcmd=0;
1524:
1.217 brouard 1525: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1526: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1527: char filelog[FILENAMELENGTH]; /* Log file */
1528: char filerest[FILENAMELENGTH];
1529: char fileregp[FILENAMELENGTH];
1530: char popfile[FILENAMELENGTH];
1531:
1532: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1533:
1.157 brouard 1534: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1535: /* struct timezone tzp; */
1536: /* extern int gettimeofday(); */
1537:
1.366 brouard 1538: /* extern time_t time(); */ /* Commented out for clang */
1539: /* struct tm tml, *gmtime(), *localtime(); */
1540:
1.157 brouard 1541:
1542: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1543: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1.349 brouard 1544: time_t rlast_btime; /* raw time */
1.366 brouard 1545: /* struct tm tm; */
1546: struct tm tm, tml;
1.157 brouard 1547:
1.126 brouard 1548: char strcurr[80], strfor[80];
1549:
1550: char *endptr;
1551: long lval;
1552: double dval;
1553:
1.362 brouard 1554: /* This for praxis gegen */
1555: /* int prin=1; */
1556: double h0=0.25;
1557: double macheps;
1558: double ffmin;
1559:
1.126 brouard 1560: #define NR_END 1
1561: #define FREE_ARG char*
1562: #define FTOL 1.0e-10
1563:
1564: #define NRANSI
1.240 brouard 1565: #define ITMAX 200
1566: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1567:
1568: #define TOL 2.0e-4
1569:
1570: #define CGOLD 0.3819660
1571: #define ZEPS 1.0e-10
1572: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1573:
1574: #define GOLD 1.618034
1575: #define GLIMIT 100.0
1576: #define TINY 1.0e-20
1577:
1578: static double maxarg1,maxarg2;
1579: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1580: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1581:
1582: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1583: #define rint(a) floor(a+0.5)
1.166 brouard 1584: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1585: #define mytinydouble 1.0e-16
1.166 brouard 1586: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1587: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1588: /* static double dsqrarg; */
1589: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1590: static double sqrarg;
1591: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1592: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1593: int agegomp= AGEGOMP;
1594:
1595: int imx;
1596: int stepm=1;
1597: /* Stepm, step in month: minimum step interpolation*/
1598:
1599: int estepm;
1600: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1601:
1602: int m,nb;
1603: long *num;
1.197 brouard 1604: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1605: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1606: covariate for which somebody answered excluding
1607: undefined. Usually 2: 0 and 1. */
1608: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1609: covariate for which somebody answered including
1610: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1611: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1612: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1613: double ***mobaverage, ***mobaverages; /* New global variable */
1.332 brouard 1614: double **precov; /* New global variable to store for each resultline, values of model covariates given by the resultlines (in order to speed up) */
1.126 brouard 1615: double *ageexmed,*agecens;
1616: double dateintmean=0;
1.296 brouard 1617: double anprojd, mprojd, jprojd; /* For eventual projections */
1618: double anprojf, mprojf, jprojf;
1.126 brouard 1619:
1.296 brouard 1620: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1621: double anbackf, mbackf, jbackf;
1622: double jintmean,mintmean,aintmean;
1.126 brouard 1623: double *weight;
1624: int **s; /* Status */
1.141 brouard 1625: double *agedc;
1.145 brouard 1626: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1627: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1628: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1629: double **coqvar; /* Fixed quantitative covariate nqv */
1.341 brouard 1630: double ***cotvar; /* Time varying covariate start at ncovcol + nqv + (1 to ntv) */
1.225 brouard 1631: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1632: double idx;
1633: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1634: /* Some documentation */
1635: /* Design original data
1636: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1637: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1638: * ntv=3 nqtv=1
1.330 brouard 1639: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1640: * For time varying covariate, quanti or dummies
1641: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1.341 brouard 1642: * cotvar[wav][ncovcol+nqv+ iv(1 to nqtv)][i]= [(1 to nqtv)][i]=(V12) quanti
1.319 brouard 1643: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1644: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1.332 brouard 1645: * covar[Vk,i], value of the Vkth fixed covariate dummy or quanti for individual i:
1.319 brouard 1646: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1647: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1648: * k= 1 2 3 4 5 6 7 8 9 10 11
1649: */
1650: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1651: /* ncovcol=1(Males=0 Females=1) nqv=1(raedyrs) ntv=2(withoutiadl=0 withiadl=1, witoutadl=0 withoutadl=1) nqtv=1(bmi) nlstate=3 ndeath=1
1652: # States 1=Coresidence, 2 Living alone, 3 Institution
1653: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1654: */
1.349 brouard 1655: /* V5+V4+ V3+V4*V3 +V5*age+V2 +V1*V2+V1*age+V1+V4*V3*age */
1656: /* kmodel 1 2 3 4 5 6 7 8 9 10 */
1657: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 3 *//*0 for simple covariate (dummy, quantitative,*/
1658: /* fixed or varying), 1 for age product, 2 for*/
1659: /* product without age, 3 for age and double product */
1660: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 3 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1661: /*(single or product without age), 2 dummy*/
1662: /* with age product, 3 quant with age product*/
1663: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 6 */
1664: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1665: /*TnsdVar[Tvar] 1 2 3 */
1666: /*Tvaraff[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1667: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1668: /*TvarsDind[nsd] 2 3 9 */ /* position K of single dummy cova */
1669: /* nsq 1 2 */ /* Counting single quantit tv */
1670: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1671: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1672: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1673: /* cptcovage 1 2 3 */ /* Counting cov*age in the model equation */
1674: /* Tage[cptcovage]=k 5 8 10 */ /* Position in the model of ith cov*age */
1.350 brouard 1675: /* model="V2+V3+V4+V6+V7+V6*V2+V7*V2+V6*V3+V7*V3+V6*V4+V7*V4+age*V2+age*V3+age*V4+age*V6+age*V7+age*V6*V2+age*V6*V3+age*V7*V3+age*V6*V4+age*V7*V4\r"*/
1676: /* p Tvard[1][1]@21 = {6, 2, 7, 2, 6, 3, 7, 3, 6, 4, 7, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0}*/
1.354 brouard 1677: /* p Tvard[2][1]@21 = {7, 2, 6, 3, 7, 3, 6, 4, 7, 4, 0 <repeats 11 times>} */
1.350 brouard 1678: /* p Tvardk[1][1]@24 = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 6, 2, 7, 2, 6, 3, 7, 3, 6, 4, 7, 4, 0, 0}*/
1679: /* p Tvardk[1][1]@22 = {0, 0, 0, 0, 0, 0, 0, 0, 6, 2, 7, 2, 6, 3, 7, 3, 6, 4, 7, 4, 0, 0} */
1.349 brouard 1680: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 brouard 1681: /* Tvardk[4][1]=4;Tvardk[4][2]=3;Tvardk[7][1]=1;Tvardk[7][2]=2 */ /* Variables of a prod at position in the model equation*/
1.319 brouard 1682: /* TvarF TvarF[1]=Tvar[6]=2, TvarF[2]=Tvar[7]=7, TvarF[3]=Tvar[9]=1 ID of fixed covariates or product V2, V1*V2, V1 */
1.320 brouard 1683: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1684: /* Type */
1685: /* V 1 2 3 4 5 */
1686: /* F F V V V */
1687: /* D Q D D Q */
1688: /* */
1689: int *TvarsD;
1.330 brouard 1690: int *TnsdVar;
1.234 brouard 1691: int *TvarsDind;
1692: int *TvarsQ;
1693: int *TvarsQind;
1694:
1.318 brouard 1695: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1696: int nresult=0;
1.258 brouard 1697: int parameterline=0; /* # of the parameter (type) line */
1.334 brouard 1698: int TKresult[MAXRESULTLINESPONE]; /* TKresult[nres]=k for each resultline nres give the corresponding combination of dummies */
1699: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model corresponds to the k3 position in the resultline */
1700: int modelresult[MAXRESULTLINESPONE][NCOVMAX];/* modelresult[k3]=k1: k1th position in the model corresponds to the k3 position in the resultline */
1701: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1702: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1703: double TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.334 brouard 1704: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1705: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.318 brouard 1706: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1.332 brouard 1707: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.318 brouard 1708:
1709: /* ncovcol=1(Males=0 Females=1) nqv=1(raedyrs) ntv=2(withoutiadl=0 withiadl=1, witoutadl=0 withoutadl=1) nqtv=1(bmi) nlstate=3 ndeath=1
1710: # States 1=Coresidence, 2 Living alone, 3 Institution
1711: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1712: */
1.234 brouard 1713: /* int *TDvar; /\**< TDvar[1]=4, TDvarF[2]=3, TDvar[3]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
1.232 brouard 1714: int *TvarF; /**< TvarF[1]=Tvar[6]=2, TvarF[2]=Tvar[7]=7, TvarF[3]=Tvar[9]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1715: int *TvarFind; /**< TvarFind[1]=6, TvarFind[2]=7, Tvarind[3]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1716: int *TvarV; /**< TvarV[1]=Tvar[1]=5, TvarV[2]=Tvar[2]=4 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1717: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1718: int *TvarA; /**< TvarA[1]=Tvar[5]=5, TvarA[2]=Tvar[8]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1719: int *TvarAind; /**< TvarindA[1]=5, TvarAind[2]=8 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 1720: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1721: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1722: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1723: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1724: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1725: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1726: int *TvarVQ; /* TvarVQ[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying quantitative variable */
1727: int *TvarVQind; /* TvarVQind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying quantitative variable */
1.339 brouard 1728: int *TvarVV; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1729: int *TvarVVind; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1.349 brouard 1730: int *TvarVVA; /* We count ncovvt time varying covariates (single or products with age) and put their name into TvarVVA */
1731: int *TvarVVAind; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1732: int *TvarAVVA; /* We count ALL ncovta time varying covariates (single or products with age) and put their name into TvarVVA */
1733: int *TvarAVVAind; /* We count ALL ncovta time varying covariates (single or products without age) and put their name into TvarVV */
1.339 brouard 1734: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
1.349 brouard 1735: /* model V1+V3+age*V1+age*V3+V1*V3 + V1*V3*age */
1736: /* Tvar={1, 3, 1, 3, 6, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
1737: /* TvarVV={3,1,3,1,3}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1738: /* TvarVVind={2,5,5,6,6}, for V3 and then the product V1*V3 is decomposed into V1 and V3 and V1*V3*age into 6,6 */
1.230 brouard 1739: int *Tvarsel; /**< Selected covariates for output */
1740: double *Tvalsel; /**< Selected modality value of covariate for output */
1.349 brouard 1741: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product, 3 age*Vn*Vm */
1.227 brouard 1742: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1743: int *Dummy; /** Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product */
1.238 brouard 1744: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1745: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1746: int *Tage;
1.227 brouard 1747: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1748: int *Tmodelind; /** Tmodelind[Tvaraff[3]]=9 for V1 position,Tvaraff[1]@9={4, 3, 1, 0, 0, 0, 0, 0, 0}, model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.230 brouard 1749: int *TmodelInvind; /** Tmodelind[Tvaraff[3]]=9 for V1 position,Tvaraff[1]@9={4, 3, 1, 0, 0, 0, 0, 0, 0}, model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1750: int *TmodelInvQind; /** Tmodelqind[1]=1 for V5(quantitative varying) position,Tvaraff[1]@9={4, 3, 1, 0, 0, 0, 0, 0, 0}, model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.145 brouard 1751: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1752: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1753: int **Tvard;
1.330 brouard 1754: int **Tvardk;
1.227 brouard 1755: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1756: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1757: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1758: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1759: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1760: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1761: double *lsurv, *lpop, *tpop;
1762:
1.231 brouard 1763: #define FD 1; /* Fixed dummy covariate */
1764: #define FQ 2; /* Fixed quantitative covariate */
1765: #define FP 3; /* Fixed product covariate */
1766: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1767: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1768: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1769: #define VD 10; /* Varying dummy covariate */
1770: #define VQ 11; /* Varying quantitative covariate */
1771: #define VP 12; /* Varying product covariate */
1772: #define VPDD 13; /* Varying product dummy*dummy covariate */
1773: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1774: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1775: #define APFD 16; /* Age product * fixed dummy covariate */
1776: #define APFQ 17; /* Age product * fixed quantitative covariate */
1777: #define APVD 18; /* Age product * varying dummy covariate */
1778: #define APVQ 19; /* Age product * varying quantitative covariate */
1779:
1780: #define FTYPE 1; /* Fixed covariate */
1781: #define VTYPE 2; /* Varying covariate (loop in wave) */
1782: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1783:
1784: struct kmodel{
1785: int maintype; /* main type */
1786: int subtype; /* subtype */
1787: };
1788: struct kmodel modell[NCOVMAX];
1789:
1.143 brouard 1790: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1791: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1792:
1793: /**************** split *************************/
1794: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1795: {
1796: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1797: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1798: */
1799: char *ss; /* pointer */
1.186 brouard 1800: int l1=0, l2=0; /* length counters */
1.126 brouard 1801:
1802: l1 = strlen(path ); /* length of path */
1803: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1804: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1805: if ( ss == NULL ) { /* no directory, so determine current directory */
1806: strcpy( name, path ); /* we got the fullname name because no directory */
1807: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1808: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1809: /* get current working directory */
1810: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1811: #ifdef WIN32
1812: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1813: #else
1814: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1815: #endif
1.126 brouard 1816: return( GLOCK_ERROR_GETCWD );
1817: }
1818: /* got dirc from getcwd*/
1819: printf(" DIRC = %s \n",dirc);
1.205 brouard 1820: } else { /* strip directory from path */
1.126 brouard 1821: ss++; /* after this, the filename */
1822: l2 = strlen( ss ); /* length of filename */
1823: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1824: strcpy( name, ss ); /* save file name */
1825: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1826: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1827: printf(" DIRC2 = %s \n",dirc);
1828: }
1829: /* We add a separator at the end of dirc if not exists */
1830: l1 = strlen( dirc ); /* length of directory */
1831: if( dirc[l1-1] != DIRSEPARATOR ){
1832: dirc[l1] = DIRSEPARATOR;
1833: dirc[l1+1] = 0;
1834: printf(" DIRC3 = %s \n",dirc);
1835: }
1836: ss = strrchr( name, '.' ); /* find last / */
1837: if (ss >0){
1838: ss++;
1839: strcpy(ext,ss); /* save extension */
1840: l1= strlen( name);
1841: l2= strlen(ss)+1;
1842: strncpy( finame, name, l1-l2);
1843: finame[l1-l2]= 0;
1844: }
1845:
1846: return( 0 ); /* we're done */
1847: }
1848:
1849:
1850: /******************************************/
1851:
1852: void replace_back_to_slash(char *s, char*t)
1853: {
1854: int i;
1855: int lg=0;
1856: i=0;
1857: lg=strlen(t);
1858: for(i=0; i<= lg; i++) {
1859: (s[i] = t[i]);
1860: if (t[i]== '\\') s[i]='/';
1861: }
1862: }
1863:
1.132 brouard 1864: char *trimbb(char *out, char *in)
1.137 brouard 1865: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1866: char *s;
1867: s=out;
1868: while (*in != '\0'){
1.137 brouard 1869: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1870: in++;
1871: }
1872: *out++ = *in++;
1873: }
1874: *out='\0';
1875: return s;
1876: }
1877:
1.351 brouard 1878: char *trimbtab(char *out, char *in)
1879: { /* Trim blanks or tabs in line but keeps first blanks if line starts with blanks */
1880: char *s;
1881: s=out;
1882: while (*in != '\0'){
1883: while( (*in == ' ' || *in == '\t')){ /* && *(in+1) != '\0'){*/
1884: in++;
1885: }
1886: *out++ = *in++;
1887: }
1888: *out='\0';
1889: return s;
1890: }
1891:
1.187 brouard 1892: /* char *substrchaine(char *out, char *in, char *chain) */
1893: /* { */
1894: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1895: /* char *s, *t; */
1896: /* t=in;s=out; */
1897: /* while ((*in != *chain) && (*in != '\0')){ */
1898: /* *out++ = *in++; */
1899: /* } */
1900:
1901: /* /\* *in matches *chain *\/ */
1902: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1903: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1904: /* } */
1905: /* in--; chain--; */
1906: /* while ( (*in != '\0')){ */
1907: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1908: /* *out++ = *in++; */
1909: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1910: /* } */
1911: /* *out='\0'; */
1912: /* out=s; */
1913: /* return out; */
1914: /* } */
1915: char *substrchaine(char *out, char *in, char *chain)
1916: {
1917: /* Substract chain 'chain' from 'in', return and output 'out' */
1.349 brouard 1918: /* in="V1+V1*age+age*age+V2", chain="+age*age" out="V1+V1*age+V2" */
1.187 brouard 1919:
1920: char *strloc;
1921:
1.349 brouard 1922: strcpy (out, in); /* out="V1+V1*age+age*age+V2" */
1923: strloc = strstr(out, chain); /* strloc points to out at "+age*age+V2" */
1924: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out); /* strloc=+age*age+V2 chain="+age*age", out="V1+V1*age+age*age+V2" */
1.187 brouard 1925: if(strloc != NULL){
1.349 brouard 1926: /* will affect out */ /* strloc+strlen(chain)=|+V2 = "V1+V1*age+age*age|+V2" */ /* Will also work in Unicodek */
1927: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1); /* move number of bytes corresponding to the length of "+V2" which is 3, plus one is 4 (including the null)*/
1928: /* equivalent to strcpy (strloc, strloc +strlen(chain)) if no overlap; Copies from "+V2" to V1+V1*age+ */
1.187 brouard 1929: }
1.349 brouard 1930: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out); /* strloc=+V2 chain="+age*age", in="V1+V1*age+age*age+V2", out="V1+V1*age+V2" */
1.187 brouard 1931: return out;
1932: }
1933:
1934:
1.145 brouard 1935: char *cutl(char *blocc, char *alocc, char *in, char occ)
1936: {
1.187 brouard 1937: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.349 brouard 1938: and alocc starts after first occurence of char 'occ' : ex cutl(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1939: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1940: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1941: */
1.160 brouard 1942: char *s, *t;
1.145 brouard 1943: t=in;s=in;
1944: while ((*in != occ) && (*in != '\0')){
1945: *alocc++ = *in++;
1946: }
1947: if( *in == occ){
1948: *(alocc)='\0';
1949: s=++in;
1950: }
1951:
1952: if (s == t) {/* occ not found */
1953: *(alocc-(in-s))='\0';
1954: in=s;
1955: }
1956: while ( *in != '\0'){
1957: *blocc++ = *in++;
1958: }
1959:
1960: *blocc='\0';
1961: return t;
1962: }
1.137 brouard 1963: char *cutv(char *blocc, char *alocc, char *in, char occ)
1964: {
1.187 brouard 1965: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1966: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1967: gives blocc="abcdef2ghi" and alocc="j".
1968: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1969: */
1970: char *s, *t;
1971: t=in;s=in;
1972: while (*in != '\0'){
1973: while( *in == occ){
1974: *blocc++ = *in++;
1975: s=in;
1976: }
1977: *blocc++ = *in++;
1978: }
1979: if (s == t) /* occ not found */
1980: *(blocc-(in-s))='\0';
1981: else
1982: *(blocc-(in-s)-1)='\0';
1983: in=s;
1984: while ( *in != '\0'){
1985: *alocc++ = *in++;
1986: }
1987:
1988: *alocc='\0';
1989: return s;
1990: }
1991:
1.126 brouard 1992: int nbocc(char *s, char occ)
1993: {
1994: int i,j=0;
1995: int lg=20;
1996: i=0;
1997: lg=strlen(s);
1998: for(i=0; i<= lg; i++) {
1.234 brouard 1999: if (s[i] == occ ) j++;
1.126 brouard 2000: }
2001: return j;
2002: }
2003:
1.349 brouard 2004: int nboccstr(char *textin, char *chain)
2005: {
2006: /* Counts the number of occurence of "chain" in string textin */
2007: /* in="+V7*V4+age*V2+age*V3+age*V4" chain="age" */
2008: char *strloc;
2009:
1.366 brouard 2010: int j=0;
1.349 brouard 2011:
2012: strloc=textin; /* strloc points to "^+V7*V4+age+..." in textin */
2013: for(;;) {
2014: strloc= strstr(strloc,chain); /* strloc points to first character of chain in textin if found. Example strloc points^ to "+V7*V4+^age" in textin */
2015: if(strloc != NULL){
2016: strloc = strloc+strlen(chain); /* strloc points to "+V7*V4+age^" in textin */
2017: j++;
2018: }else
2019: break;
2020: }
2021: return j;
2022:
2023: }
1.137 brouard 2024: /* void cutv(char *u,char *v, char*t, char occ) */
2025: /* { */
2026: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
2027: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
2028: /* gives u="abcdef2ghi" and v="j" *\/ */
2029: /* int i,lg,j,p=0; */
2030: /* i=0; */
2031: /* lg=strlen(t); */
2032: /* for(j=0; j<=lg-1; j++) { */
2033: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
2034: /* } */
1.126 brouard 2035:
1.137 brouard 2036: /* for(j=0; j<p; j++) { */
2037: /* (u[j] = t[j]); */
2038: /* } */
2039: /* u[p]='\0'; */
1.126 brouard 2040:
1.137 brouard 2041: /* for(j=0; j<= lg; j++) { */
2042: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
2043: /* } */
2044: /* } */
1.126 brouard 2045:
1.160 brouard 2046: #ifdef _WIN32
2047: char * strsep(char **pp, const char *delim)
2048: {
2049: char *p, *q;
2050:
2051: if ((p = *pp) == NULL)
2052: return 0;
2053: if ((q = strpbrk (p, delim)) != NULL)
2054: {
2055: *pp = q + 1;
2056: *q = '\0';
2057: }
2058: else
2059: *pp = 0;
2060: return p;
2061: }
2062: #endif
2063:
1.126 brouard 2064: /********************** nrerror ********************/
2065:
2066: void nrerror(char error_text[])
2067: {
2068: fprintf(stderr,"ERREUR ...\n");
2069: fprintf(stderr,"%s\n",error_text);
2070: exit(EXIT_FAILURE);
2071: }
2072: /*********************** vector *******************/
2073: double *vector(int nl, int nh)
2074: {
2075: double *v;
2076: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
2077: if (!v) nrerror("allocation failure in vector");
2078: return v-nl+NR_END;
2079: }
2080:
2081: /************************ free vector ******************/
2082: void free_vector(double*v, int nl, int nh)
2083: {
2084: free((FREE_ARG)(v+nl-NR_END));
2085: }
2086:
2087: /************************ivector *******************************/
2088: int *ivector(long nl,long nh)
2089: {
2090: int *v;
2091: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
2092: if (!v) nrerror("allocation failure in ivector");
2093: return v-nl+NR_END;
2094: }
2095:
2096: /******************free ivector **************************/
2097: void free_ivector(int *v, long nl, long nh)
2098: {
2099: free((FREE_ARG)(v+nl-NR_END));
2100: }
2101:
2102: /************************lvector *******************************/
2103: long *lvector(long nl,long nh)
2104: {
2105: long *v;
2106: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
2107: if (!v) nrerror("allocation failure in ivector");
2108: return v-nl+NR_END;
2109: }
2110:
2111: /******************free lvector **************************/
2112: void free_lvector(long *v, long nl, long nh)
2113: {
2114: free((FREE_ARG)(v+nl-NR_END));
2115: }
2116:
2117: /******************* imatrix *******************************/
2118: int **imatrix(long nrl, long nrh, long ncl, long nch)
2119: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
2120: {
2121: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
2122: int **m;
2123:
2124: /* allocate pointers to rows */
2125: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
2126: if (!m) nrerror("allocation failure 1 in matrix()");
2127: m += NR_END;
2128: m -= nrl;
2129:
2130:
2131: /* allocate rows and set pointers to them */
2132: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
2133: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2134: m[nrl] += NR_END;
2135: m[nrl] -= ncl;
2136:
2137: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
2138:
2139: /* return pointer to array of pointers to rows */
2140: return m;
2141: }
2142:
2143: /****************** free_imatrix *************************/
1.366 brouard 2144: /* void free_imatrix(m,nrl,nrh,ncl,nch); */
2145: /* int **m; */
2146: /* long nch,ncl,nrh,nrl; */
2147: void free_imatrix(int **m,long nrl, long nrh, long ncl, long nch)
2148: /* free an int matrix allocated by imatrix() */
2149: {
2150: free((FREE_ARG) (m[nrl]+ncl-NR_END));
2151: free((FREE_ARG) (m+nrl-NR_END));
2152: }
1.126 brouard 2153:
2154: /******************* matrix *******************************/
2155: double **matrix(long nrl, long nrh, long ncl, long nch)
2156: {
2157: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
2158: double **m;
2159:
2160: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2161: if (!m) nrerror("allocation failure 1 in matrix()");
2162: m += NR_END;
2163: m -= nrl;
2164:
2165: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2166: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2167: m[nrl] += NR_END;
2168: m[nrl] -= ncl;
2169:
2170: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2171: return m;
1.145 brouard 2172: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
2173: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
2174: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 2175: */
2176: }
2177:
2178: /*************************free matrix ************************/
2179: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
2180: {
2181: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2182: free((FREE_ARG)(m+nrl-NR_END));
2183: }
2184:
2185: /******************* ma3x *******************************/
2186: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
2187: {
2188: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
2189: double ***m;
2190:
2191: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2192: if (!m) nrerror("allocation failure 1 in matrix()");
2193: m += NR_END;
2194: m -= nrl;
2195:
2196: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2197: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2198: m[nrl] += NR_END;
2199: m[nrl] -= ncl;
2200:
2201: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2202:
2203: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
2204: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
2205: m[nrl][ncl] += NR_END;
2206: m[nrl][ncl] -= nll;
2207: for (j=ncl+1; j<=nch; j++)
2208: m[nrl][j]=m[nrl][j-1]+nlay;
2209:
2210: for (i=nrl+1; i<=nrh; i++) {
2211: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
2212: for (j=ncl+1; j<=nch; j++)
2213: m[i][j]=m[i][j-1]+nlay;
2214: }
2215: return m;
2216: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
2217: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
2218: */
2219: }
2220:
2221: /*************************free ma3x ************************/
2222: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
2223: {
2224: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
2225: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2226: free((FREE_ARG)(m+nrl-NR_END));
2227: }
2228:
2229: /*************** function subdirf ***********/
2230: char *subdirf(char fileres[])
2231: {
2232: /* Caution optionfilefiname is hidden */
2233: strcpy(tmpout,optionfilefiname);
2234: strcat(tmpout,"/"); /* Add to the right */
2235: strcat(tmpout,fileres);
2236: return tmpout;
2237: }
2238:
2239: /*************** function subdirf2 ***********/
2240: char *subdirf2(char fileres[], char *preop)
2241: {
1.314 brouard 2242: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
2243: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 2244: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 2245: /* Caution optionfilefiname is hidden */
2246: strcpy(tmpout,optionfilefiname);
2247: strcat(tmpout,"/");
2248: strcat(tmpout,preop);
2249: strcat(tmpout,fileres);
2250: return tmpout;
2251: }
2252:
2253: /*************** function subdirf3 ***********/
2254: char *subdirf3(char fileres[], char *preop, char *preop2)
2255: {
2256:
2257: /* Caution optionfilefiname is hidden */
2258: strcpy(tmpout,optionfilefiname);
2259: strcat(tmpout,"/");
2260: strcat(tmpout,preop);
2261: strcat(tmpout,preop2);
2262: strcat(tmpout,fileres);
2263: return tmpout;
2264: }
1.213 brouard 2265:
2266: /*************** function subdirfext ***********/
2267: char *subdirfext(char fileres[], char *preop, char *postop)
2268: {
2269:
2270: strcpy(tmpout,preop);
2271: strcat(tmpout,fileres);
2272: strcat(tmpout,postop);
2273: return tmpout;
2274: }
1.126 brouard 2275:
1.213 brouard 2276: /*************** function subdirfext3 ***********/
2277: char *subdirfext3(char fileres[], char *preop, char *postop)
2278: {
2279:
2280: /* Caution optionfilefiname is hidden */
2281: strcpy(tmpout,optionfilefiname);
2282: strcat(tmpout,"/");
2283: strcat(tmpout,preop);
2284: strcat(tmpout,fileres);
2285: strcat(tmpout,postop);
2286: return tmpout;
2287: }
2288:
1.162 brouard 2289: char *asc_diff_time(long time_sec, char ascdiff[])
2290: {
2291: long sec_left, days, hours, minutes;
2292: days = (time_sec) / (60*60*24);
2293: sec_left = (time_sec) % (60*60*24);
2294: hours = (sec_left) / (60*60) ;
2295: sec_left = (sec_left) %(60*60);
2296: minutes = (sec_left) /60;
2297: sec_left = (sec_left) % (60);
2298: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2299: return ascdiff;
2300: }
2301:
1.126 brouard 2302: /***************** f1dim *************************/
2303: extern int ncom;
2304: extern double *pcom,*xicom;
2305: extern double (*nrfunc)(double []);
2306:
2307: double f1dim(double x)
2308: {
2309: int j;
2310: double f;
2311: double *xt;
2312:
2313: xt=vector(1,ncom);
2314: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2315: f=(*nrfunc)(xt);
2316: free_vector(xt,1,ncom);
2317: return f;
2318: }
2319:
2320: /*****************brent *************************/
2321: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2322: {
2323: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2324: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2325: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2326: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2327: * returned function value.
2328: */
1.126 brouard 2329: int iter;
2330: double a,b,d,etemp;
1.159 brouard 2331: double fu=0,fv,fw,fx;
1.164 brouard 2332: double ftemp=0.;
1.126 brouard 2333: double p,q,r,tol1,tol2,u,v,w,x,xm;
2334: double e=0.0;
2335:
2336: a=(ax < cx ? ax : cx);
2337: b=(ax > cx ? ax : cx);
2338: x=w=v=bx;
2339: fw=fv=fx=(*f)(x);
2340: for (iter=1;iter<=ITMAX;iter++) {
2341: xm=0.5*(a+b);
2342: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2343: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2344: printf(".");fflush(stdout);
2345: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2346: #ifdef DEBUGBRENT
1.126 brouard 2347: printf("br %d,x=%.10e xm=%.10e b=%.10e a=%.10e tol=%.10e tol1=%.10e tol2=%.10e x-xm=%.10e fx=%.12e fu=%.12e,fw=%.12e,ftemp=%.12e,ftol=%.12e\n",iter,x,xm,b,a,tol,tol1,tol2,(x-xm),fx,fu,fw,ftemp,ftol);
2348: fprintf(ficlog,"br %d,x=%.10e xm=%.10e b=%.10e a=%.10e tol=%.10e tol1=%.10e tol2=%.10e x-xm=%.10e fx=%.12e fu=%.12e,fw=%.12e,ftemp=%.12e,ftol=%.12e\n",iter,x,xm,b,a,tol,tol1,tol2,(x-xm),fx,fu,fw,ftemp,ftol);
2349: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2350: #endif
2351: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2352: *xmin=x;
2353: return fx;
2354: }
2355: ftemp=fu;
2356: if (fabs(e) > tol1) {
2357: r=(x-w)*(fx-fv);
2358: q=(x-v)*(fx-fw);
2359: p=(x-v)*q-(x-w)*r;
2360: q=2.0*(q-r);
2361: if (q > 0.0) p = -p;
2362: q=fabs(q);
2363: etemp=e;
2364: e=d;
2365: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2366: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2367: else {
1.224 brouard 2368: d=p/q;
2369: u=x+d;
2370: if (u-a < tol2 || b-u < tol2)
2371: d=SIGN(tol1,xm-x);
1.126 brouard 2372: }
2373: } else {
2374: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2375: }
2376: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2377: fu=(*f)(u);
2378: if (fu <= fx) {
2379: if (u >= x) a=x; else b=x;
2380: SHFT(v,w,x,u)
1.183 brouard 2381: SHFT(fv,fw,fx,fu)
2382: } else {
2383: if (u < x) a=u; else b=u;
2384: if (fu <= fw || w == x) {
1.224 brouard 2385: v=w;
2386: w=u;
2387: fv=fw;
2388: fw=fu;
1.183 brouard 2389: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2390: v=u;
2391: fv=fu;
1.183 brouard 2392: }
2393: }
1.126 brouard 2394: }
2395: nrerror("Too many iterations in brent");
2396: *xmin=x;
2397: return fx;
2398: }
2399:
2400: /****************** mnbrak ***********************/
2401:
2402: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2403: double (*func)(double))
1.183 brouard 2404: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2405: the downhill direction (defined by the function as evaluated at the initial points) and returns
2406: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2407: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2408: */
1.126 brouard 2409: double ulim,u,r,q, dum;
2410: double fu;
1.187 brouard 2411:
1.366 brouard 2412: /* double scale=10.; */
2413: /* int iterscale=0; */
1.187 brouard 2414:
2415: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2416: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2417:
2418:
2419: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2420: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2421: /* *bx = *ax - (*ax - *bx)/scale; */
2422: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2423: /* } */
2424:
1.126 brouard 2425: if (*fb > *fa) {
2426: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2427: SHFT(dum,*fb,*fa,dum)
2428: }
1.126 brouard 2429: *cx=(*bx)+GOLD*(*bx-*ax);
2430: *fc=(*func)(*cx);
1.183 brouard 2431: #ifdef DEBUG
1.224 brouard 2432: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2433: fprintf(ficlog,"mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1.183 brouard 2434: #endif
1.224 brouard 2435: while (*fb > *fc) { /* Declining a,b,c with fa> fb > fc. If fc=inf it exits and if flat fb=fc it exits too.*/
1.126 brouard 2436: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2437: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2438: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2439: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2440: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2441: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2442: fu=(*func)(u);
1.163 brouard 2443: #ifdef DEBUG
2444: /* f(x)=A(x-u)**2+f(u) */
2445: double A, fparabu;
2446: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2447: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2448: printf("\nmnbrak (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf), (*u=%.12f, fu=%.12lf, fparabu=%.12f, q=%lf < %lf=r)\n",*ax,*fa,*bx,*fb,*cx,*fc,u,fu, fparabu,q,r);
2449: fprintf(ficlog,"\nmnbrak (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf), (*u=%.12f, fu=%.12lf, fparabu=%.12f, q=%lf < %lf=r)\n",*ax,*fa,*bx,*fb,*cx,*fc,u,fu, fparabu,q,r);
1.183 brouard 2450: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2451: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2452: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2453: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2454: #endif
1.184 brouard 2455: #ifdef MNBRAKORIGINAL
1.183 brouard 2456: #else
1.191 brouard 2457: /* if (fu > *fc) { */
2458: /* #ifdef DEBUG */
2459: /* printf("mnbrak4 fu > fc \n"); */
2460: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2461: /* #endif */
2462: /* /\* SHFT(u,*cx,*cx,u) /\\* ie a=c, c=u and u=c; in that case, next SHFT(a,b,c,u) will give a=b=b, b=c=u, c=u=c and *\\/ *\/ */
2463: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2464: /* dum=u; /\* Shifting c and u *\/ */
2465: /* u = *cx; */
2466: /* *cx = dum; */
2467: /* dum = fu; */
2468: /* fu = *fc; */
2469: /* *fc =dum; */
2470: /* } else { /\* end *\/ */
2471: /* #ifdef DEBUG */
2472: /* printf("mnbrak3 fu < fc \n"); */
2473: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2474: /* #endif */
2475: /* dum=u; /\* Shifting c and u *\/ */
2476: /* u = *cx; */
2477: /* *cx = dum; */
2478: /* dum = fu; */
2479: /* fu = *fc; */
2480: /* *fc =dum; */
2481: /* } */
1.224 brouard 2482: #ifdef DEBUGMNBRAK
2483: double A, fparabu;
2484: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2485: fparabu= *fa - A*(*ax-u)*(*ax-u);
2486: printf("\nmnbrak35 ax=%lf fa=%lf bx=%lf fb=%lf, u=%lf fp=%lf fu=%lf < or >= fc=%lf cx=%lf, q=%lf < %lf=r \n",*ax, *fa, *bx,*fb,u,fparabu,fu,*fc,*cx,q,r);
2487: fprintf(ficlog,"\nmnbrak35 ax=%lf fa=%lf bx=%lf fb=%lf, u=%lf fp=%lf fu=%lf < or >= fc=%lf cx=%lf, q=%lf < %lf=r \n",*ax, *fa, *bx,*fb,u,fparabu,fu,*fc,*cx,q,r);
1.183 brouard 2488: #endif
1.191 brouard 2489: dum=u; /* Shifting c and u */
2490: u = *cx;
2491: *cx = dum;
2492: dum = fu;
2493: fu = *fc;
2494: *fc =dum;
1.183 brouard 2495: #endif
1.162 brouard 2496: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2497: #ifdef DEBUG
1.224 brouard 2498: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2499: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2500: #endif
1.126 brouard 2501: fu=(*func)(u);
2502: if (fu < *fc) {
1.183 brouard 2503: #ifdef DEBUG
1.224 brouard 2504: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2505: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2506: #endif
2507: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2508: SHFT(*fb,*fc,fu,(*func)(u))
2509: #ifdef DEBUG
2510: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2511: #endif
2512: }
1.162 brouard 2513: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2514: #ifdef DEBUG
1.224 brouard 2515: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2516: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2517: #endif
1.126 brouard 2518: u=ulim;
2519: fu=(*func)(u);
1.183 brouard 2520: } else { /* u could be left to b (if r > q parabola has a maximum) */
2521: #ifdef DEBUG
1.224 brouard 2522: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2523: fprintf(ficlog,"\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1.183 brouard 2524: #endif
1.126 brouard 2525: u=(*cx)+GOLD*(*cx-*bx);
2526: fu=(*func)(u);
1.224 brouard 2527: #ifdef DEBUG
2528: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2529: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2530: #endif
1.183 brouard 2531: } /* end tests */
1.126 brouard 2532: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2533: SHFT(*fa,*fb,*fc,fu)
2534: #ifdef DEBUG
1.224 brouard 2535: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2536: fprintf(ficlog, "\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1.183 brouard 2537: #endif
2538: } /* end while; ie return (a, b, c, fa, fb, fc) such that a < b < c with f(a) > f(b) and fb < f(c) */
1.126 brouard 2539: }
2540:
2541: /*************** linmin ************************/
1.162 brouard 2542: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2543: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2544: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2545: the value of func at the returned location p . This is actually all accomplished by calling the
2546: routines mnbrak and brent .*/
1.126 brouard 2547: int ncom;
2548: double *pcom,*xicom;
2549: double (*nrfunc)(double []);
2550:
1.224 brouard 2551: #ifdef LINMINORIGINAL
1.126 brouard 2552: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2553: #else
2554: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2555: #endif
1.126 brouard 2556: {
2557: double brent(double ax, double bx, double cx,
2558: double (*f)(double), double tol, double *xmin);
2559: double f1dim(double x);
2560: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2561: double *fc, double (*func)(double));
2562: int j;
2563: double xx,xmin,bx,ax;
2564: double fx,fb,fa;
1.187 brouard 2565:
1.203 brouard 2566: #ifdef LINMINORIGINAL
2567: #else
2568: double scale=10., axs, xxs; /* Scale added for infinity */
2569: #endif
2570:
1.126 brouard 2571: ncom=n;
2572: pcom=vector(1,n);
2573: xicom=vector(1,n);
2574: nrfunc=func;
2575: for (j=1;j<=n;j++) {
2576: pcom[j]=p[j];
1.202 brouard 2577: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2578: }
1.187 brouard 2579:
1.203 brouard 2580: #ifdef LINMINORIGINAL
2581: xx=1.;
2582: #else
2583: axs=0.0;
2584: xxs=1.;
2585: do{
2586: xx= xxs;
2587: #endif
1.187 brouard 2588: ax=0.;
2589: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2590: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2591: /* xt[x,j]=pcom[j]+x*xicom[j] f(ax) = f(xt(a,j=1,n)) = f(p(j) + 0 * xi(j)) and f(xx) = f(xt(x, j=1,n)) = f(p(j) + 1 * xi(j)) */
2592: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2593: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2594: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2595: /* Find a bracket a,x,b in direction n=xi ie xicom, order may change. Scale is [0:xxs*xi[j]] et non plus [0:xi[j]]*/
1.203 brouard 2596: #ifdef LINMINORIGINAL
2597: #else
2598: if (fx != fx){
1.224 brouard 2599: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2600: printf("|");
2601: fprintf(ficlog,"|");
1.203 brouard 2602: #ifdef DEBUGLINMIN
1.224 brouard 2603: printf("\nLinmin NAN : input [axs=%lf:xxs=%lf], mnbrak outputs fx=%lf <(fb=%lf and fa=%lf) with xx=%lf in [ax=%lf:bx=%lf] \n", axs, xxs, fx,fb, fa, xx, ax, bx);
1.203 brouard 2604: #endif
2605: }
1.224 brouard 2606: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2607: #endif
2608:
1.191 brouard 2609: #ifdef DEBUGLINMIN
2610: printf("\nLinmin after mnbrak: ax=%12.7f xx=%12.7f bx=%12.7f fa=%12.2f fx=%12.2f fb=%12.2f\n", ax,xx,bx,fa,fx,fb);
1.202 brouard 2611: fprintf(ficlog,"\nLinmin after mnbrak: ax=%12.7f xx=%12.7f bx=%12.7f fa=%12.2f fx=%12.2f fb=%12.2f\n", ax,xx,bx,fa,fx,fb);
1.191 brouard 2612: #endif
1.224 brouard 2613: #ifdef LINMINORIGINAL
2614: #else
1.317 brouard 2615: if(fb == fx){ /* Flat function in the direction */
2616: xmin=xx;
1.224 brouard 2617: *flat=1;
1.317 brouard 2618: }else{
1.224 brouard 2619: *flat=0;
2620: #endif
2621: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2622: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2623: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2624: /* fmin = f(p[j] + xmin * xi[j]) */
2625: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2626: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2627: #ifdef DEBUG
1.224 brouard 2628: printf("retour brent from bracket (a=%lf fa=%lf, xx=%lf fx=%lf, b=%lf fb=%lf): fret=%lf xmin=%lf\n",ax,fa,xx,fx,bx,fb,*fret,xmin);
2629: fprintf(ficlog,"retour brent from bracket (a=%lf fa=%lf, xx=%lf fx=%lf, b=%lf fb=%lf): fret=%lf xmin=%lf\n",ax,fa,xx,fx,bx,fb,*fret,xmin);
2630: #endif
2631: #ifdef LINMINORIGINAL
2632: #else
2633: }
1.126 brouard 2634: #endif
1.191 brouard 2635: #ifdef DEBUGLINMIN
2636: printf("linmin end ");
1.202 brouard 2637: fprintf(ficlog,"linmin end ");
1.191 brouard 2638: #endif
1.126 brouard 2639: for (j=1;j<=n;j++) {
1.203 brouard 2640: #ifdef LINMINORIGINAL
2641: xi[j] *= xmin;
2642: #else
2643: #ifdef DEBUGLINMIN
2644: if(xxs <1.0)
2645: printf(" before xi[%d]=%12.8f", j,xi[j]);
2646: #endif
2647: xi[j] *= xmin*xxs; /* xi rescaled by xmin and number of loops: if xmin=-1.237 and xi=(1,0,...,0) xi=(-1.237,0,...,0) */
2648: #ifdef DEBUGLINMIN
2649: if(xxs <1.0)
2650: printf(" after xi[%d]=%12.8f, xmin=%12.8f, ax=%12.8f, xx=%12.8f, bx=%12.8f, xxs=%12.8f", j,xi[j], xmin, ax, xx, bx,xxs );
2651: #endif
2652: #endif
1.187 brouard 2653: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2654: }
1.191 brouard 2655: #ifdef DEBUGLINMIN
1.203 brouard 2656: printf("\n");
1.191 brouard 2657: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2658: fprintf(ficlog,"Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.191 brouard 2659: for (j=1;j<=n;j++) {
1.202 brouard 2660: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2661: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2662: if(j % ncovmodel == 0){
1.191 brouard 2663: printf("\n");
1.202 brouard 2664: fprintf(ficlog,"\n");
2665: }
1.191 brouard 2666: }
1.203 brouard 2667: #else
1.191 brouard 2668: #endif
1.126 brouard 2669: free_vector(xicom,1,n);
2670: free_vector(pcom,1,n);
2671: }
2672:
1.359 brouard 2673: /**** praxis gegen ****/
2674:
2675: /* This has been tested by Visual C from Microsoft and works */
2676: /* meaning tha valgrind could be wrong */
2677: /*********************************************************************/
2678: /* f u n c t i o n p r a x i s */
2679: /* */
2680: /* praxis is a general purpose routine for the minimization of a */
2681: /* function in several variables. the algorithm used is a modifi- */
2682: /* cation of conjugate gradient search method by powell. the changes */
2683: /* are due to r.p. brent, who gives an algol-w program, which served */
2684: /* as a basis for this function. */
2685: /* */
2686: /* references: */
2687: /* - powell, m.j.d., 1964. an efficient method for finding */
2688: /* the minimum of a function in several variables without */
2689: /* calculating derivatives, computer journal, 7, 155-162 */
2690: /* - brent, r.p., 1973. algorithms for minimization without */
2691: /* derivatives, prentice hall, englewood cliffs. */
2692: /* */
2693: /* problems, suggestions or improvements are always wellcome */
2694: /* karl gegenfurtner 07/08/87 */
2695: /* c - version */
2696: /*********************************************************************/
2697: /* */
2698: /* usage: min = praxis(tol, macheps, h, n, prin, x, func) */
2699: /* macheps has been suppressed because it is replaced by DBL_EPSILON */
2700: /* and if it was an argument of praxis (as it is in original brent) */
2701: /* it should be declared external */
2702: /* usage: min = praxis(tol, h, n, prin, x, func) */
2703: /* was min = praxis(fun, x, n); */
2704: /* */
2705: /* fun the function to be minimized. fun is called from */
2706: /* praxis with x and n as arguments */
2707: /* x a double array containing the initial guesses for */
2708: /* the minimum, which will contain the solution on */
2709: /* return */
2710: /* n an integer specifying the number of unknown */
2711: /* parameters */
2712: /* min praxis returns the least calculated value of fun */
2713: /* */
2714: /* some additional global variables control some more aspects of */
2715: /* the inner workings of praxis. setting them is optional, they */
2716: /* are all set to some reasonable default values given below. */
2717: /* */
2718: /* prin controls the printed output from the routine. */
2719: /* 0 -> no output */
2720: /* 1 -> print only starting and final values */
2721: /* 2 -> detailed map of the minimization process */
2722: /* 3 -> print also eigenvalues and vectors of the */
2723: /* search directions */
2724: /* the default value is 1 */
2725: /* tol is the tolerance allowed for the precision of the */
2726: /* solution. praxis returns if the criterion */
2727: /* 2 * ||x[k]-x[k-1]|| <= sqrt(macheps) * ||x[k]|| + tol */
2728: /* is fulfilled more than ktm times. */
2729: /* the default value depends on the machine precision */
2730: /* ktm see just above. default is 1, and a value of 4 leads */
2731: /* to a very(!) cautious stopping criterion. */
2732: /* h0 or step is a steplength parameter and should be set equal */
2733: /* to the expected distance from the solution. */
2734: /* exceptionally small or large values of step lead to */
2735: /* slower convergence on the first few iterations */
2736: /* the default value for step is 1.0 */
2737: /* scbd is a scaling parameter. 1.0 is the default and */
2738: /* indicates no scaling. if the scales for the different */
2739: /* parameters are very different, scbd should be set to */
2740: /* a value of about 10.0. */
2741: /* illc should be set to true (1) if the problem is known to */
2742: /* be ill-conditioned. the default is false (0). this */
2743: /* variable is automatically set, when praxis finds */
2744: /* the problem to be ill-conditioned during iterations. */
2745: /* maxfun is the maximum number of calls to fun allowed. praxis */
2746: /* will return after maxfun calls to fun even when the */
2747: /* minimum is not yet found. the default value of 0 */
2748: /* indicates no limit on the number of calls. */
2749: /* this return condition is only checked every n */
2750: /* iterations. */
2751: /* */
2752: /*********************************************************************/
2753:
2754: #include <math.h>
2755: #include <stdio.h>
2756: #include <stdlib.h>
2757: #include <float.h> /* for DBL_EPSILON */
2758: /* #include "machine.h" */
2759:
2760:
2761: /* extern void minfit(int n, double eps, double tol, double **ab, double q[]); */
2762: /* extern void minfit(int n, double eps, double tol, double ab[N][N], double q[]); */
2763: /* control parameters */
2764: /* control parameters */
2765: #define SQREPSILON 1.0e-19
2766: /* #define EPSILON 1.0e-8 */ /* in main */
2767:
2768: double tol = SQREPSILON,
2769: scbd = 1.0,
2770: step = 1.0;
2771: int ktm = 1,
2772: /* prin = 2, */
2773: maxfun = 0,
2774: illc = 0;
2775:
2776: /* some global variables */
2777: static int i, j, k, k2, nl, nf, kl, kt;
2778: /* static double s; */
2779: double sl, dn, dmin,
2780: fx, f1, lds, ldt, sf, df,
2781: qf1, qd0, qd1, qa, qb, qc,
2782: m2, m4, small_windows, vsmall, large,
2783: vlarge, ldfac, t2;
2784: /* static double d[N], y[N], z[N], */
2785: /* q0[N], q1[N], v[N][N]; */
2786:
2787: static double *d, *y, *z;
2788: static double *q0, *q1, **v;
2789: double *tflin; /* used in flin: return (*fun)(tflin, n); */
2790: double *e; /* used in minfit, don't konw how to free memory and thus made global */
2791: /* static double s, sl, dn, dmin, */
2792: /* fx, f1, lds, ldt, sf, df, */
2793: /* qf1, qd0, qd1, qa, qb, qc, */
2794: /* m2, m4, small, vsmall, large, */
2795: /* vlarge, ldfac, t2; */
2796: /* static double d[N], y[N], z[N], */
2797: /* q0[N], q1[N], v[N][N]; */
2798:
2799: /* these will be set by praxis to point to it's arguments */
2800: static int prin; /* added */
2801: static int n;
2802: static double *x;
1.366 brouard 2803: static double (*fun)(double *x); /* New for clang */
2804: /* static double (*fun)(); */
1.359 brouard 2805: /* static double (*fun)(double *x, int n); */
2806:
2807: /* these will be set by praxis to the global control parameters */
2808: /* static double h, macheps, t; */
2809: extern double macheps;
2810: static double h;
2811: static double t;
2812:
2813: static double
2814: drandom() /* return random no between 0 and 1 */
2815: {
2816: return (double)(rand()%(8192*2))/(double)(8192*2);
2817: }
2818:
2819: static void sort() /* d and v in descending order */
2820: {
2821: int k, i, j;
2822: double s;
2823:
2824: for (i=1; i<=n-1; i++) {
2825: k = i; s = d[i];
2826: for (j=i+1; j<=n; j++) {
2827: if (d[j] > s) {
2828: k = j;
2829: s = d[j];
2830: }
2831: }
2832: if (k > i) {
2833: d[k] = d[i];
2834: d[i] = s;
2835: for (j=1; j<=n; j++) {
2836: s = v[j][i];
2837: v[j][i] = v[j][k];
2838: v[j][k] = s;
2839: }
2840: }
2841: }
2842: }
2843:
2844: double randbrent ( int *naught )
2845: {
2846: double ran1, ran3[127], half;
2847: int ran2, q, r, i, j;
2848: int init=0; /* false */
2849: double rr;
2850: /* REAL*8 RAN1,RAN3(127),HALF */
2851:
2852: /* INTEGER RAN2,Q,R */
2853: /* LOGICAL INIT */
2854: /* DATA INIT/.FALSE./ */
2855: /* IF (INIT) GO TO 3 */
2856: if(!init){
2857: /* R = MOD(NAUGHT,8190) + 1 *//* 1804289383 rand () */
2858: r = *naught % 8190 + 1;/* printf(" naught r %d %d",*naught,r); */
2859: ran2=127;
2860: for(i=ran2; i>0; i--){
2861: /* RAN2 = 128 */
2862: /* DO 2 I=1,127 */
2863: ran2 = ran2-1;
2864: /* RAN2 = RAN2 - 1 */
2865: ran1 = -pow(2.0,55);
2866: /* RAN1 = -2.D0**55 */
2867: /* DO 1 J=1,7 */
2868: for(j=1; j<=7;j++){
2869: /* R = MOD(1756*R,8191) */
2870: r = (1756*r) % 8191;/* printf(" i=%d (1756*r)%8191=%d",j,r); */
2871: q=r/32;
2872: /* Q = R/32 */
2873: /* 1 RAN1 = (RAN1 + Q)*(1.0D0/256) */
2874: ran1 =(ran1+q)*(1.0/256);
2875: }
2876: /* 2 RAN3(RAN2) = RAN1 */
2877: ran3[ran2] = ran1; /* printf(" ran2=%d ran1=%.7g \n",ran2,ran1); */
2878: }
2879: /* INIT = .TRUE. */
2880: init=1;
2881: /* 3 IF (RAN2.EQ.1) RAN2 = 128 */
2882: }
2883: if(ran2 == 0) ran2 = 126;
2884: else ran2 = ran2 -1;
2885: /* RAN2 = RAN2 - 1 */
2886: /* RAN1 = RAN1 + RAN3(RAN2) */
2887: ran1 = ran1 + ran3[ran2];/* printf("BIS ran2=%d ran1=%.7g \n",ran2,ran1); */
2888: half= 0.5;
2889: /* HALF = .5D0 */
2890: /* IF (RAN1.GE.0.D0) HALF = -HALF */
2891: if(ran1 >= 0.) half =-half;
2892: ran1 = ran1 +half;
2893: ran3[ran2] = ran1;
2894: rr= ran1+0.5;
2895: /* RAN1 = RAN1 + HALF */
2896: /* RAN3(RAN2) = RAN1 */
2897: /* RANDOM = RAN1 + .5D0 */
2898: /* r = ( ( double ) ( *seed ) ) * 4.656612875E-10; */
2899: return rr;
2900: }
2901: static void matprint(char *s, double **v, int m, int n)
2902: /* char *s; */
2903: /* double v[N][N]; */
2904: {
2905: #define INCX 8
2906: int i;
2907:
2908: int i2hi;
2909: int ihi;
2910: int ilo;
2911: int i2lo;
2912: int jlo=1;
2913: int j;
2914: int j2hi;
2915: int jhi;
2916: int j2lo;
2917: ilo=1;
2918: ihi=n;
2919: jlo=1;
2920: jhi=n;
2921:
2922: printf ("\n" );
2923: printf ("%s\n", s );
2924: for ( j2lo = jlo; j2lo <= jhi; j2lo = j2lo + INCX )
2925: {
2926: j2hi = j2lo + INCX - 1;
2927: if ( n < j2hi )
2928: {
2929: j2hi = n;
2930: }
2931: if ( jhi < j2hi )
2932: {
2933: j2hi = jhi;
2934: }
2935:
2936: /* fprintf ( ficlog, "\n" ); */
2937: printf ("\n" );
2938: /*
2939: For each column J in the current range...
2940:
2941: Write the header.
2942: */
2943: /* fprintf ( ficlog, " Col: "); */
2944: printf ("Col:");
2945: for ( j = j2lo; j <= j2hi; j++ )
2946: {
2947: /* fprintf ( ficlog, " %7d ", j - 1 ); */
2948: /* printf (" %9d ", j - 1 ); */
2949: printf (" %9d ", j );
2950: }
2951: /* fprintf ( ficlog, "\n" ); */
2952: /* fprintf ( ficlog, " Row\n" ); */
2953: /* fprintf ( ficlog, "\n" ); */
2954: printf ("\n" );
2955: printf (" Row\n" );
2956: printf ("\n" );
2957: /*
2958: Determine the range of the rows in this strip.
2959: */
2960: if ( 1 < ilo ){
2961: i2lo = ilo;
2962: }else{
2963: i2lo = 1;
2964: }
2965: if ( m < ihi ){
2966: i2hi = m;
2967: }else{
2968: i2hi = ihi;
2969: }
2970:
2971: for ( i = i2lo; i <= i2hi; i++ ){
2972: /*
2973: Print out (up to) 5 entries in row I, that lie in the current strip.
2974: */
2975: /* fprintf ( ficlog, "%5d:", i - 1 ); */
2976: /* printf ("%5d:", i - 1 ); */
2977: printf ("%5d:", i );
2978: for ( j = j2lo; j <= j2hi; j++ )
2979: {
2980: /* fprintf ( ficlog, " %14g", a[i-1+(j-1)*m] ); */
2981: /* printf ("%14.7g ", a[i-1+(j-1)*m] ); */
2982: /* printf("%14.7f ", v[i-1][j-1]); */
2983: printf("%14.7f ", v[i][j]);
2984: /* fprintf ( stdout, " %14g", a[i-1+(j-1)*m] ); */
2985: }
2986: /* fprintf ( ficlog, "\n" ); */
2987: printf ("\n" );
2988: }
2989: }
2990:
2991: /* printf("%s\n", s); */
2992: /* for (k=0; k<n; k++) { */
2993: /* for (i=0; i<n; i++) { */
2994: /* /\* printf("%20.10e ", v[k][i]); *\/ */
2995: /* } */
2996: /* printf("\n"); */
2997: /* } */
2998: #undef INCX
2999: }
3000:
3001: void vecprint(char *s, double *x, int n)
3002: /* char *s; */
3003: /* double x[N]; */
3004: {
3005: int i=0;
3006:
3007: printf(" %s", s);
3008: /* for (i=0; i<n; i++) */
3009: for (i=1; i<=n; i++)
3010: printf (" %14.7g", x[i] );
3011: /* printf(" %8d: %14g\n", i, x[i]); */
3012: printf ("\n" );
3013: }
3014:
3015: static void print() /* print a line of traces */
3016: {
3017:
3018:
3019: printf("\n");
3020: /* printf("... chi square reduced to ... %20.10e\n", fx); */
3021: /* printf("... after %u function calls ...\n", nf); */
3022: /* printf("... including %u linear searches ...\n", nl); */
3023: printf("%10d %10d%14.7g",nl, nf, fx);
3024: vecprint("... current values of x ...", x, n);
3025: }
3026: /* static void print2(int n, double *x, int prin, double fx, int nf, int nl) */ /* print a line of traces */
3027: static void print2() /* print a line of traces */
3028: {
1.366 brouard 3029: int i; /* double fmin=0.; */
1.359 brouard 3030:
3031: /* printf("\n"); */
3032: /* printf("... chi square reduced to ... %20.10e\n", fx); */
3033: /* printf("... after %u function calls ...\n", nf); */
3034: /* printf("... including %u linear searches ...\n", nl); */
3035: /* printf("%10d %10d%14.7g",nl, nf, fx); */
1.363 brouard 3036: /* printf ( "\n" ); */
1.359 brouard 3037: printf ( " Linear searches %d", nl );
1.364 brouard 3038: fprintf (ficlog, " Linear searches %d", nl );
1.359 brouard 3039: /* printf ( " Linear searches %d\n", nl ); */
3040: /* printf ( " Function evaluations %d\n", nf ); */
3041: /* printf ( " Function value FX = %g\n", fx ); */
3042: printf ( " Function evaluations %d", nf );
3043: printf ( " Function value FX = %.12lf\n", fx );
1.363 brouard 3044: fprintf (ficlog, " Function evaluations %d", nf );
3045: fprintf (ficlog, " Function value FX = %.12lf\n", fx );
1.359 brouard 3046: #ifdef DEBUGPRAX
3047: printf("n=%d prin=%d\n",n,prin);
3048: #endif
1.363 brouard 3049: /* if(fx <= fmin) printf(" UNDEFINED "); else printf("%14.7g",log(fx-fmin)); */
1.359 brouard 3050: if ( n <= 4 || 2 < prin )
3051: {
3052: /* for(i=1;i<=n;i++)printf("%14.7g",x[i-1]); */
1.363 brouard 3053: for(i=1;i<=n;i++){
1.364 brouard 3054: printf(" %14.7g",x[i]);
3055: fprintf(ficlog," %14.7g",x[i]);
1.363 brouard 3056: }
1.359 brouard 3057: /* r8vec_print ( n, x, " X:" ); */
3058: }
3059: printf("\n");
1.363 brouard 3060: fprintf(ficlog,"\n");
1.359 brouard 3061: }
3062:
3063:
3064: /* #ifdef MSDOS */
3065: /* static double tflin[N]; */
3066: /* #endif */
3067:
3068: static double flin(double l, int j)
3069: /* double l; */
3070: {
3071: int i;
3072: /* #ifndef MSDOS */
3073: /* double tflin[N]; */
3074: /* #endif */
3075: /* double *tflin; */ /* Be careful to put tflin on a vector n */
3076:
3077: /* j is used from 0 to n-1 and can be -1 for parabolic search */
3078:
3079: /* if (j != -1) { /\* linear search *\/ */
3080: if (j > 0) { /* linear search */
3081: /* for (i=0; i<n; i++){ */
3082: for (i=1; i<=n; i++){
3083: tflin[i] = x[i] + l *v[i][j];
3084: #ifdef DEBUGPRAX
3085: /* printf(" flin i=%14d t=%14.7f x=%14.7f l=%14.7f v[%d,%d]=%14.7f nf=%14d\n",i+1, tflin[i],x[i],l,i,j,v[i][j],nf); */
3086: printf(" flin i=%14d t=%14.7f x=%14.7f l=%14.7f v[%d,%d]=%14.7f nf=%14d\n",i, tflin[i],x[i],l,i,j,v[i][j],nf);
3087: #endif
3088: }
3089: }
3090: else { /* search along parabolic space curve */
3091: qa = l*(l-qd1)/(qd0*(qd0+qd1));
3092: qb = (l+qd0)*(qd1-l)/(qd0*qd1);
3093: qc = l*(l+qd0)/(qd1*(qd0+qd1));
3094: #ifdef DEBUGPRAX
3095: printf(" search along a parabolic space curve. j=%14d nf=%14d l=%14.7f qd0=%14.7f qd1=%14.7f\n",j,nf,l,qd0,qd1);
3096: #endif
3097: /* for (i=0; i<n; i++){ */
3098: for (i=1; i<=n; i++){
3099: tflin[i] = qa*q0[i]+qb*x[i]+qc*q1[i];
3100: #ifdef DEBUGPRAX
3101: /* printf(" parabole i=%14d t(i)=%14.7f q0=%14.7f x=%14.7f q1=%14.7f\n",i+1,tflin[i],q0[i],x[i],q1[i]); */
3102: printf(" parabole i=%14d t(i)=%14.7e q0=%14.7e x=%14.7e q1=%14.7e\n",i,tflin[i],q0[i],x[i],q1[i]);
3103: #endif
3104: }
3105: }
3106: nf++;
3107:
3108: #ifdef NR_SHIFT
3109: return (*fun)((tflin-1), n);
3110: #else
3111: /* return (*fun)(tflin, n);*/
3112: return (*fun)(tflin);
3113: #endif
3114: }
3115:
3116: void minny(int j, int nits, double *d2, double *x1, double f1, int fk)
3117: /* double *d2, *x1, f1; */
3118: {
3119: /* here j is from 0 to n-1 and can be -1 for parabolic search */
3120: /* MINIMIZES F FROM X IN THE DIRECTION V(*,J) */
3121: /* UNLESS J<1, WHEN A QUADRATIC SEARCH IS DONE */
3122: /* IN THE PLANE DEFINED BY Q0, Q1 AND X. */
3123: /* D2 AN APPROXIMATION TO HALF F'' (OR ZERO), */
3124: /* X1 AN ESTIMATE OF DISTANCE TO MINIMUM, */
3125: /* RETURNED AS THE DISTANCE FOUND. */
3126: /* IF FK = TRUE THEN F1 IS FLIN(X1), OTHERWISE */
3127: /* X1 AND F1 ARE IGNORED ON ENTRY UNLESS FINAL */
3128: /* FX > F1. NITS CONTROLS THE NUMBER OF TIMES */
3129: /* AN ATTEMPT IS MADE TO HALVE THE INTERVAL. */
3130: /* SIDE EFFECTS: USES AND ALTERS X, FX, NF, NL. */
3131: /* IF J < 1 USES VARIABLES Q... . */
3132: /* USES H, N, T, M2, M4, LDT, DMIN, MACHEPS; */
3133: int k, i, dz;
3134: double x2, xm, f0, f2, fm, d1, t2, sf1, sx1;
3135: double s;
3136: double macheps;
3137: macheps=pow(16.0,-13.0);
3138: sf1 = f1; sx1 = *x1;
3139: k = 0; xm = 0.0; fm = f0 = fx; dz = *d2 < macheps;
3140: /* h=1.0;*/ /* To be revised */
3141: #ifdef DEBUGPRAX
3142: /* printf("min macheps=%14g h=%14g step=%14g t=%14g fx=%14g\n",macheps,h, step,t, fx); */
3143: /* Where is fx coming from */
3144: printf(" min macheps=%14g h=%14g t=%14g fx=%.9lf dirj=%d\n",macheps, h, t, fx, j);
3145: matprint(" min vectors:",v,n,n);
3146: #endif
3147: /* find step size */
3148: s = 0.;
3149: /* for (i=0; i<n; i++) s += x[i]*x[i]; */
3150: for (i=1; i<=n; i++) s += x[i]*x[i];
3151: s = sqrt(s);
3152: if (dz)
3153: t2 = m4*sqrt(fabs(fx)/dmin + s*ldt) + m2*ldt;
3154: else
3155: t2 = m4*sqrt(fabs(fx)/(*d2) + s*ldt) + m2*ldt;
3156: s = s*m4 + t;
3157: if (dz && t2 > s) t2 = s;
3158: if (t2 < small_windows) t2 = small_windows;
3159: if (t2 > 0.01*h) t2 = 0.01 * h;
3160: if (fk && f1 <= fm) {
3161: xm = *x1;
3162: fm = f1;
3163: }
3164: #ifdef DEBUGPRAX
3165: printf(" additional flin X1=%14.7f t2=%14.7f *f1=%14.7f fm=%14.7f fk=%d\n",*x1,t2,f1,fm,fk);
3166: #endif
3167: if (!fk || fabs(*x1) < t2) {
3168: *x1 = (*x1 >= 0 ? t2 : -t2);
3169: /* *x1 = (*x1 > 0 ? t2 : -t2); */ /* kind of error */
3170: #ifdef DEBUGPRAX
3171: printf(" additional flin X1=%16.10e dirj=%d fk=%d\n",*x1, j, fk);
3172: #endif
3173: f1 = flin(*x1, j);
3174: #ifdef DEBUGPRAX
3175: printf(" after flin f1=%18.12e dirj=%d fk=%d\n",f1, j,fk);
3176: #endif
3177: }
3178: if (f1 <= fm) {
3179: xm = *x1;
3180: fm = f1;
3181: }
3182: L0: /*L0 loop or next */
3183: /*
3184: Evaluate FLIN at another point and estimate the second derivative.
3185: */
3186: if (dz) {
3187: x2 = (f0 < f1 ? -(*x1) : 2*(*x1));
3188: #ifdef DEBUGPRAX
3189: printf(" additional second flin x2=%14.8e x1=%14.8e f0=%14.8e f1=%18.12e dirj=%d\n",x2,*x1,f0,f1,j);
3190: #endif
3191: f2 = flin(x2, j);
3192: #ifdef DEBUGPRAX
3193: printf(" additional second flin x2=%16.10e x1=%16.10e f1=%18.12e f0=%18.10e f2=%18.10e fm=%18.10e\n",x2, *x1, f1,f0,f2,fm);
3194: #endif
3195: if (f2 <= fm) {
3196: xm = x2;
3197: fm = f2;
3198: }
3199: /* d2 is the curvature or double difference f1 doesn't seem to be accurately computed */
3200: *d2 = (x2*(f1-f0) - (*x1)*(f2-f0))/((*x1)*x2*((*x1)-x2));
3201: #ifdef DEBUGPRAX
3202: double d11,d12;
3203: d11=(f1-f0)/(*x1);d12=(f2-f0)/x2;
3204: printf(" d11=%18.12e d12=%18.12e d11-d12=%18.12e x1-x2=%18.12e (d11-d12)/(x2-(*x1))=%18.12e\n", d11 ,d12, d11-d12, x2-(*x1), (d11-d12)/(x2-(*x1)));
3205: printf(" original computing f1=%18.12e *d2=%16.10e f0=%18.12e f1-f0=%16.10e f2-f0=%16.10e\n",f1,*d2,f0,f1-f0, f2-f0);
3206: double ff1=7.783920622852e+04;
3207: double f1mf0=9.0344736236e-05;
3208: *d2 = (f1mf0)/ (*x1)/((*x1)-x2) - (f2-f0)/x2/((*x1)-x2);
3209: /* *d2 = (ff1-f0)/ (*x1)/((*x1)-x2) - (f2-f0)/x2/((*x1)-x2); */
3210: printf(" simpliff computing *d2=%16.10e f1mf0=%18.12e,f1=f0+f1mf0=%18.12e\n",*d2,f1mf0,f0+f1mf0);
3211: *d2 = ((f1-f0)/ (*x1) - (f2-f0)/x2)/((*x1)-x2);
3212: printf(" overlifi computing *d2=%16.10e\n",*d2);
3213: #endif
3214: *d2 = ((f1-f0)/ (*x1) - (f2-f0)/x2)/((*x1)-x2);
3215: }
3216: #ifdef DEBUGPRAX
3217: printf(" additional second flin xm=%14.8e fm=%14.8e *d2=%14.8e\n",xm, fm,*d2);
3218: #endif
3219: /*
3220: Estimate the first derivative at 0.
3221: */
3222: d1 = (f1-f0)/(*x1) - *x1**d2; dz = 1;
3223: /*
3224: Predict the minimum.
3225: */
3226: if (*d2 <= small_windows) {
3227: x2 = (d1 < 0 ? h : -h);
3228: }
3229: else {
3230: x2 = - 0.5*d1/(*d2);
3231: }
3232: #ifdef DEBUGPRAX
3233: printf(" AT d1=%14.8e d2=%14.8e small=%14.8e dz=%d x1=%14.8e x2=%14.8e\n",d1,*d2,small_windows,dz,*x1,x2);
3234: #endif
3235: if (fabs(x2) > h)
3236: x2 = (x2 > 0 ? h : -h);
3237: L1: /* L1 or try loop */
3238: #ifdef DEBUGPRAX
3239: printf(" AT predicted minimum flin x2=%14.8e x1=%14.8e K=%14d NITS=%14d dirj=%d\n",x2,*x1,k,nits,j);
3240: #endif
3241: f2 = flin(x2, j); /* x[i]+x2*v[i][j] */
3242: #ifdef DEBUGPRAX
3243: printf(" after flin f0=%14.8e f1=%14.8e f2=%14.8e fm=%14.8e\n",f0,f1,f2, fm);
3244: #endif
3245: if ((k < nits) && (f2 > f0)) {
3246: #ifdef DEBUGPRAX
3247: printf(" NO SUCCESS SO TRY AGAIN;\n");
3248: #endif
3249: k++;
3250: if ((f0 < f1) && (*x1*x2 > 0.0))
3251: goto L0; /* or next */
3252: x2 *= 0.5;
3253: goto L1;
3254: }
3255: nl++;
3256: #ifdef DEBUGPRAX
3257: printf(" bebeBE end of min x1=%14.8e x2=%14.8e f1=%14.8e f2=%14.8e f0=%14.8e fm=%14.8e d2=%14.8e\n",*x1, x2, f1, f2, f0, fm, *d2);
3258: #endif
3259: if (f2 > fm) x2 = xm; else fm = f2;
3260: if (fabs(x2*(x2-*x1)) > small_windows) {
3261: *d2 = (x2*(f1-f0) - *x1*(fm-f0))/(*x1*x2*(*x1-x2));
3262: }
3263: else {
3264: if (k > 0) *d2 = 0;
3265: }
3266: #ifdef DEBUGPRAX
1.362 brouard 3267: printf(" bebe end of min x1 might be very wrong x1=%14.8e fx=%14.8e d2=%14.8e\n",*x1, fx, *d2);
1.359 brouard 3268: #endif
3269: if (*d2 <= small_windows) *d2 = small_windows;
3270: *x1 = x2; fx = fm;
3271: if (sf1 < fx) {
3272: fx = sf1;
3273: *x1 = sx1;
3274: }
3275: /*
3276: Update X for linear search.
3277: */
3278: #ifdef DEBUGPRAX
3279: printf(" end of min x1=%14.8e fx=%14.8e d2=%14.8e\n",*x1, fx, *d2);
3280: #endif
3281:
3282: /* if (j != -1) */
3283: /* for (i=0; i<n; i++) */
3284: /* x[i] += (*x1)*v[i][j]; */
3285: if (j > 0)
3286: for (i=1; i<=n; i++)
3287: x[i] += (*x1)*v[i][j];
3288: }
3289:
3290: void quad() /* look for a minimum along the curve q0, q1, q2 */
3291: {
3292: int i;
3293: double l, s;
3294:
3295: s = fx; fx = qf1; qf1 = s; qd1 = 0.0;
3296: /* for (i=0; i<n; i++) { */
3297: for (i=1; i<=n; i++) {
3298: s = x[i]; l = q1[i]; x[i] = l; q1[i] = s;
3299: qd1 = qd1 + (s-l)*(s-l);
3300: }
3301: s = 0.0; qd1 = sqrt(qd1); l = qd1;
3302: #ifdef DEBUGPRAX
3303: printf(" QUAD after sqrt qd1=%14.8e \n",qd1);
3304: #endif
3305:
3306: if (qd0>0.0 && qd1>0.0 &&nl>=3*n*n) {
3307: #ifdef DEBUGPRAX
3308: printf(" QUAD before min value=%14.8e \n",qf1);
3309: #endif
3310: /* min(-1, 2, &s, &l, qf1, 1); */
3311: minny(0, 2, &s, &l, qf1, 1);
3312: qa = l*(l-qd1)/(qd0*(qd0+qd1));
3313: qb = (l+qd0)*(qd1-l)/(qd0*qd1);
3314: qc = l*(l+qd0)/(qd1*(qd0+qd1));
3315: }
3316: else {
3317: fx = qf1; qa = qb = 0.0; qc = 1.0;
3318: }
3319: #ifdef DEBUGPRAX
3320: printf("after eventual min qd0=%14.8e qd1=%14.8e nl=%d\n",qd0, qd1,nl);
3321: #endif
3322: qd0 = qd1;
3323: /* for (i=0; i<n; i++) { */
3324: for (i=1; i<=n; i++) {
3325: s = q0[i]; q0[i] = x[i];
3326: x[i] = qa*s + qb*x[i] + qc*q1[i];
3327: }
3328: #ifdef DEBUGQUAD
3329: vecprint ( " X after QUAD:" , x, n );
3330: #endif
3331: }
3332:
3333: /* void minfit(int n, double eps, double tol, double ab[N][N], double q[]) */
3334: void minfit(int n, double eps, double tol, double **ab, double q[])
3335: /* int n; */
3336: /* double eps, tol, ab[N][N], q[N]; */
3337: {
3338: int l, kt, l2, i, j, k;
3339: double c, f, g, h, s, x, y, z;
3340: /* double eps; */
3341: /* #ifndef MSDOS */
3342: /* double e[N]; /\* plenty of stack on a vax *\/ */
3343: /* #endif */
3344: /* double *e; */
3345: /* e=vector(0,n-1); /\* should be freed somewhere but gotos *\/ */
3346:
3347: /* householder's reduction to bidiagonal form */
3348:
3349: if(n==1){
3350: /* q[1-1]=ab[1-1][1-1]; */
3351: /* ab[1-1][1-1]=1.0; */
3352: q[1]=ab[1][1];
3353: ab[1][1]=1.0;
3354: return; /* added from hardt */
3355: }
3356: /* eps=macheps; */ /* added */
3357: x = g = 0.0;
3358: #ifdef DEBUGPRAX
3359: matprint (" HOUSE holder:", ab, n, n);
3360: #endif
3361:
3362: /* for (i=0; i<n; i++) { /\* FOR I := 1 UNTIL N DO *\/ */
3363: for (i=1; i<=n; i++) { /* FOR I := 1 UNTIL N DO */
3364: e[i] = g; s = 0.0; l = i+1;
3365: /* for (j=i; j<n; j++) /\* FOR J := I UNTIL N DO S := S*AB(J,I)**2; *\/ /\* not correct *\/ */
3366: for (j=i; j<=n; j++) /* FOR J := I UNTIL N DO S := S*AB(J,I)**2; */ /* not correct */
3367: s += ab[j][i] * ab[j][i];
3368: #ifdef DEBUGPRAXFIN
3369: printf("i=%d s=%d %.7g tol=%.7g",i,s,tol);
3370: #endif
3371: if (s < tol) {
3372: g = 0.0;
3373: }
3374: else {
3375: /* f = ab[i][i]; */
3376: f = ab[i][i];
3377: if (f < 0.0)
3378: g = sqrt(s);
3379: else
3380: g = -sqrt(s);
3381: /* h = f*g - s; ab[i][i] = f - g; */
3382: h = f*g - s; ab[i][i] = f - g;
3383: /* for (j=l; j<n; j++) { */ /* FOR J := L UNTIL N DO */ /* wrong */
3384: for (j=l; j<=n; j++) {
3385: f = 0.0;
3386: /* for (k=i; k<n; k++) /\* FOR K := I UNTIL N DO *\/ /\* wrong *\/ */
3387: for (k=i; k<=n; k++) /* FOR K := I UNTIL N DO */
3388: /* f += ab[k][i] * ab[k][j]; */
3389: f += ab[k][i] * ab[k][j];
3390: f /= h;
3391: for (k=i; k<=n; k++) /* FOR K := I UNTIL N DO */
3392: /* for (k=i; k<n; k++)/\* FOR K := I UNTIL N DO *\/ /\* wrong *\/ */
3393: ab[k][j] += f * ab[k][i];
3394: /* ab[k][j] += f * ab[k][i]; */
3395: #ifdef DEBUGPRAX
3396: printf("Holder J=%d F=%.7g",j,f);
3397: #endif
3398: }
3399: } /* end s */
3400: /* q[i] = g; s = 0.0; */
3401: q[i] = g; s = 0.0;
3402: #ifdef DEBUGPRAX
3403: printf(" I Q=%d %.7g",i,q[i]);
3404: #endif
3405:
3406: /* if (i < n) */
3407: /* if (i <= n) /\* I is always lower or equal to n wasn't in golub reinsch*\/ */
3408: /* for (j=l; j<n; j++) */
3409: for (j=l; j<=n; j++)
3410: s += ab[i][j] * ab[i][j];
3411: /* s += ab[i][j] * ab[i][j]; */
3412: if (s < tol) {
3413: g = 0.0;
3414: }
3415: else {
3416: if(i<n)
3417: /* f = ab[i][i+1]; */ /* Brent golub overflow */
3418: f = ab[i][i+1];
3419: if (f < 0.0)
3420: g = sqrt(s);
3421: else
3422: g = - sqrt(s);
3423: h = f*g - s;
3424: /* h = f*g - s; ab[i][i+1] = f - g; */ /* Overflow for i=n Error in Golub too but not Burkardt*/
3425: /* for (j=l; j<n; j++) */
3426: /* e[j] = ab[i][j]/h; */
3427: if(i<n){
3428: ab[i][i+1] = f - g;
3429: for (j=l; j<=n; j++)
3430: e[j] = ab[i][j]/h;
3431: /* for (j=l; j<n; j++) { */
3432: for (j=l; j<=n; j++) {
3433: s = 0.0;
3434: /* for (k=l; k<n; k++) s += ab[j][k]*ab[i][k]; */
3435: for (k=l; k<=n; k++) s += ab[j][k]*ab[i][k];
3436: /* for (k=l; k<n; k++) ab[j][k] += s * e[k]; */
3437: for (k=l; k<=n; k++) ab[j][k] += s * e[k];
3438: } /* END J */
3439: } /* END i <n */
3440: } /* end s */
3441: /* y = fabs(q[i]) + fabs(e[i]); */
3442: y = fabs(q[i]) + fabs(e[i]);
3443: if (y > x) x = y;
3444: #ifdef DEBUGPRAX
3445: printf(" I Y=%d %.7g",i,y);
3446: #endif
3447: #ifdef DEBUGPRAX
3448: printf(" i=%d e(i) %.7g",i,e[i]);
3449: #endif
3450: } /* end i */
3451: /*
3452: Accumulation of right hand transformations */
3453: /* for (i=n-1; i >= 0; i--) { */ /* FOR I := N STEP -1 UNTIL 1 DO */
3454: /* We should avoid the overflow in Golub */
3455: /* ab[n-1][n-1] = 1.0; */
3456: /* g = e[n-1]; */
3457: ab[n][n] = 1.0;
3458: g = e[n];
3459: l = n;
3460:
3461: /* for (i=n; i >= 1; i--) { */
3462: for (i=n-1; i >= 1; i--) { /* n-1 loops, different from brent and golub*/
3463: if (g != 0.0) {
3464: /* h = ab[i-1][i]*g; */
3465: h = ab[i][i+1]*g;
3466: for (j=l; j<=n; j++) ab[j][i] = ab[i][j] / h;
3467: for (j=l; j<=n; j++) {
3468: /* h = ab[i][i+1]*g; */
3469: /* for (j=l; j<n; j++) ab[j][i] = ab[i][j] / h; */
3470: /* for (j=l; j<n; j++) { */
3471: s = 0.0;
3472: /* for (k=l; k<n; k++) s += ab[i][k] * ab[k][j]; */
3473: /* for (k=l; k<n; k++) ab[k][j] += s * ab[k][i]; */
3474: for (k=l; k<=n; k++) s += ab[i][k] * ab[k][j];
3475: for (k=l; k<=n; k++) ab[k][j] += s * ab[k][i];
3476: }/* END J */
3477: }/* END G */
3478: /* for (j=l; j<n; j++) */
3479: /* ab[i][j] = ab[j][i] = 0.0; */
3480: /* ab[i][i] = 1.0; g = e[i]; l = i; */
3481: for (j=l; j<=n; j++)
3482: ab[i][j] = ab[j][i] = 0.0;
3483: ab[i][i] = 1.0; g = e[i]; l = i;
3484: }/* END I */
3485: #ifdef DEBUGPRAX
3486: matprint (" HOUSE accumulation:",ab,n, n );
3487: #endif
3488:
3489: /* diagonalization to bidiagonal form */
3490: eps *= x;
3491: /* for (k=n-1; k>= 0; k--) { */
3492: for (k=n; k>= 1; k--) {
3493: kt = 0;
3494: TestFsplitting:
3495: #ifdef DEBUGPRAX
3496: printf(" TestFsplitting: k=%d kt=%d\n",k,kt);
3497: /* for(i=1;i<=n;i++)printf(" e(%d)=%.14f",i,e[i]);printf("\n"); */
3498: #endif
3499: kt = kt+1;
3500: /* TestFsplitting: */
3501: /* if (++kt > 30) { */
3502: if (kt > 30) {
3503: e[k] = 0.0;
3504: fprintf(stderr, "\n+++ MINFIT - Fatal error\n");
3505: fprintf ( stderr, " The QR algorithm failed to converge.\n" );
3506: }
3507: /* for (l2=k; l2>=0; l2--) { */
3508: for (l2=k; l2>=1; l2--) {
3509: l = l2;
3510: #ifdef DEBUGPRAX
3511: printf(" l e(l)< eps %d %.7g %.7g ",l,e[l], eps);
3512: #endif
3513: /* if (fabs(e[l]) <= eps) */
3514: if (fabs(e[l]) <= eps)
3515: goto TestFconvergence;
3516: /* if (fabs(q[l-1]) <= eps)*/ /* missing if ( 1 < l ){ *//* printf(" q(l-1)< eps %d %.7g %.7g ",l-1,q[l-2], eps); */
3517: if (fabs(q[l-1]) <= eps)
3518: break; /* goto Cancellation; */
3519: }
3520: Cancellation:
3521: #ifdef DEBUGPRAX
3522: printf(" Cancellation:\n");
3523: #endif
3524: c = 0.0; s = 1.0;
3525: for (i=l; i<=k; i++) {
3526: f = s * e[i]; e[i] *= c;
3527: /* f = s * e[i]; e[i] *= c; */
3528: if (fabs(f) <= eps)
3529: goto TestFconvergence;
3530: /* g = q[i]; */
3531: g = q[i];
3532: if (fabs(f) < fabs(g)) {
3533: double fg = f/g;
3534: h = fabs(g)*sqrt(1.0+fg*fg);
3535: }
3536: else {
3537: double gf = g/f;
3538: h = (f!=0.0 ? fabs(f)*sqrt(1.0+gf*gf) : 0.0);
3539: }
3540: /* COMMENT: THE ABOVE REPLACES Q(I):=H:=LONGSQRT(G*G+F*F) */
3541: /* WHICH MAY GIVE INCORRECT RESULTS IF THE */
3542: /* SQUARES UNDERFLOW OR IF F = G = 0; */
3543:
3544: /* q[i] = h; */
3545: q[i] = h;
3546: if (h == 0.0) { h = 1.0; g = 1.0; }
3547: c = g/h; s = -f/h;
3548: }
3549: TestFconvergence:
3550: #ifdef DEBUGPRAX
3551: printf(" TestFconvergence: l=%d k=%d\n",l,k);
3552: #endif
3553: /* z = q[k]; */
3554: z = q[k];
3555: if (l == k)
3556: goto Convergence;
3557: /* shift from bottom 2x2 minor */
3558: /* x = q[l]; y = q[k-l]; g = e[k-1]; h = e[k]; */ /* Error */
3559: x = q[l]; y = q[k-1]; g = e[k-1]; h = e[k];
3560: f = ((y-z)*(y+z) + (g-h)*(g+h)) / (2.0*h*y);
3561: g = sqrt(f*f+1.0);
3562: if (f <= 0.0)
3563: f = ((x-z)*(x+z) + h*(y/(f-g)-h))/x;
3564: else
3565: f = ((x-z)*(x+z) + h*(y/(f+g)-h))/x;
3566: /* next qr transformation */
3567: s = c = 1.0;
3568: for (i=l+1; i<=k; i++) {
3569: #ifdef DEBUGPRAXQR
3570: printf(" Before Mid TestFconvergence: l+1=%d i=%d k=%d h=%.6e e(i)=%14.8f e(i-1)=%14.8f\n",l+1,i,k, h, e[i],e[i-1]);
3571: #endif
3572: /* g = e[i]; y = q[i]; h = s*g; g *= c; */
3573: g = e[i]; y = q[i]; h = s*g; g *= c;
3574: if (fabs(f) < fabs(h)) {
3575: double fh = f/h;
3576: z = fabs(h) * sqrt(1.0 + fh*fh);
3577: }
3578: else {
3579: double hf = h/f;
3580: z = (f!=0.0 ? fabs(f)*sqrt(1.0+hf*hf) : 0.0);
3581: }
3582: /* e[i-1] = z; */
3583: e[i-1] = z;
3584: #ifdef DEBUGPRAXQR
3585: printf(" Mid TestFconvergence: l+1=%d i=%d k=%d h=%.6e e(i)=%14.8f e(i-1)=%14.8f\n",l+1,i,k, h, e[i],e[i-1]);
3586: #endif
3587: if (z == 0.0)
3588: f = z = 1.0;
3589: c = f/z; s = h/z;
3590: f = x*c + g*s; g = - x*s + g*c; h = y*s;
3591: y *= c;
3592: /* for (j=0; j<n; j++) { */
3593: /* x = ab[j][i-1]; z = ab[j][i]; */
3594: /* ab[j][i-1] = x*c + z*s; */
3595: /* ab[j][i] = - x*s + z*c; */
3596: /* } */
3597: for (j=1; j<=n; j++) {
3598: x = ab[j][i-1]; z = ab[j][i];
3599: ab[j][i-1] = x*c + z*s;
3600: ab[j][i] = - x*s + z*c;
3601: }
3602: if (fabs(f) < fabs(h)) {
3603: double fh = f/h;
3604: z = fabs(h) * sqrt(1.0 + fh*fh);
3605: }
3606: else {
3607: double hf = h/f;
3608: z = (f!=0.0 ? fabs(f)*sqrt(1.0+hf*hf) : 0.0);
3609: }
3610: #ifdef DEBUGPRAXQR
3611: printf(" qr transformation z f h=%.7g %.7g %.7g i=%d k=%d\n",z,f,h, i, k);
3612: #endif
3613: q[i-1] = z;
3614: if (z == 0.0)
3615: z = f = 1.0;
3616: c = f/z; s = h/z;
3617: f = c*g + s*y; /* f can be very small */
3618: x = - s*g + c*y;
3619: }
3620: /* e[l] = 0.0; e[k] = f; q[k] = x; */
3621: e[l] = 0.0; e[k] = f; q[k] = x;
3622: #ifdef DEBUGPRAXQR
3623: printf(" aftermid loop l=%d k=%d e(l)=%7g e(k)=%.7g q(k)=%.7g x=%.7g\n",l,k,e[l],e[k],q[k],x);
3624: #endif
3625: goto TestFsplitting;
3626: Convergence:
3627: #ifdef DEBUGPRAX
3628: printf(" Convergence:\n");
3629: #endif
3630: if (z < 0.0) {
3631: /* q[k] = - z; */
3632: /* for (j=0; j<n; j++) ab[j][k] = - ab[j][k]; */
3633: q[k] = - z;
3634: for (j=1; j<=n; j++) ab[j][k] = - ab[j][k];
3635: }/* END Z */
3636: }/* END K */
3637: } /* END MINFIT */
3638:
3639:
3640: double praxis(double tol, double macheps, double h0, int _n, int _prin, double *_x, double (*_fun)(double *_x))
3641: /* double praxis(double tol, double macheps, double h0, int _n, int _prin, double *_x, double (*_fun)(double *_x, int _n)) */
3642: /* double praxis(double (*_fun)(), double _x[], int _n) */
3643: /* double (*_fun)(); */
3644: /* double _x[N]; */
3645: /* double (*_fun)(); */
3646: /* double _x[N]; */
3647: {
3648: /* init global extern variables and parameters */
3649: /* double *d, *y, *z, */
3650: /* *q0, *q1, **v; */
3651: /* double *tflin; /\* used in flin: return (*fun)(tflin, n); *\/ */
3652: /* double *e; /\* used in minfit, don't konw how to free memory and thus made global *\/ */
3653:
3654:
3655: int seed; /* added */
3656: int biter=0;
3657: double r;
3658: double randbrent( int (*));
3659: double s, sf;
3660:
3661: h = h0; /* step; */
3662: t = tol;
3663: scbd = 1.0;
3664: illc = 0;
3665: ktm = 1;
3666:
3667: macheps = DBL_EPSILON;
3668: /* prin=4; */
3669: #ifdef DEBUGPRAX
3670: printf("Praxis macheps=%14g h=%14g step=%14g tol=%14g\n",macheps,h, h0,tol);
3671: #endif
3672: n = _n;
3673: x = _x;
3674: prin = _prin;
3675: fun = _fun;
3676: d=vector(1, n);
3677: y=vector(1, n);
3678: z=vector(1, n);
3679: q0=vector(1, n);
3680: q1=vector(1, n);
3681: e=vector(1, n);
3682: tflin=vector(1, n);
3683: v=matrix(1, n, 1, n);
3684: for(i=1;i<=n;i++){d[i]=y[i]=z[i]=q0[0]=e[i]=tflin[i]=0.;}
3685: small_windows = (macheps) * (macheps); vsmall = small_windows*small_windows;
3686: large = 1.0/small_windows; vlarge = 1.0/vsmall;
3687: m2 = sqrt(macheps); m4 = sqrt(m2);
3688: seed = 123456789; /* added */
3689: ldfac = (illc ? 0.1 : 0.01);
3690: for(i=1;i<=n;i++) z[i]=0.; /* Was missing in Gegenfurtner as well as Brent's algol or fortran */
3691: nl = kt = 0; nf = 1;
3692: #ifdef NR_SHIFT
3693: fx = (*fun)((x-1), n);
3694: #else
3695: fx = (*fun)(x);
3696: #endif
3697: qf1 = fx;
3698: t2 = small_windows + fabs(t); t = t2; dmin = small_windows;
3699: #ifdef DEBUGPRAX
3700: printf("praxis2 macheps=%14g h=%14g step=%14g small=%14g t=%14g\n",macheps,h, h0,small_windows, t);
3701: #endif
3702: if (h < 100.0*t) h = 100.0*t;
3703: #ifdef DEBUGPRAX
3704: printf("praxis3 macheps=%14g h=%14g step=%14g small=%14g t=%14g\n",macheps,h, h0,small_windows, t);
3705: #endif
3706: ldt = h;
3707: /* for (i=0; i<n; i++) for (j=0; j<n; j++) */
3708: for (i=1; i<=n; i++) for (j=1; j<=n; j++)
3709: v[i][j] = (i == j ? 1.0 : 0.0);
3710: d[1] = 0.0; qd0 = 0.0;
3711: /* for (i=0; i<n; i++) q1[i] = x[i]; */
3712: for (i=1; i<=n; i++) q1[i] = x[i];
3713: if (prin > 1) {
3714: printf("\n------------- enter function praxis -----------\n");
3715: printf("... current parameter settings ...\n");
3716: printf("... scaling ... %20.10e\n", scbd);
3717: printf("... tol ... %20.10e\n", t);
3718: printf("... maxstep ... %20.10e\n", h);
3719: printf("... illc ... %20u\n", illc);
3720: printf("... ktm ... %20u\n", ktm);
3721: printf("... maxfun ... %20u\n", maxfun);
3722: }
3723: if (prin) print2();
3724:
3725: mloop:
3726: biter++; /* Added to count the loops */
3727: /* sf = d[0]; */
3728: /* s = d[0] = 0.0; */
3729: printf("\n Big iteration %d \n",biter);
3730: fprintf(ficlog,"\n Big iteration %d \n",biter);
3731: sf = d[1];
3732: s = d[1] = 0.0;
3733:
3734: /* minimize along first direction V(*,1) */
3735: #ifdef DEBUGPRAX
3736: printf(" Minimize along the first direction V(*,1). illc=%d\n",illc);
3737: /* fprintf(ficlog," Minimize along the first direction V(*,1).\n"); */
3738: #endif
3739: #ifdef DEBUGPRAX2
3740: printf("praxis4 macheps=%14g h=%14g step=%14g small=%14g t=%14g\n",macheps,h, h0,small_windows, t);
3741: #endif
3742: /* min(0, 2, &d[0], &s, fx, 0); /\* mac heps not global *\/ */
1.362 brouard 3743: minny(1, 2, &d[1], &s, fx, 0); /* mac heps not global it seems that fx doesn't correspond to f(s=*x1) */
1.359 brouard 3744: #ifdef DEBUGPRAX
3745: printf("praxis5 macheps=%14g h=%14g looks at sign of s=%14g fx=%14g\n",macheps,h, s,fx);
3746: #endif
3747: if (s <= 0.0)
3748: /* for (i=0; i < n; i++) */
3749: for (i=1; i <= n; i++)
3750: v[i][1] = -v[i][1];
3751: /* if ((sf <= (0.9 * d[0])) || ((0.9 * sf) >= d[0])) */
3752: if ((sf <= (0.9 * d[1])) || ((0.9 * sf) >= d[1]))
3753: /* for (i=1; i<n; i++) */
3754: for (i=2; i<=n; i++)
3755: d[i] = 0.0;
3756: /* for (k=1; k<n; k++) { */
3757: for (k=2; k<=n; k++) {
3758: /*
3759: The inner loop starts here.
3760: */
3761: #ifdef DEBUGPRAX
3762: printf(" The inner loop here from k=%d to n=%d.\n",k,n);
3763: /* fprintf(ficlog," The inner loop here from k=%d to n=%d.\n",k,n); */
3764: #endif
3765: /* for (i=0; i<n; i++) */
3766: for (i=1; i<=n; i++)
3767: y[i] = x[i];
3768: sf = fx;
3769: #ifdef DEBUGPRAX
3770: printf(" illc=%d and kt=%d and ktm=%d\n", illc, kt, ktm);
3771: #endif
3772: illc = illc || (kt > 0);
3773: next:
3774: kl = k;
3775: df = 0.0;
3776: if (illc) { /* random step to get off resolution valley */
3777: #ifdef DEBUGPRAX
3778: printf(" A random step follows, to avoid resolution valleys.\n");
3779: matprint(" before rand, vectors:",v,n,n);
3780: #endif
3781: for (i=1; i<=n; i++) {
3782: #ifdef NOBRENTRAND
3783: r = drandom();
3784: #else
3785: seed=i;
3786: /* seed=i+1; */
3787: #ifdef DEBUGRAND
3788: printf(" Random seed=%d, brent i=%d",seed,i); /* YYYY i=5 j=1 vji= -0.0001170073 */
3789: #endif
3790: r = randbrent ( &seed );
3791: #endif
3792: #ifdef DEBUGRAND
3793: printf(" Random r=%.7g \n",r);
3794: #endif
3795: z[i] = (0.1 * ldt + t2 * pow(10.0,(double)kt)) * (r - 0.5);
3796: /* z[i] = (0.1 * ldt + t2 * pow(10.0,(double)kt)) * (drandom() - 0.5); */
3797:
3798: s = z[i];
3799: for (j=1; j <= n; j++)
3800: x[j] += s * v[j][i];
3801: }
3802: #ifdef DEBUGRAND
3803: matprint(" after rand, vectors:",v,n,n);
3804: #endif
3805: #ifdef NR_SHIFT
3806: fx = (*fun)((x-1), n);
3807: #else
1.366 brouard 3808: fx = (*fun)(x);
1.359 brouard 3809: #endif
3810: /* fx = (*func) ( (x-1) ); *//* This for func which is computed from x[1] and not from x[0] xm1=(x-1)*/
3811: nf++;
3812: }
3813: /* minimize along non-conjugate directions */
3814: #ifdef DEBUGPRAX
3815: printf(" Minimize along the 'non-conjugate' directions (dots printed) V(*,%d),...,V(*,%d).\n",k,n);
3816: /* fprintf(ficlog," Minimize along the 'non-conjugate' directions (dots printed) V(*,%d),...,V(*,%d).\n",k,n); */
3817: #endif
3818: /* for (k2=k; k2<n; k2++) { /\* Be careful here k2 <=n ? *\/ */
3819: for (k2=k; k2<=n; k2++) { /* Be careful here k2 <=n ? */
3820: sl = fx;
3821: s = 0.0;
3822: #ifdef DEBUGPRAX
3823: printf(" Minimize along the 'NON-CONJUGATE' true direction k2=%14d fx=%14.7f\n",k2, fx);
3824: matprint(" before min vectors:",v,n,n);
3825: #endif
3826: /* min(k2, 2, &d[k2], &s, fx, 0); */
3827: /* jsearch=k2-1; */
3828: /* min(jsearch, 2, &d[jsearch], &s, fx, 0); */
3829: minny(k2, 2, &d[k2], &s, fx, 0);
3830: #ifdef DEBUGPRAX
3831: printf(" . D(%d)=%14.7f d[k2]=%14.7f z[k2]=%14.7f illc=%14d fx=%14.7f\n",k2,d[k2],d[k2],z[k2],illc,fx);
3832: #endif
3833: if (illc) {
3834: /* double szk = s + z[k2]; */
3835: /* s = d[k2] * szk*szk; */
3836: double szk = s + z[k2];
3837: s = d[k2] * szk*szk;
3838: }
3839: else
3840: s = sl - fx;
3841: /* if (df < s) { */
3842: if (df <= s) {
3843: df = s;
3844: kl = k2;
3845: #ifdef DEBUGPRAX
3846: printf(" df=%.7g and choose kl=%d \n",df,kl); /* UUUU */
3847: #endif
3848: }
3849: } /* end loop k2 */
3850: /*
3851: If there was not much improvement on the first try, set
3852: ILLC = true and start the inner loop again.
3853: */
3854: #ifdef DEBUGPRAX
3855: printf(" If there was not much improvement on the first try, set ILLC = true and start the inner loop again. illc=%d\n",illc);
3856: /* fprintf(ficlog," If there was not much improvement on the first try, set ILLC = true and start the inner loop again.\n"); */
3857: #endif
3858: if (!illc && (df < fabs(100.0 * (macheps) * fx))) {
3859: #ifdef DEBUGPRAX
3860: printf("\n NO SUCCESS because DF is small, starts inner loop with same K(=%d), fabs( 100.0 * machep(=%.10e) * fx(=%.9e) )=%.9e > df(=%.9e) break illc=%d\n", k, macheps, fx, fabs ( 100.0 * macheps * fx ), df, illc);
3861: #endif
3862: illc = 1;
3863: goto next;
3864: }
3865: #ifdef DEBUGPRAX
3866: printf("\n SUCCESS, BREAKS inner loop K(=%d) because DF is big, fabs( 100.0 * machep(=%.10e) * fx(=%.9e) )=%.9e <= df(=%.9e) break illc=%d\n", k, macheps, fx, fabs ( 100.0 * macheps * fx ), df, illc);
3867: #endif
3868:
3869: /* if ((k == 1) && (prin > 1)){ /\* be careful k=2 *\/ */
3870: if ((k == 2) && (prin > 1)){ /* be careful k=2 */
3871: #ifdef DEBUGPRAX
3872: printf(" NEW D The second difference array d:\n" );
3873: /* fprintf(ficlog, " NEW D The second difference array d:\n" ); */
3874: #endif
3875: vecprint(" NEW D The second difference array d:",d,n);
3876: }
3877: /* minimize along conjugate directions */
3878: /*
3879: Minimize along the "conjugate" directions V(*,1),...,V(*,K-1).
3880: */
3881: #ifdef DEBUGPRAX
3882: printf("Minimize along the 'conjugate' directions V(*,1),...,V(*,K-1=%d).\n",k-1);
3883: /* fprintf(ficlog,"Minimize along the 'conjugate' directions V(*,1),...,V(*,K-1=%d).\n",k-1); */
3884: #endif
3885: /* for (k2=0; k2<=k-1; k2++) { */
3886: for (k2=1; k2<=k-1; k2++) {
3887: s = 0.0;
3888: /* min(k2-1, 2, &d[k2-1], &s, fx, 0); */
3889: minny(k2, 2, &d[k2], &s, fx, 0);
3890: }
3891: f1 = fx;
3892: fx = sf;
3893: lds = 0.0;
3894: /* for (i=0; i<n; i++) { */
3895: for (i=1; i<=n; i++) {
3896: sl = x[i];
3897: x[i] = y[i];
3898: y[i] = sl - y[i];
3899: sl = y[i];
3900: lds = lds + sl*sl;
3901: }
3902: lds = sqrt(lds);
3903: #ifdef DEBUGPRAX
3904: printf("Minimization done 'conjugate', shifted all points, computed lds=%.8f\n",lds);
3905: #endif
3906: /*
3907: Discard direction V(*,kl).
3908:
3909: If no random step was taken, V(*,KL) is the "non-conjugate"
3910: direction along which the greatest improvement was made.
3911: */
3912: if (lds > small_windows) {
3913: #ifdef DEBUGPRAX
3914: printf("lds big enough to throw direction V(*,kl=%d). If no random step was taken, V(*,KL) is the 'non-conjugate' direction along which the greatest improvement was made.\n",kl);
3915: matprint(" before shift new conjugate vectors:",v,n,n);
3916: #endif
3917: for (i=kl-1; i>=k; i--) {
3918: /* for (j=0; j < n; j++) */
3919: for (j=1; j <= n; j++)
3920: /* v[j][i+1] = v[j][i]; */ /* This is v[j][i+1]=v[j][i] i=kl-1 to k */
3921: v[j][i+1] = v[j][i]; /* This is v[j][i+1]=v[j][i] i=kl-1 to k */
3922: /* v[j][i+1] = v[j][i]; */
3923: /* d[i+1] = d[i];*/ /* last is d[k+1]= d[k] */
3924: d[i+1] = d[i]; /* last is d[k]= d[k-1] */
3925: }
3926: #ifdef DEBUGPRAX
3927: matprint(" after shift new conjugate vectors:",v,n,n);
3928: #endif /* d[k] = 0.0; */
3929: d[k] = 0.0;
3930: for (i=1; i <= n; i++)
3931: v[i][k] = y[i] / lds;
3932: /* v[i][k] = y[i] / lds; */
3933: #ifdef DEBUGPRAX
3934: printf("Minimize along the new 'conjugate' direction V(*,k=%d), which is the normalized vector: (new x) - (old x). d2=%14.7g lds=%.10f\n",k,d[k],lds);
3935: /* fprintf(ficlog,"Minimize along the new 'conjugate' direction V(*,k=%d), which is the normalized vector: (new x) - (old x).\n",k); */
3936: matprint(" before min new conjugate vectors:",v,n,n);
3937: #endif
3938: /* min(k-1, 4, &d[k-1], &lds, f1, 1); */
3939: minny(k, 4, &d[k], &lds, f1, 1);
3940: #ifdef DEBUGPRAX
3941: printf(" after min d(k)=%d %.7g lds=%14f\n",k,d[k],lds);
3942: matprint(" after min vectors:",v,n,n);
3943: #endif
3944: if (lds <= 0.0) {
3945: lds = -lds;
3946: #ifdef DEBUGPRAX
3947: printf(" lds changed sign lds=%.14f k=%d\n",lds,k);
3948: #endif
3949: /* for (i=0; i<n; i++) */
3950: /* v[i][k] = -v[i][k]; */
3951: for (i=1; i<=n; i++)
3952: v[i][k] = -v[i][k];
3953: }
3954: }
3955: ldt = ldfac * ldt;
3956: if (ldt < lds)
3957: ldt = lds;
3958: if (prin > 0){
3959: #ifdef DEBUGPRAX
3960: printf(" k=%d",k);
3961: /* fprintf(ficlog," k=%d",k); */
3962: #endif
3963: print2();/* n, x, prin, fx, nf, nl ); */
3964: }
3965: t2 = 0.0;
3966: /* for (i=0; i<n; i++) */
3967: for (i=1; i<=n; i++)
3968: t2 += x[i]*x[i];
3969: t2 = m2 * sqrt(t2) + t;
3970: /*
3971: See whether the length of the step taken since starting the
3972: inner loop exceeds half the tolerance.
3973: */
3974: #ifdef DEBUGPRAX
3975: printf("See if step length exceeds half the tolerance.\n"); /* ZZZZZ */
3976: /* fprintf(ficlog,"See if step length exceeds half the tolerance.\n"); */
3977: #endif
3978: if (ldt > (0.5 * t2))
3979: kt = 0;
3980: else
3981: kt++;
3982: #ifdef DEBUGPRAX
3983: printf("if kt=%d >? ktm=%d gotoL2 loop\n",kt,ktm);
3984: #endif
3985: if (kt > ktm){
3986: if ( 0 < prin ){
3987: /* printf("\nr8vec_print\n X:\n"); */
3988: /* fprintf(ficlog,"\nr8vec_print\n X:\n"); */
3989: vecprint ("END X:", x, n );
3990: }
3991: goto fret;
3992: }
3993: #ifdef DEBUGPRAX
3994: matprint(" end of L2 loop vectors:",v,n,n);
3995: #endif
3996:
3997: }
3998: /* printf("The inner loop ends here.\n"); */
3999: /* fprintf(ficlog,"The inner loop ends here.\n"); */
4000: /*
4001: The inner loop ends here.
4002:
4003: Try quadratic extrapolation in case we are in a curved valley.
4004: */
4005: #ifdef DEBUGPRAX
4006: printf("Try QUAD ratic extrapolation in case we are in a curved valley.\n");
4007: #endif
4008: /* try quadratic extrapolation in case */
4009: /* we are stuck in a curved valley */
4010: quad();
4011: dn = 0.0;
4012: /* for (i=0; i<n; i++) { */
4013: for (i=1; i<=n; i++) {
4014: d[i] = 1.0 / sqrt(d[i]);
4015: if (dn < d[i])
4016: dn = d[i];
4017: }
4018: if (prin > 2)
4019: matprint(" NEW DIRECTIONS vectors:",v,n,n);
4020: /* for (j=0; j<n; j++) { */
4021: for (j=1; j<=n; j++) {
4022: s = d[j] / dn;
4023: /* for (i=0; i < n; i++) */
4024: for (i=1; i <= n; i++)
4025: v[i][j] *= s;
4026: }
4027:
4028: if (scbd > 1.0) { /* scale axis to reduce condition number */
4029: #ifdef DEBUGPRAX
4030: printf("Scale the axes to try to reduce the condition number.\n");
4031: #endif
4032: /* fprintf(ficlog,"Scale the axes to try to reduce the condition number.\n"); */
4033: s = vlarge;
4034: /* for (i=0; i<n; i++) { */
4035: for (i=1; i<=n; i++) {
4036: sl = 0.0;
4037: /* for (j=0; j < n; j++) */
4038: for (j=1; j <= n; j++)
4039: sl += v[i][j]*v[i][j];
4040: z[i] = sqrt(sl);
4041: if (z[i] < m4)
4042: z[i] = m4;
4043: if (s > z[i])
4044: s = z[i];
4045: }
4046: /* for (i=0; i<n; i++) { */
4047: for (i=1; i<=n; i++) {
4048: sl = s / z[i];
4049: z[i] = 1.0 / sl;
4050: if (z[i] > scbd) {
4051: sl = 1.0 / scbd;
4052: z[i] = scbd;
4053: }
4054: }
4055: }
4056: for (i=1; i<=n; i++)
4057: /* for (j=0; j<=i-1; j++) { */
4058: /* for (j=1; j<=i; j++) { */
4059: for (j=1; j<=i-1; j++) {
4060: s = v[i][j];
4061: v[i][j] = v[j][i];
4062: v[j][i] = s;
4063: }
4064: #ifdef DEBUGPRAX
4065: printf(" Calculate a new set of orthogonal directions before repeating the main loop.\n Transpose V for MINFIT:...\n");
4066: #endif
4067: /*
4068: MINFIT finds the singular value decomposition of V.
4069:
4070: This gives the principal values and principal directions of the
4071: approximating quadratic form without squaring the condition number.
4072: */
4073: #ifdef DEBUGPRAX
4074: printf(" MINFIT finds the singular value decomposition of V. \n This gives the principal values and principal directions of the\n approximating quadratic form without squaring the condition number...\n");
4075: #endif
4076:
4077: minfit(n, macheps, vsmall, v, d);
4078: /* for(i=0; i<n;i++)printf(" %14.7g",d[i]); */
4079: /* v is overwritten with R. */
4080: /*
4081: Unscale the axes.
4082: */
4083: if (scbd > 1.0) {
4084: #ifdef DEBUGPRAX
4085: printf(" Unscale the axes.\n");
4086: #endif
4087: /* for (i=0; i<n; i++) { */
4088: for (i=1; i<=n; i++) {
4089: s = z[i];
4090: /* for (j=0; j<n; j++) */
4091: for (j=1; j<=n; j++)
4092: v[i][j] *= s;
4093: }
4094: /* for (i=0; i<n; i++) { */
4095: for (i=1; i<=n; i++) {
4096: s = 0.0;
4097: /* for (j=0; j<n; j++) */
4098: for (j=1; j<=n; j++)
4099: s += v[j][i]*v[j][i];
4100: s = sqrt(s);
4101: d[i] *= s;
4102: s = 1.0 / s;
4103: /* for (j=0; j<n; j++) */
4104: for (j=1; j<=n; j++)
4105: v[j][i] *= s;
4106: }
4107: }
4108: /* for (i=0; i<n; i++) { */
4109: double dni; /* added for compatibility with buckhardt but not brent */
4110: for (i=1; i<=n; i++) {
4111: dni=dn*d[i]; /* added for compatibility with buckhardt but not brent */
4112: if ((dn * d[i]) > large)
4113: d[i] = vsmall;
4114: else if ((dn * d[i]) < small_windows)
4115: d[i] = vlarge;
4116: else
4117: d[i] = 1.0 / dni / dni; /* added for compatibility with buckhardt but not brent */
4118: /* d[i] = pow(dn * d[i],-2.0); */
4119: }
4120: #ifdef DEBUGPRAX
4121: vecprint ("\n Before sort Eigenvalues of a:",d,n );
4122: #endif
4123:
4124: sort(); /* the new eigenvalues and eigenvectors */
4125: #ifdef DEBUGPRAX
4126: vecprint( " After sort the eigenvalues ....\n", d, n);
4127: matprint( " After sort the eigenvectors....\n", v, n,n);
4128: #endif
4129: #ifdef DEBUGPRAX
4130: printf(" Determine the smallest eigenvalue.\n");
4131: #endif
4132: /* dmin = d[n-1]; */
4133: dmin = d[n];
4134: if (dmin < small_windows)
4135: dmin = small_windows;
4136: /*
4137: The ratio of the smallest to largest eigenvalue determines whether
4138: the system is ill conditioned.
4139: */
4140:
4141: /* illc = (m2 * d[0]) > dmin; */
4142: illc = (m2 * d[1]) > dmin;
4143: #ifdef DEBUGPRAX
4144: printf(" The ratio of the smallest to largest eigenvalue determines whether\n the system is ill conditioned=%d . dmin=%.10lf < m2=%.10lf * d[1]=%.10lf \n",illc, dmin,m2, d[1]);
4145: #endif
4146:
4147: if ((prin > 2) && (scbd > 1.0))
4148: vecprint("\n The scale factors:",z,n);
4149: if (prin > 2)
4150: vecprint(" Principal values (EIGEN VALUES OF A) of the quadratic form:",d,n);
4151: if (prin > 2)
4152: matprint(" The principal axes (EIGEN VECTORS OF A:",v,n, n);
4153:
4154: if ((maxfun > 0) && (nf > maxfun)) {
4155: if (prin)
4156: printf("\n... maximum number of function calls reached ...\n");
4157: goto fret;
4158: }
4159: #ifdef DEBUGPRAX
4160: printf("Goto main loop\n");
4161: #endif
4162: goto mloop; /* back to main loop */
4163:
4164: fret:
4165: if (prin > 0) {
4166: vecprint("\n X:", x, n);
4167: /* printf("\n... ChiSq reduced to %20.10e ...\n", fx); */
4168: /* printf("... after %20u function calls.\n", nf); */
4169: }
4170: free_vector(d, 1, n);
4171: free_vector(y, 1, n);
4172: free_vector(z, 1, n);
4173: free_vector(q0, 1, n);
4174: free_vector(q1, 1, n);
4175: free_matrix(v, 1, n, 1, n);
4176: /* double *d, *y, *z, */
4177: /* *q0, *q1, **v; */
4178: free_vector(tflin, 1, n);
4179: /* double *tflin; /\* used in flin: return (*fun)(tflin, n); *\/ */
4180: free_vector(e, 1, n);
4181: /* double *e; /\* used in minfit, don't konw how to free memory and thus made global *\/ */
4182:
4183: return(fx);
4184: }
4185:
4186: /* end praxis gegen */
1.126 brouard 4187:
4188: /*************** powell ************************/
1.162 brouard 4189: /*
1.317 brouard 4190: Minimization of a function func of n variables. Input consists in an initial starting point
4191: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
4192: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
4193: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 4194: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
4195: function value at p , and iter is the number of iterations taken. The routine linmin is used.
4196: */
1.224 brouard 4197: #ifdef LINMINORIGINAL
4198: #else
4199: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 4200: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 4201: #endif
1.126 brouard 4202: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
4203: double (*func)(double []))
4204: {
1.224 brouard 4205: #ifdef LINMINORIGINAL
4206: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 4207: double (*func)(double []));
1.224 brouard 4208: #else
1.241 brouard 4209: void linmin(double p[], double xi[], int n, double *fret,
4210: double (*func)(double []),int *flat);
1.224 brouard 4211: #endif
1.239 brouard 4212: int i,ibig,j,jk,k;
1.126 brouard 4213: double del,t,*pt,*ptt,*xit;
1.181 brouard 4214: double directest;
1.126 brouard 4215: double fp,fptt;
4216: double *xits;
4217: int niterf, itmp;
1.349 brouard 4218: int Bigter=0, nBigterf=1;
4219:
1.126 brouard 4220: pt=vector(1,n);
4221: ptt=vector(1,n);
4222: xit=vector(1,n);
4223: xits=vector(1,n);
4224: *fret=(*func)(p);
4225: for (j=1;j<=n;j++) pt[j]=p[j];
1.338 brouard 4226: rcurr_time = time(NULL);
4227: fp=(*fret); /* Initialisation */
1.126 brouard 4228: for (*iter=1;;++(*iter)) {
4229: ibig=0;
4230: del=0.0;
1.157 brouard 4231: rlast_time=rcurr_time;
1.349 brouard 4232: rlast_btime=rcurr_time;
1.157 brouard 4233: /* (void) gettimeofday(&curr_time,&tzp); */
4234: rcurr_time = time(NULL);
4235: curr_time = *localtime(&rcurr_time);
1.337 brouard 4236: /* printf("\nPowell iter=%d -2*LL=%.12f gain=%.12f=%.3g %ld sec. %ld sec.",*iter,*fret, fp-*fret,fp-*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout); */
4237: /* fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f gain=%.12f=%.3g %ld sec. %ld sec.",*iter,*fret, fp-*fret,fp-*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog); */
1.359 brouard 4238: /* Bigter=(*iter - *iter % ncovmodel)/ncovmodel +1; /\* Big iteration, i.e on ncovmodel cycle *\/ */
4239: Bigter=(*iter - (*iter-1) % n)/n +1; /* Big iteration, i.e on ncovmodel cycle */
1.349 brouard 4240: printf("\nPowell iter=%d Big Iter=%d -2*LL=%.12f gain=%.3lg %ld sec. %ld sec.",*iter,Bigter,*fret,fp-*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
4241: fprintf(ficlog,"\nPowell iter=%d Big Iter=%d -2*LL=%.12f gain=%.3lg %ld sec. %ld sec.",*iter,Bigter,*fret,fp-*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
4242: fprintf(ficrespow,"%d %d %.12f %d",*iter,Bigter, *fret,curr_time.tm_sec-start_time.tm_sec);
1.324 brouard 4243: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 4244: for (i=1;i<=n;i++) {
1.126 brouard 4245: fprintf(ficrespow," %.12lf", p[i]);
4246: }
1.239 brouard 4247: fprintf(ficrespow,"\n");fflush(ficrespow);
4248: printf("\n#model= 1 + age ");
4249: fprintf(ficlog,"\n#model= 1 + age ");
4250: if(nagesqr==1){
1.241 brouard 4251: printf(" + age*age ");
4252: fprintf(ficlog," + age*age ");
1.239 brouard 4253: }
1.362 brouard 4254: for(j=1;j <=ncovmodel-2-nagesqr;j++){
1.239 brouard 4255: if(Typevar[j]==0) {
4256: printf(" + V%d ",Tvar[j]);
4257: fprintf(ficlog," + V%d ",Tvar[j]);
4258: }else if(Typevar[j]==1) {
4259: printf(" + V%d*age ",Tvar[j]);
4260: fprintf(ficlog," + V%d*age ",Tvar[j]);
4261: }else if(Typevar[j]==2) {
4262: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
4263: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.349 brouard 4264: }else if(Typevar[j]==3) {
4265: printf(" + V%d*V%d*age ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
4266: fprintf(ficlog," + V%d*V%d*age ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.239 brouard 4267: }
4268: }
1.126 brouard 4269: printf("\n");
1.239 brouard 4270: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
4271: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 4272: fprintf(ficlog,"\n");
1.239 brouard 4273: for(i=1,jk=1; i <=nlstate; i++){
4274: for(k=1; k <=(nlstate+ndeath); k++){
4275: if (k != i) {
4276: printf("%d%d ",i,k);
4277: fprintf(ficlog,"%d%d ",i,k);
4278: for(j=1; j <=ncovmodel; j++){
4279: printf("%12.7f ",p[jk]);
4280: fprintf(ficlog,"%12.7f ",p[jk]);
4281: jk++;
4282: }
4283: printf("\n");
4284: fprintf(ficlog,"\n");
4285: }
4286: }
4287: }
1.241 brouard 4288: if(*iter <=3 && *iter >1){
1.157 brouard 4289: tml = *localtime(&rcurr_time);
4290: strcpy(strcurr,asctime(&tml));
4291: rforecast_time=rcurr_time;
1.126 brouard 4292: itmp = strlen(strcurr);
4293: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 4294: strcurr[itmp-1]='\0';
1.162 brouard 4295: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 4296: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.349 brouard 4297: for(nBigterf=1;nBigterf<=31;nBigterf+=10){
4298: niterf=nBigterf*ncovmodel;
4299: /* rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time); */
1.241 brouard 4300: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
4301: forecast_time = *localtime(&rforecast_time);
4302: strcpy(strfor,asctime(&forecast_time));
4303: itmp = strlen(strfor);
4304: if(strfor[itmp-1]=='\n')
4305: strfor[itmp-1]='\0';
1.349 brouard 4306: printf(" - if your program needs %d BIG iterations (%d iterations) to converge, convergence will be \n reached in %s i.e.\n on %s (current time is %s);\n",nBigterf, niterf, asc_diff_time(rforecast_time-rcurr_time,tmpout),strfor,strcurr);
4307: fprintf(ficlog," - if your program needs %d BIG iterations (%d iterations) to converge, convergence will be \n reached in %s i.e.\n on %s (current time is %s);\n",nBigterf, niterf, asc_diff_time(rforecast_time-rcurr_time,tmpout),strfor,strcurr);
1.126 brouard 4308: }
4309: }
1.359 brouard 4310: for (i=1;i<=n;i++) { /* For each direction i, maximisation after loading directions */
4311: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales. xi is not changed but one dim xit */
4312:
4313: fptt=(*fret); /* Computes likelihood for parameters xit */
1.126 brouard 4314: #ifdef DEBUG
1.203 brouard 4315: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
4316: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 4317: #endif
1.203 brouard 4318: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 4319: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 4320: #ifdef LINMINORIGINAL
1.359 brouard 4321: linmin(p,xit,n,fret,func); /* New point i minimizing in direction xit, i has coordinates p[j].*/
1.357 brouard 4322: /* xit[j] gives the n coordinates of direction i as input.*/
4323: /* *fret gives the maximum value on direction xit */
1.224 brouard 4324: #else
4325: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.359 brouard 4326: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.224 brouard 4327: #endif
1.359 brouard 4328: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 4329: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.359 brouard 4330: /* because that direction will be replaced unless the gain del is small */
4331: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
4332: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
4333: /* with the new direction. */
4334: del=fabs(fptt-(*fret));
4335: ibig=i;
1.126 brouard 4336: }
4337: #ifdef DEBUG
4338: printf("%d %.12e",i,(*fret));
4339: fprintf(ficlog,"%d %.12e",i,(*fret));
4340: for (j=1;j<=n;j++) {
1.359 brouard 4341: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
4342: printf(" x(%d)=%.12e",j,xit[j]);
4343: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 4344: }
4345: for(j=1;j<=n;j++) {
1.359 brouard 4346: printf(" p(%d)=%.12e",j,p[j]);
4347: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 4348: }
4349: printf("\n");
4350: fprintf(ficlog,"\n");
4351: #endif
1.187 brouard 4352: } /* end loop on each direction i */
1.357 brouard 4353: /* Convergence test will use last linmin estimation (fret) and compare to former iteration (fp) */
1.188 brouard 4354: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.359 brouard 4355: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 4356: for(j=1;j<=n;j++) {
4357: if(flatdir[j] >0){
4358: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
4359: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 4360: }
1.319 brouard 4361: /* printf("\n"); */
4362: /* fprintf(ficlog,"\n"); */
4363: }
1.243 brouard 4364: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
4365: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 4366: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
4367: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
4368: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
4369: /* decreased of more than 3.84 */
4370: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
4371: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
4372: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 4373:
1.188 brouard 4374: /* Starting the program with initial values given by a former maximization will simply change */
4375: /* the scales of the directions and the directions, because the are reset to canonical directions */
4376: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
4377: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 4378: #ifdef DEBUG
4379: int k[2],l;
4380: k[0]=1;
4381: k[1]=-1;
4382: printf("Max: %.12e",(*func)(p));
4383: fprintf(ficlog,"Max: %.12e",(*func)(p));
4384: for (j=1;j<=n;j++) {
4385: printf(" %.12e",p[j]);
4386: fprintf(ficlog," %.12e",p[j]);
4387: }
4388: printf("\n");
4389: fprintf(ficlog,"\n");
4390: for(l=0;l<=1;l++) {
4391: for (j=1;j<=n;j++) {
4392: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
4393: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
4394: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
4395: }
4396: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
4397: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
4398: }
4399: #endif
4400:
4401: free_vector(xit,1,n);
4402: free_vector(xits,1,n);
4403: free_vector(ptt,1,n);
4404: free_vector(pt,1,n);
4405: return;
1.192 brouard 4406: } /* enough precision */
1.240 brouard 4407: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.359 brouard 4408: for (j=1;j<=n;j++) { /* Computes the extrapolated point and value f3, P_0 + 2 (P_n-P_0)=2Pn-P0 and xit is direction Pn-P0 */
1.126 brouard 4409: ptt[j]=2.0*p[j]-pt[j];
1.359 brouard 4410: xit[j]=p[j]-pt[j]; /* Coordinate j of last direction xi_n=P_n-P_0 */
4411: #ifdef DEBUG
4412: printf("\n %d xit=%12.7g p=%12.7g pt=%12.7g ",j,xit[j],p[j],pt[j]);
4413: #endif
4414: pt[j]=p[j]; /* New P0 is Pn */
4415: }
4416: #ifdef DEBUG
4417: printf("\n");
4418: #endif
1.181 brouard 4419: fptt=(*func)(ptt); /* f_3 */
1.359 brouard 4420: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in directions until some iterations are done */
1.224 brouard 4421: if (*iter <=4) {
1.225 brouard 4422: #else
4423: #endif
1.224 brouard 4424: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 4425: #else
1.161 brouard 4426: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 4427: #endif
1.162 brouard 4428: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 4429: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 4430: /* Let f"(x2) be the 2nd derivative equal everywhere. */
4431: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
4432: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 4433: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
4434: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
4435: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 4436: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 4437: /* Even if f3 <f1, directest can be negative and t >0 */
4438: /* mu² and del² are equal when f3=f1 */
1.359 brouard 4439: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
4440: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
4441: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
4442: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 4443: #ifdef NRCORIGINAL
4444: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
4445: #else
4446: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del); /* Intel compiler doesn't work on one line; bug reported */
1.161 brouard 4447: t= t- del*SQR(fp-fptt);
1.183 brouard 4448: #endif
1.202 brouard 4449: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 4450: #ifdef DEBUG
1.181 brouard 4451: printf("t1= %.12lf, t2= %.12lf, t=%.12lf directest=%.12lf\n", 2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del),del*SQR(fp-fptt),t,directest);
4452: fprintf(ficlog,"t1= %.12lf, t2= %.12lf, t=%.12lf directest=%.12lf\n", 2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del),del*SQR(fp-fptt),t,directest);
1.161 brouard 4453: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
4454: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
4455: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
4456: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
4457: printf("tt= %.12lf, t=%.12lf\n",2.0*(fp-2.0*(*fret)+fptt)*(fp-(*fret)-del)*(fp-(*fret)-del)-del*(fp-fptt)*(fp-fptt),t);
4458: fprintf(ficlog, "tt= %.12lf, t=%.12lf\n",2.0*(fp-2.0*(*fret)+fptt)*(fp-(*fret)-del)*(fp-(*fret)-del)-del*(fp-fptt)*(fp-fptt),t);
4459: #endif
1.183 brouard 4460: #ifdef POWELLORIGINAL
4461: if (t < 0.0) { /* Then we use it for new direction */
1.361 brouard 4462: #else /* Not POWELLOriginal but Brouard's */
1.182 brouard 4463: if (directest*t < 0.0) { /* Contradiction between both tests */
1.359 brouard 4464: printf("directest= %.12lf (if <0 we include P0 Pn as new direction), t= %.12lf, f1= %.12lf,f2= %.12lf,f3= %.12lf, del= %.12lf\n",directest, t, fp,(*fret),fptt,del);
1.192 brouard 4465: printf("f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
1.224 brouard 4466: fprintf(ficlog,"directest= %.12lf (if directest<0 or t<0 we include P0 Pn as new direction), t= %.12lf, f1= %.12lf,f2= %.12lf,f3= %.12lf, del= %.12lf\n",directest, t, fp,(*fret),fptt, del);
1.192 brouard 4467: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
4468: }
1.361 brouard 4469: if (directest < 0.0) { /* Then we use (P0, Pn) for new direction Xi_n or Xi_iBig */
1.181 brouard 4470: #endif
1.191 brouard 4471: #ifdef DEBUGLINMIN
1.234 brouard 4472: printf("Before linmin in direction P%d-P0\n",n);
4473: for (j=1;j<=n;j++) {
4474: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
4475: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
4476: if(j % ncovmodel == 0){
4477: printf("\n");
4478: fprintf(ficlog,"\n");
4479: }
4480: }
1.224 brouard 4481: #endif
4482: #ifdef LINMINORIGINAL
1.234 brouard 4483: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 4484: #else
1.234 brouard 4485: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
4486: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 4487: #endif
1.234 brouard 4488:
1.191 brouard 4489: #ifdef DEBUGLINMIN
1.234 brouard 4490: for (j=1;j<=n;j++) {
4491: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
4492: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
4493: if(j % ncovmodel == 0){
4494: printf("\n");
4495: fprintf(ficlog,"\n");
4496: }
4497: }
1.224 brouard 4498: #endif
1.234 brouard 4499: for (j=1;j<=n;j++) {
4500: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
4501: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
4502: }
1.361 brouard 4503:
4504: /* #else */
4505: /* for (i=1;i<=n-1;i++) { */
4506: /* for (j=1;j<=n;j++) { */
4507: /* xi[j][i]=xi[j][i+1]; /\* Standard method of conjugate directions, not Powell who changes the nth direction by p0 pn . *\/ */
4508: /* } */
4509: /* } */
4510: /* for (j=1;j<=n;j++) { */
4511: /* xi[j][n]=xit[j]; /\* and this nth direction by the by the average p_0 p_n *\/ */
4512: /* } */
4513: /* /\* for (j=1;j<=n-1;j++) { *\/ */
4514: /* /\* xi[j][1]=xi[j][j+1]; /\\* Standard method of conjugate directions *\\/ *\/ */
4515: /* /\* xi[j][n]=xit[j]; /\\* and this nth direction by the by the average p_0 p_n *\\/ *\/ */
4516: /* /\* } *\/ */
4517: /* #endif */
1.224 brouard 4518: #ifdef LINMINORIGINAL
4519: #else
1.234 brouard 4520: for (j=1, flatd=0;j<=n;j++) {
4521: if(flatdir[j]>0)
4522: flatd++;
4523: }
4524: if(flatd >0){
1.255 brouard 4525: printf("%d flat directions: ",flatd);
4526: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 4527: for (j=1;j<=n;j++) {
4528: if(flatdir[j]>0){
4529: printf("%d ",j);
4530: fprintf(ficlog,"%d ",j);
4531: }
4532: }
4533: printf("\n");
4534: fprintf(ficlog,"\n");
1.319 brouard 4535: #ifdef FLATSUP
4536: free_vector(xit,1,n);
4537: free_vector(xits,1,n);
4538: free_vector(ptt,1,n);
4539: free_vector(pt,1,n);
4540: return;
4541: #endif
1.361 brouard 4542: } /* endif(flatd >0) */
4543: #endif /* LINMINORIGINAL */
1.234 brouard 4544: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
4545: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
4546:
1.126 brouard 4547: #ifdef DEBUG
1.234 brouard 4548: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
4549: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
4550: for(j=1;j<=n;j++){
4551: printf(" %lf",xit[j]);
4552: fprintf(ficlog," %lf",xit[j]);
4553: }
4554: printf("\n");
4555: fprintf(ficlog,"\n");
1.126 brouard 4556: #endif
1.192 brouard 4557: } /* end of t or directest negative */
1.359 brouard 4558: printf(" Directest is positive, P_n-P_0 does not increase the conjugacy. n=%d\n",n);
4559: fprintf(ficlog," Directest is positive, P_n-P_0 does not increase the conjugacy. n=%d\n",n);
1.224 brouard 4560: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 4561: #else
1.234 brouard 4562: } /* end if (fptt < fp) */
1.192 brouard 4563: #endif
1.225 brouard 4564: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 4565: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 4566: #else
1.224 brouard 4567: #endif
1.234 brouard 4568: } /* loop iteration */
1.126 brouard 4569: }
1.234 brouard 4570:
1.126 brouard 4571: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 4572:
1.235 brouard 4573: double **prevalim(double **prlim, int nlstate, double x[], double age, double **oldm, double **savm, double ftolpl, int *ncvyear, int ij, int nres)
1.234 brouard 4574: {
1.338 brouard 4575: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij . Nicely done
1.279 brouard 4576: * (and selected quantitative values in nres)
4577: * by left multiplying the unit
4578: * matrix by transitions matrix until convergence is reached with precision ftolpl
4579: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
4580: * Wx is row vector: population in state 1, population in state 2, population dead
4581: * or prevalence in state 1, prevalence in state 2, 0
4582: * newm is the matrix after multiplications, its rows are identical at a factor.
4583: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
4584: * Output is prlim.
4585: * Initial matrix pimij
4586: */
1.206 brouard 4587: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
4588: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
4589: /* 0, 0 , 1} */
4590: /*
4591: * and after some iteration: */
4592: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
4593: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
4594: /* 0, 0 , 1} */
4595: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
4596: /* {0.51571254859325999, 0.4842874514067399, */
4597: /* 0.51326036147820708, 0.48673963852179264} */
4598: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 4599:
1.332 brouard 4600: int i, ii,j,k, k1;
1.209 brouard 4601: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.366 brouard 4602: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b); /* test */ /* for clang */
4603: /* double **matprod2(); */ /* test */
4604: /* double **out, cov[NCOVMAX+1], **pmij(); */ /* **pmmij is a global variable feeded with oldms etc */
4605: double **out, cov[NCOVMAX+1], **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 4606: double **newm;
1.209 brouard 4607: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 4608: int ncvloop=0;
1.288 brouard 4609: int first=0;
1.169 brouard 4610:
1.209 brouard 4611: min=vector(1,nlstate);
4612: max=vector(1,nlstate);
4613: meandiff=vector(1,nlstate);
4614:
1.218 brouard 4615: /* Starting with matrix unity */
1.126 brouard 4616: for (ii=1;ii<=nlstate+ndeath;ii++)
4617: for (j=1;j<=nlstate+ndeath;j++){
4618: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4619: }
1.169 brouard 4620:
4621: cov[1]=1.;
4622:
4623: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 4624: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 4625: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 4626: ncvloop++;
1.126 brouard 4627: newm=savm;
4628: /* Covariates have to be included here again */
1.138 brouard 4629: cov[2]=agefin;
1.319 brouard 4630: if(nagesqr==1){
4631: cov[3]= agefin*agefin;
4632: }
1.332 brouard 4633: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
4634: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
4635: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.349 brouard 4636: if(Typevar[k1]==1 || Typevar[k1]==3){ /* A product with age */
1.332 brouard 4637: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
4638: }else{
4639: cov[2+nagesqr+k1]=precov[nres][k1];
4640: }
4641: }/* End of loop on model equation */
4642:
4643: /* Start of old code (replaced by a loop on position in the model equation */
4644: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only of the model *\/ */
4645: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
4646: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; *\/ */
4647: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])]; */
4648: /* /\* model = 1 +age + V1*V3 + age*V1 + V2 + V1 + age*V2 + V3 + V3*age + V1*V2 */
4649: /* * k 1 2 3 4 5 6 7 8 */
4650: /* *cov[] 1 2 3 4 5 6 7 8 9 10 */
4651: /* *TypeVar[k] 2 1 0 0 1 0 1 2 */
4652: /* *Dummy[k] 0 2 0 0 2 0 2 0 */
4653: /* *Tvar[k] 4 1 2 1 2 3 3 5 */
4654: /* *nsd=3 (1) (2) (3) */
4655: /* *TvarsD[nsd] [1]=2 1 3 */
4656: /* *TnsdVar [2]=2 [1]=1 [3]=3 */
4657: /* *TvarsDind[nsd](=k) [1]=3 [2]=4 [3]=6 */
4658: /* *Tage[] [1]=1 [2]=2 [3]=3 */
4659: /* *Tvard[] [1][1]=1 [2][1]=1 */
4660: /* * [1][2]=3 [2][2]=2 */
4661: /* *Tprod[](=k) [1]=1 [2]=8 */
4662: /* *TvarsDp(=Tvar) [1]=1 [2]=2 [3]=3 [4]=5 */
4663: /* *TvarD (=k) [1]=1 [2]=3 [3]=4 [3]=6 [4]=6 */
4664: /* *TvarsDpType */
4665: /* *si model= 1 + age + V3 + V2*age + V2 + V3*age */
4666: /* * nsd=1 (1) (2) */
4667: /* *TvarsD[nsd] 3 2 */
4668: /* *TnsdVar (3)=1 (2)=2 */
4669: /* *TvarsDind[nsd](=k) [1]=1 [2]=3 */
4670: /* *Tage[] [1]=2 [2]= 3 */
4671: /* *\/ */
4672: /* /\* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; *\/ */
4673: /* /\* printf("prevalim Dummy combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov=%lf codtabm(%d,Tvar[%d])=%d \n",ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); *\/ */
4674: /* } */
4675: /* for (k=1; k<=nsq;k++) { /\* For single quantitative varying covariates only of the model *\/ */
4676: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
4677: /* /\* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline *\/ */
4678: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
4679: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][resultmodel[nres][k1]] */
4680: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
4681: /* /\* printf("prevalim Quantitative k=%d TvarsQind[%d]=%d, TvarsQ[%d]=V%d,Tqresult[%d][%d]=%f\n",k,k,TvarsQind[k],k,TvarsQ[k],nres,k,Tqresult[nres][k]); *\/ */
4682: /* } */
4683: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
4684: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
4685: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
4686: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
4687: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
4688: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
4689: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
4690: /* } */
4691: /* /\* printf("prevalim Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); *\/ */
4692: /* } */
4693: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
4694: /* /\* printf("prevalim Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2]); *\/ */
4695: /* if(Dummy[Tvard[k][1]]==0){ */
4696: /* if(Dummy[Tvard[k][2]]==0){ */
4697: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
4698: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
4699: /* }else{ */
4700: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
4701: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
4702: /* } */
4703: /* }else{ */
4704: /* if(Dummy[Tvard[k][2]]==0){ */
4705: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
4706: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
4707: /* }else{ */
4708: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
4709: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
4710: /* } */
4711: /* } */
4712: /* } /\* End product without age *\/ */
4713: /* ENd of old code */
1.138 brouard 4714: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
4715: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
4716: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 4717: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4718: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 4719: /* age and covariate values of ij are in 'cov' */
1.142 brouard 4720: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 4721:
1.126 brouard 4722: savm=oldm;
4723: oldm=newm;
1.209 brouard 4724:
4725: for(j=1; j<=nlstate; j++){
4726: max[j]=0.;
4727: min[j]=1.;
4728: }
4729: for(i=1;i<=nlstate;i++){
4730: sumnew=0;
4731: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
4732: for(j=1; j<=nlstate; j++){
4733: prlim[i][j]= newm[i][j]/(1-sumnew);
4734: max[j]=FMAX(max[j],prlim[i][j]);
4735: min[j]=FMIN(min[j],prlim[i][j]);
4736: }
4737: }
4738:
1.126 brouard 4739: maxmax=0.;
1.209 brouard 4740: for(j=1; j<=nlstate; j++){
4741: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
4742: maxmax=FMAX(maxmax,meandiff[j]);
4743: /* printf(" age= %d meandiff[%d]=%f, agefin=%d max[%d]=%f min[%d]=%f maxmax=%f\n", (int)age, j, meandiff[j],(int)agefin, j, max[j], j, min[j],maxmax); */
1.169 brouard 4744: } /* j loop */
1.203 brouard 4745: *ncvyear= (int)age- (int)agefin;
1.208 brouard 4746: /* printf("maxmax=%lf maxmin=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, maxmin, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.126 brouard 4747: if(maxmax < ftolpl){
1.209 brouard 4748: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
4749: free_vector(min,1,nlstate);
4750: free_vector(max,1,nlstate);
4751: free_vector(meandiff,1,nlstate);
1.126 brouard 4752: return prlim;
4753: }
1.288 brouard 4754: } /* agefin loop */
1.208 brouard 4755: /* After some age loop it doesn't converge */
1.288 brouard 4756: if(!first){
4757: first=1;
4758: printf("Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d). Others in log file only...\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
1.317 brouard 4759: fprintf(ficlog, "Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d).\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
4760: }else if (first >=1 && first <10){
4761: fprintf(ficlog, "Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d).\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
4762: first++;
4763: }else if (first ==10){
4764: fprintf(ficlog, "Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d).\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
4765: printf("Warning: the stable prevalence dit not converge. This warning came too often, IMaCh will stop notifying, even in its log file. Look at the graphs to appreciate the non convergence.\n");
4766: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
4767: first++;
1.288 brouard 4768: }
4769:
1.359 brouard 4770: /* Try to lower 'ftol', for example from 1.e-8 to 6.e-9.\n", ftolpl,
4771: * (int)age, (int)delaymax, (int)agefin, ncvloop,
4772: * (int)age-(int)agefin); */
1.209 brouard 4773: free_vector(min,1,nlstate);
4774: free_vector(max,1,nlstate);
4775: free_vector(meandiff,1,nlstate);
1.208 brouard 4776:
1.169 brouard 4777: return prlim; /* should not reach here */
1.126 brouard 4778: }
4779:
1.217 brouard 4780:
4781: /**** Back Prevalence limit (stable or period prevalence) ****************/
4782:
1.218 brouard 4783: /* double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ageminpar, double agemaxpar, double **oldm, double **savm, double **dnewm, double **doldm, double **dsavm, double ftolpl, int *ncvyear, int ij) */
4784: /* double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double **oldm, double **savm, double **dnewm, double **doldm, double **dsavm, double ftolpl, int *ncvyear, int ij) */
1.242 brouard 4785: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 4786: {
1.264 brouard 4787: /* Computes the prevalence limit in each live state at age x and for covariate combination ij (<=2**cptcoveff) by left multiplying the unit
1.217 brouard 4788: matrix by transitions matrix until convergence is reached with precision ftolpl */
4789: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
4790: /* Wx is row vector: population in state 1, population in state 2, population dead */
4791: /* or prevalence in state 1, prevalence in state 2, 0 */
4792: /* newm is the matrix after multiplications, its rows are identical at a factor */
4793: /* Initial matrix pimij */
4794: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
4795: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
4796: /* 0, 0 , 1} */
4797: /*
4798: * and after some iteration: */
4799: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
4800: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
4801: /* 0, 0 , 1} */
4802: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
4803: /* {0.51571254859325999, 0.4842874514067399, */
4804: /* 0.51326036147820708, 0.48673963852179264} */
4805: /* If we start from prlim again, prlim tends to a constant matrix */
4806:
1.359 brouard 4807: int i, ii,j, k1;
1.247 brouard 4808: int first=0;
1.217 brouard 4809: double *min, *max, *meandiff, maxmax,sumnew=0.;
4810: /* double **matprod2(); */ /* test */
1.366 brouard 4811: double **out, cov[NCOVMAX+1], **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij);
4812: /* double **out, cov[NCOVMAX+1], **bmij(); */ /* Deprecated in clang */
1.217 brouard 4813: double **newm;
1.218 brouard 4814: double **dnewm, **doldm, **dsavm; /* for use */
4815: double **oldm, **savm; /* for use */
4816:
1.217 brouard 4817: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
4818: int ncvloop=0;
4819:
4820: min=vector(1,nlstate);
4821: max=vector(1,nlstate);
4822: meandiff=vector(1,nlstate);
4823:
1.266 brouard 4824: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
4825: oldm=oldms; savm=savms;
4826:
4827: /* Starting with matrix unity */
4828: for (ii=1;ii<=nlstate+ndeath;ii++)
4829: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 4830: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4831: }
4832:
4833: cov[1]=1.;
4834:
4835: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
4836: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 4837: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 4838: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
4839: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 4840: ncvloop++;
1.218 brouard 4841: newm=savm; /* oldm should be kept from previous iteration or unity at start */
4842: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 4843: /* Covariates have to be included here again */
4844: cov[2]=agefin;
1.319 brouard 4845: if(nagesqr==1){
1.217 brouard 4846: cov[3]= agefin*agefin;;
1.319 brouard 4847: }
1.332 brouard 4848: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.349 brouard 4849: if(Typevar[k1]==1 || Typevar[k1]==3){ /* A product with age */
1.332 brouard 4850: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.242 brouard 4851: }else{
1.332 brouard 4852: cov[2+nagesqr+k1]=precov[nres][k1];
1.242 brouard 4853: }
1.332 brouard 4854: }/* End of loop on model equation */
4855:
4856: /* Old code */
4857:
4858: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
4859: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
4860: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; */
4861: /* /\* printf("bprevalim Dummy agefin=%.0f combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov[%d]=%lf codtabm(%d,Tvar[%d])=%d \n",agefin,ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],2+nagesqr+TvarsDind[k],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); *\/ */
4862: /* } */
4863: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
4864: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
4865: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
4866: /* /\* /\\* printf("prevalim ij=%d k=%d Tvar[%d]=%d nbcode=%d cov=%lf codtabm(%d,Tvar[%d])=%d \n",ij,k, k, Tvar[k],nbcode[Tvar[k]][codtabm(ij,Tvar[k])],cov[2+k], ij, k, codtabm(ij,Tvar[k])]); *\\/ *\/ */
4867: /* /\* } *\/ */
4868: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
4869: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
4870: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
4871: /* /\* printf("prevalim Quantitative k=%d TvarsQind[%d]=%d, TvarsQ[%d]=V%d,Tqresult[%d][%d]=%f\n",k,k,TvarsQind[k],k,TvarsQ[k],nres,k,Tqresult[nres][k]); *\/ */
4872: /* } */
4873: /* /\* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; *\/ */
4874: /* /\* for (k=1; k<=cptcovprod;k++) /\\* Useless *\\/ *\/ */
4875: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\\/ *\/ */
4876: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
4877: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
4878: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ ERROR ???*\/ */
4879: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
4880: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
4881: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
4882: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
4883: /* } */
4884: /* /\* printf("prevalim Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); *\/ */
4885: /* } */
4886: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
4887: /* /\* printf("prevalim Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2]); *\/ */
4888: /* if(Dummy[Tvard[k][1]]==0){ */
4889: /* if(Dummy[Tvard[k][2]]==0){ */
4890: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
4891: /* }else{ */
4892: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
4893: /* } */
4894: /* }else{ */
4895: /* if(Dummy[Tvard[k][2]]==0){ */
4896: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
4897: /* }else{ */
4898: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
4899: /* } */
4900: /* } */
4901: /* } */
1.217 brouard 4902:
4903: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
4904: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
4905: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
4906: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4907: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 4908: /* ij should be linked to the correct index of cov */
4909: /* age and covariate values ij are in 'cov', but we need to pass
4910: * ij for the observed prevalence at age and status and covariate
4911: * number: prevacurrent[(int)agefin][ii][ij]
4912: */
4913: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, ageminpar, agemaxpar, dnewm, doldm, dsavm,ij)); /\* Bug Valgrind *\/ */
4914: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij)); /\* Bug Valgrind *\/ */
4915: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij)); /* Bug Valgrind */
1.268 brouard 4916: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 4917: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
4918: /* for(i=1; i<=nlstate+ndeath; i++) { */
4919: /* printf("%d newm= ",i); */
4920: /* for(j=1;j<=nlstate+ndeath;j++) { */
4921: /* printf("%f ",newm[i][j]); */
4922: /* } */
4923: /* printf("oldm * "); */
4924: /* for(j=1;j<=nlstate+ndeath;j++) { */
4925: /* printf("%f ",oldm[i][j]); */
4926: /* } */
1.268 brouard 4927: /* printf(" bmmij "); */
1.266 brouard 4928: /* for(j=1;j<=nlstate+ndeath;j++) { */
4929: /* printf("%f ",pmmij[i][j]); */
4930: /* } */
4931: /* printf("\n"); */
4932: /* } */
4933: /* } */
1.217 brouard 4934: savm=oldm;
4935: oldm=newm;
1.266 brouard 4936:
1.217 brouard 4937: for(j=1; j<=nlstate; j++){
4938: max[j]=0.;
4939: min[j]=1.;
4940: }
4941: for(j=1; j<=nlstate; j++){
4942: for(i=1;i<=nlstate;i++){
1.234 brouard 4943: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
4944: bprlim[i][j]= newm[i][j];
4945: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
4946: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 4947: }
4948: }
1.218 brouard 4949:
1.217 brouard 4950: maxmax=0.;
4951: for(i=1; i<=nlstate; i++){
1.318 brouard 4952: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 4953: maxmax=FMAX(maxmax,meandiff[i]);
4954: /* printf("Back age= %d meandiff[%d]=%f, agefin=%d max[%d]=%f min[%d]=%f maxmax=%f\n", (int)age, i, meandiff[i],(int)agefin, i, max[i], i, min[i],maxmax); */
1.268 brouard 4955: } /* i loop */
1.217 brouard 4956: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 4957: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 4958: if(maxmax < ftolpl){
1.220 brouard 4959: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 4960: free_vector(min,1,nlstate);
4961: free_vector(max,1,nlstate);
4962: free_vector(meandiff,1,nlstate);
4963: return bprlim;
4964: }
1.288 brouard 4965: } /* agefin loop */
1.217 brouard 4966: /* After some age loop it doesn't converge */
1.288 brouard 4967: if(!first){
1.247 brouard 4968: first=1;
4969: printf("Warning: the back stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. Others in log file only...\n\
4970: Oldest age to start was %d-%d=%d, ncvloop=%d, ncvyear=%d\n", (int)age, maxmax, ftolpl, delaymax, (int)age, (int)delaymax, (int)agefin, ncvloop, *ncvyear);
4971: }
4972: fprintf(ficlog,"Warning: the back stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. \n\
1.217 brouard 4973: Oldest age to start was %d-%d=%d, ncvloop=%d, ncvyear=%d\n", (int)age, maxmax, ftolpl, delaymax, (int)age, (int)delaymax, (int)agefin, ncvloop, *ncvyear);
4974: /* Try to lower 'ftol', for example from 1.e-8 to 6.e-9.\n", ftolpl, (int)age, (int)delaymax, (int)agefin, ncvloop, (int)age-(int)agefin); */
4975: free_vector(min,1,nlstate);
4976: free_vector(max,1,nlstate);
4977: free_vector(meandiff,1,nlstate);
4978:
4979: return bprlim; /* should not reach here */
4980: }
4981:
1.126 brouard 4982: /*************** transition probabilities ***************/
4983:
4984: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
4985: {
1.138 brouard 4986: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 4987: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 4988: model to the ncovmodel covariates (including constant and age).
4989: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
4990: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
4991: ncth covariate in the global vector x is given by the formula:
4992: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
4993: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
4994: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
4995: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 4996: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 4997: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 4998: Sum on j ps[i][j] should equal to 1.
1.138 brouard 4999: */
5000: double s1, lnpijopii;
1.126 brouard 5001: /*double t34;*/
1.164 brouard 5002: int i,j, nc, ii, jj;
1.126 brouard 5003:
1.223 brouard 5004: for(i=1; i<= nlstate; i++){
5005: for(j=1; j<i;j++){
5006: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
5007: /*lnpijopii += param[i][j][nc]*cov[nc];*/
5008: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
5009: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
5010: }
5011: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 5012: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 5013: }
5014: for(j=i+1; j<=nlstate+ndeath;j++){
5015: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
5016: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
5017: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
5018: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
5019: }
5020: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 5021: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 5022: }
5023: }
1.218 brouard 5024:
1.223 brouard 5025: for(i=1; i<= nlstate; i++){
5026: s1=0;
5027: for(j=1; j<i; j++){
1.339 brouard 5028: /* printf("debug1 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 5029: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
5030: }
5031: for(j=i+1; j<=nlstate+ndeath; j++){
1.339 brouard 5032: /* printf("debug2 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 5033: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
5034: }
5035: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
5036: ps[i][i]=1./(s1+1.);
5037: /* Computing other pijs */
5038: for(j=1; j<i; j++)
1.325 brouard 5039: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 5040: for(j=i+1; j<=nlstate+ndeath; j++)
5041: ps[i][j]= exp(ps[i][j])*ps[i][i];
5042: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
5043: } /* end i */
1.218 brouard 5044:
1.223 brouard 5045: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
5046: for(jj=1; jj<= nlstate+ndeath; jj++){
5047: ps[ii][jj]=0;
5048: ps[ii][ii]=1;
5049: }
5050: }
1.294 brouard 5051:
5052:
1.223 brouard 5053: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
5054: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
5055: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
5056: /* } */
5057: /* printf("\n "); */
5058: /* } */
5059: /* printf("\n ");printf("%lf ",cov[2]);*/
5060: /*
5061: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 5062: goto end;*/
1.266 brouard 5063: return ps; /* Pointer is unchanged since its call */
1.126 brouard 5064: }
5065:
1.218 brouard 5066: /*************** backward transition probabilities ***************/
5067:
5068: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ageminpar, double agemaxpar, double ***dnewm, double **doldm, double **dsavm, int ij ) */
5069: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
5070: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
5071: {
1.302 brouard 5072: /* Computes the backward probability at age agefin, cov[2], and covariate combination 'ij'. In fact cov is already filled and x too.
1.266 brouard 5073: * Call to pmij(cov and x), call to cross prevalence, sums and inverses, left multiply, and returns in **ps as well as **bmij.
1.222 brouard 5074: */
1.359 brouard 5075: int ii, j;
1.222 brouard 5076:
1.366 brouard 5077: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate);
5078: /* double **pmij(); */ /* No more for clang */
1.222 brouard 5079: double sumnew=0.;
1.218 brouard 5080: double agefin;
1.292 brouard 5081: double k3=0.; /* constant of the w_x diagonal matrix (in order for B to sum to 1 even for death state) */
1.222 brouard 5082: double **dnewm, **dsavm, **doldm;
5083: double **bbmij;
5084:
1.218 brouard 5085: doldm=ddoldms; /* global pointers */
1.222 brouard 5086: dnewm=ddnewms;
5087: dsavm=ddsavms;
1.318 brouard 5088:
5089: /* Debug */
5090: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 5091: agefin=cov[2];
1.268 brouard 5092: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 5093: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 5094: the observed prevalence (with this covariate ij) at beginning of transition */
5095: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 5096:
5097: /* P_x */
1.325 brouard 5098: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 5099: /* outputs pmmij which is a stochastic matrix in row */
5100:
5101: /* Diag(w_x) */
1.292 brouard 5102: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 5103: sumnew=0.;
1.269 brouard 5104: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 5105: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 5106: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 5107: sumnew+=prevacurrent[(int)agefin][ii][ij];
5108: }
5109: if(sumnew >0.01){ /* At least some value in the prevalence */
5110: for (ii=1;ii<=nlstate+ndeath;ii++){
5111: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 5112: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 5113: }
5114: }else{
5115: for (ii=1;ii<=nlstate+ndeath;ii++){
5116: for (j=1;j<=nlstate+ndeath;j++)
5117: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
5118: }
5119: /* if(sumnew <0.9){ */
5120: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
5121: /* } */
5122: }
5123: k3=0.0; /* We put the last diagonal to 0 */
5124: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
5125: doldm[ii][ii]= k3;
5126: }
5127: /* End doldm, At the end doldm is diag[(w_i)] */
5128:
1.292 brouard 5129: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
5130: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 5131:
1.292 brouard 5132: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 5133: /* w1 p11 + w2 p21 only on live states N1./N..*N11/N1. + N2./N..*N21/N2.=(N11+N21)/N..=N.1/N.. */
1.222 brouard 5134: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 5135: sumnew=0.;
1.222 brouard 5136: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 5137: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 5138: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 5139: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 5140: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 5141: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 5142: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 5143: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 5144: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 5145: /* }else */
1.268 brouard 5146: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
5147: } /*End ii */
5148: } /* End j, At the end dsavm is diag[1/(w_1p1i+w_2 p2i)] for ALL states even if the sum is only for live states */
5149:
1.292 brouard 5150: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 5151: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 5152: /* end bmij */
1.266 brouard 5153: return ps; /*pointer is unchanged */
1.218 brouard 5154: }
1.217 brouard 5155: /*************** transition probabilities ***************/
5156:
1.218 brouard 5157: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 5158: {
5159: /* According to parameters values stored in x and the covariate's values stored in cov,
5160: computes the probability to be observed in state j being in state i by appying the
5161: model to the ncovmodel covariates (including constant and age).
5162: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
5163: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
5164: ncth covariate in the global vector x is given by the formula:
5165: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
5166: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
5167: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
5168: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
5169: Outputs ps[i][j] the probability to be observed in j being in j according to
5170: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
5171: */
5172: double s1, lnpijopii;
5173: /*double t34;*/
5174: int i,j, nc, ii, jj;
5175:
1.234 brouard 5176: for(i=1; i<= nlstate; i++){
5177: for(j=1; j<i;j++){
5178: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
5179: /*lnpijopii += param[i][j][nc]*cov[nc];*/
5180: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
5181: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
5182: }
5183: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
5184: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
5185: }
5186: for(j=i+1; j<=nlstate+ndeath;j++){
5187: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
5188: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
5189: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
5190: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
5191: }
5192: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
5193: }
5194: }
5195:
5196: for(i=1; i<= nlstate; i++){
5197: s1=0;
5198: for(j=1; j<i; j++){
5199: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
5200: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
5201: }
5202: for(j=i+1; j<=nlstate+ndeath; j++){
5203: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
5204: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
5205: }
5206: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
5207: ps[i][i]=1./(s1+1.);
5208: /* Computing other pijs */
5209: for(j=1; j<i; j++)
5210: ps[i][j]= exp(ps[i][j])*ps[i][i];
5211: for(j=i+1; j<=nlstate+ndeath; j++)
5212: ps[i][j]= exp(ps[i][j])*ps[i][i];
5213: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
5214: } /* end i */
5215:
5216: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
5217: for(jj=1; jj<= nlstate+ndeath; jj++){
5218: ps[ii][jj]=0;
5219: ps[ii][ii]=1;
5220: }
5221: }
1.296 brouard 5222: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 5223: for(jj=1; jj<= nlstate+ndeath; jj++){
5224: s1=0.;
5225: for(ii=1; ii<= nlstate+ndeath; ii++){
5226: s1+=ps[ii][jj];
5227: }
5228: for(ii=1; ii<= nlstate; ii++){
5229: ps[ii][jj]=ps[ii][jj]/s1;
5230: }
5231: }
5232: /* Transposition */
5233: for(jj=1; jj<= nlstate+ndeath; jj++){
5234: for(ii=jj; ii<= nlstate+ndeath; ii++){
5235: s1=ps[ii][jj];
5236: ps[ii][jj]=ps[jj][ii];
5237: ps[jj][ii]=s1;
5238: }
5239: }
5240: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
5241: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
5242: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
5243: /* } */
5244: /* printf("\n "); */
5245: /* } */
5246: /* printf("\n ");printf("%lf ",cov[2]);*/
5247: /*
5248: for(i=1; i<= npar; i++) printf("%f ",x[i]);
5249: goto end;*/
5250: return ps;
1.217 brouard 5251: }
5252:
5253:
1.126 brouard 5254: /**************** Product of 2 matrices ******************/
5255:
1.145 brouard 5256: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 5257: {
5258: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
5259: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
5260: /* in, b, out are matrice of pointers which should have been initialized
5261: before: only the contents of out is modified. The function returns
5262: a pointer to pointers identical to out */
1.145 brouard 5263: int i, j, k;
1.126 brouard 5264: for(i=nrl; i<= nrh; i++)
1.145 brouard 5265: for(k=ncolol; k<=ncoloh; k++){
5266: out[i][k]=0.;
5267: for(j=ncl; j<=nch; j++)
5268: out[i][k] +=in[i][j]*b[j][k];
5269: }
1.126 brouard 5270: return out;
5271: }
5272:
5273:
5274: /************* Higher Matrix Product ***************/
5275:
1.235 brouard 5276: double ***hpxij(double ***po, int nhstepm, double age, int hstepm, double *x, int nlstate, int stepm, double **oldm, double **savm, int ij, int nres )
1.126 brouard 5277: {
1.336 brouard 5278: /* Already optimized with precov.
5279: Computes the transition matrix starting at age 'age' and dummies values in each resultline (loop on ij to find the corresponding combination) to over
1.126 brouard 5280: 'nhstepm*hstepm*stepm' months (i.e. until
5281: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
5282: nhstepm*hstepm matrices.
5283: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
5284: (typically every 2 years instead of every month which is too big
5285: for the memory).
5286: Model is determined by parameters x and covariates have to be
5287: included manually here.
5288:
5289: */
5290:
1.359 brouard 5291: int i, j, d, h, k1;
1.131 brouard 5292: double **out, cov[NCOVMAX+1];
1.126 brouard 5293: double **newm;
1.187 brouard 5294: double agexact;
1.359 brouard 5295: /*double agebegin, ageend;*/
1.126 brouard 5296:
5297: /* Hstepm could be zero and should return the unit matrix */
5298: for (i=1;i<=nlstate+ndeath;i++)
5299: for (j=1;j<=nlstate+ndeath;j++){
5300: oldm[i][j]=(i==j ? 1.0 : 0.0);
5301: po[i][j][0]=(i==j ? 1.0 : 0.0);
5302: }
5303: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
5304: for(h=1; h <=nhstepm; h++){
5305: for(d=1; d <=hstepm; d++){
5306: newm=savm;
5307: /* Covariates have to be included here again */
5308: cov[1]=1.;
1.214 brouard 5309: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 5310: cov[2]=agexact;
1.319 brouard 5311: if(nagesqr==1){
1.227 brouard 5312: cov[3]= agexact*agexact;
1.319 brouard 5313: }
1.330 brouard 5314: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
5315: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
5316: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.349 brouard 5317: if(Typevar[k1]==1 || Typevar[k1]==3){ /* A product with age */
1.332 brouard 5318: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
5319: }else{
5320: cov[2+nagesqr+k1]=precov[nres][k1];
5321: }
5322: }/* End of loop on model equation */
5323: /* Old code */
5324: /* if( Dummy[k1]==0 && Typevar[k1]==0 ){ /\* Single dummy *\/ */
5325: /* /\* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) *\/ */
5326: /* /\* for (k=1; k<=nsd;k++) { /\\* For single dummy covariates only *\\/ *\/ */
5327: /* /\* /\\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\\/ *\/ */
5328: /* /\* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 *\/ */
5329: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
5330: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
5331: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
5332: /* /\* nsd 1 2 3 *\/ /\* Counting single dummies covar fixed or tv *\/ */
5333: /* /\*TvarsD[nsd] 4 3 1 *\/ /\* ID of single dummy cova fixed or timevary*\/ */
5334: /* /\*TvarsDind[k] 2 3 9 *\/ /\* position K of single dummy cova *\/ */
5335: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] *\/ */
5336: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
5337: /* /\* printf("hpxij Dummy combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov=%lf codtabm(%d,TnsdVar[TvarsD[%d])=%d \n",ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,TnsdVar[TvarsD[k]])); *\/ */
5338: /* printf("hpxij Dummy combi=%d k1=%d Tvar[%d]=V%d cov[2+%d+%d]=%lf resultmodel[nres][%d]=%d nres/nresult=%d/%d \n",ij,k1,k1, Tvar[k1],nagesqr,k1,cov[2+nagesqr+k1],k1,resultmodel[nres][k1],nres,nresult); */
5339: /* printf("hpxij new Dummy precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
5340: /* }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /\* Single quantitative variables *\/ */
5341: /* /\* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline *\/ */
5342: /* cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]]; */
5343: /* /\* for (k=1; k<=nsq;k++) { /\\* For single varying covariates only *\\/ *\/ */
5344: /* /\* /\\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\\/ *\/ */
5345: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
5346: /* printf("hPxij Quantitative k1=%d resultmodel[nres][%d]=%d,Tqresult[%d][%d]=%f\n",k1,k1,resultmodel[nres][k1],nres,resultmodel[nres][k1],Tqresult[nres][resultmodel[nres][k1]]); */
5347: /* printf("hpxij new Quanti precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
5348: /* }else if( Dummy[k1]==2 ){ /\* For dummy with age product *\/ */
5349: /* /\* Tvar[k1] Variable in the age product age*V1 is 1 *\/ */
5350: /* /\* [Tinvresult[nres][V1] is its value in the resultline nres *\/ */
5351: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvar[k1]]*cov[2]; */
5352: /* printf("DhPxij Dummy with age k1=%d Tvar[%d]=%d TinvDoQresult[nres=%d][%d]=%.f age=%.2f,cov[2+%d+%d]=%.3f\n",k1,k1,Tvar[k1],nres,TinvDoQresult[nres][Tvar[k1]],cov[2],nagesqr,k1,cov[2+nagesqr+k1]); */
5353: /* printf("hpxij new Dummy with age product precov[nres=%d][k1=%d]=%.4f * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
5354:
5355: /* /\* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; *\/ */
5356: /* /\* for (k=1; k<=cptcovage;k++){ /\\* For product with age V1+V1*age +V4 +age*V3 *\\/ *\/ */
5357: /* /\* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*\/ */
5358: /* /\* *\/ */
1.330 brouard 5359: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
5360: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
5361: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
1.332 brouard 5362: /* /\*cptcovage=2 1 2 *\/ */
5363: /* /\*Tage[k]= 5 8 *\/ */
5364: /* }else if( Dummy[k1]==3 ){ /\* For quant with age product *\/ */
5365: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
5366: /* printf("QhPxij Quant with age k1=%d resultmodel[nres][%d]=%d,Tqresult[%d][%d]=%f\n",k1,k1,resultmodel[nres][k1],nres,resultmodel[nres][k1],Tqresult[nres][resultmodel[nres][k1]]); */
5367: /* printf("hpxij new Quanti with age product precov[nres=%d][k1=%d] * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
5368: /* /\* if(Dummy[Tage[k]]== 2){ /\\* dummy with age *\\/ *\/ */
5369: /* /\* /\\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\\* dummy with age *\\\/ *\\/ *\/ */
5370: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\\/ *\/ */
5371: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\\/ *\/ */
5372: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
5373: /* /\* printf("hPxij Age combi=%d k=%d cptcovage=%d Tage[%d]=%d Tvar[Tage[%d]]=V%d nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[Tvar[Tage[k]]]])]=%d nres=%d\n",ij,k,cptcovage,k,Tage[k],k,Tvar[Tage[k]], nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[Tvar[Tage[k]]])],nres); *\/ */
5374: /* /\* } else if(Dummy[Tage[k]]== 3){ /\\* quantitative with age *\\/ *\/ */
5375: /* /\* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; *\/ */
5376: /* /\* } *\/ */
5377: /* /\* printf("hPxij Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); *\/ */
5378: /* }else if(Typevar[k1]==2 ){ /\* For product (not with age) *\/ */
5379: /* /\* for (k=1; k<=cptcovprod;k++){ /\\* For product without age *\\/ *\/ */
5380: /* /\* /\\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\\/ *\/ */
5381: /* /\* /\\* k 1 2 3 4 5 6 7 8 9 *\\/ *\/ */
5382: /* /\* /\\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\\/ *\/ */
5383: /* /\* /\\*cptcovprod=1 1 2 *\\/ *\/ */
5384: /* /\* /\\*Tprod[]= 4 7 *\\/ *\/ */
5385: /* /\* /\\*Tvard[][1] 4 1 *\\/ *\/ */
5386: /* /\* /\\*Tvard[][2] 3 2 *\\/ *\/ */
1.330 brouard 5387:
1.332 brouard 5388: /* /\* printf("hPxij Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]=%d nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]=%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2],nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])],nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]); *\/ */
5389: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
5390: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]]; */
5391: /* printf("hPxij Prod ij=%d k1=%d cov[2+%d+%d]=%.5f Tvard[%d][1]=V%d * Tvard[%d][2]=V%d ; TinvDoQresult[nres][Tvardk[k1][1]]=%.4f * TinvDoQresult[nres][Tvardk[k1][1]]=%.4f\n",ij,k1,nagesqr,k1,cov[2+nagesqr+k1],k1,Tvardk[k1][1], k1,Tvardk[k1][2], TinvDoQresult[nres][Tvardk[k1][1]], TinvDoQresult[nres][Tvardk[k1][2]]); */
5392: /* printf("hpxij new Product no age product precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
5393:
5394: /* /\* if(Dummy[Tvardk[k1][1]]==0){ *\/ */
5395: /* /\* if(Dummy[Tvardk[k1][2]]==0){ /\\* Product of dummies *\\/ *\/ */
5396: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
5397: /* /\* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; *\/ */
5398: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])]; *\/ */
5399: /* /\* }else{ /\\* Product of dummy by quantitative *\\/ *\/ */
5400: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; *\/ */
5401: /* /\* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; *\/ */
5402: /* /\* } *\/ */
5403: /* /\* }else{ /\\* Product of quantitative by...*\\/ *\/ */
5404: /* /\* if(Dummy[Tvard[k][2]]==0){ /\\* quant by dummy *\\/ *\/ */
5405: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\\/ *\/ */
5406: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
5407: /* /\* }else{ /\\* Product of two quant *\\/ *\/ */
5408: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\\/ *\/ */
5409: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
5410: /* /\* } *\/ */
5411: /* /\* }/\\*end of products quantitative *\\/ *\/ */
5412: /* }/\*end of products *\/ */
5413: /* } /\* End of loop on model equation *\/ */
1.235 brouard 5414: /* for (k=1; k<=cptcovn;k++) */
5415: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
5416: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
5417: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
5418: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
5419: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 5420:
5421:
1.126 brouard 5422: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
5423: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 5424: /* right multiplication of oldm by the current matrix */
1.126 brouard 5425: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
5426: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 5427: /* if((int)age == 70){ */
5428: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
5429: /* for(i=1; i<=nlstate+ndeath; i++) { */
5430: /* printf("%d pmmij ",i); */
5431: /* for(j=1;j<=nlstate+ndeath;j++) { */
5432: /* printf("%f ",pmmij[i][j]); */
5433: /* } */
5434: /* printf(" oldm "); */
5435: /* for(j=1;j<=nlstate+ndeath;j++) { */
5436: /* printf("%f ",oldm[i][j]); */
5437: /* } */
5438: /* printf("\n"); */
5439: /* } */
5440: /* } */
1.126 brouard 5441: savm=oldm;
5442: oldm=newm;
5443: }
5444: for(i=1; i<=nlstate+ndeath; i++)
5445: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 5446: po[i][j][h]=newm[i][j];
5447: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 5448: }
1.128 brouard 5449: /*printf("h=%d ",h);*/
1.126 brouard 5450: } /* end h */
1.267 brouard 5451: /* printf("\n H=%d \n",h); */
1.126 brouard 5452: return po;
5453: }
5454:
1.217 brouard 5455: /************* Higher Back Matrix Product ***************/
1.218 brouard 5456: /* double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, double **oldm, double **savm, double **dnewm, double **doldm, double **dsavm, int ij ) */
1.267 brouard 5457: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij, int nres )
1.217 brouard 5458: {
1.332 brouard 5459: /* For dummy covariates given in each resultline (for historical, computes the corresponding combination ij),
5460: computes the transition matrix starting at age 'age' over
1.217 brouard 5461: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 5462: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
5463: nhstepm*hstepm matrices.
5464: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
5465: (typically every 2 years instead of every month which is too big
1.217 brouard 5466: for the memory).
1.218 brouard 5467: Model is determined by parameters x and covariates have to be
1.266 brouard 5468: included manually here. Then we use a call to bmij(x and cov)
5469: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 5470: */
1.217 brouard 5471:
1.359 brouard 5472: int i, j, d, h, k1;
1.366 brouard 5473: double **out, cov[NCOVMAX+1], **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij);
5474: /* double **out, cov[NCOVMAX+1], **bmij(); */ /* No more for clang */
1.266 brouard 5475: double **newm, ***newmm;
1.217 brouard 5476: double agexact;
1.359 brouard 5477: /*double agebegin, ageend;*/
1.222 brouard 5478: double **oldm, **savm;
1.217 brouard 5479:
1.266 brouard 5480: newmm=po; /* To be saved */
5481: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 5482: /* Hstepm could be zero and should return the unit matrix */
5483: for (i=1;i<=nlstate+ndeath;i++)
5484: for (j=1;j<=nlstate+ndeath;j++){
5485: oldm[i][j]=(i==j ? 1.0 : 0.0);
5486: po[i][j][0]=(i==j ? 1.0 : 0.0);
5487: }
5488: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
5489: for(h=1; h <=nhstepm; h++){
5490: for(d=1; d <=hstepm; d++){
5491: newm=savm;
5492: /* Covariates have to be included here again */
5493: cov[1]=1.;
1.271 brouard 5494: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 5495: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 5496: /* Debug */
5497: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 5498: cov[2]=agexact;
1.332 brouard 5499: if(nagesqr==1){
1.222 brouard 5500: cov[3]= agexact*agexact;
1.332 brouard 5501: }
5502: /** New code */
5503: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.349 brouard 5504: if(Typevar[k1]==1 || Typevar[k1]==3){ /* A product with age */
1.332 brouard 5505: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.325 brouard 5506: }else{
1.332 brouard 5507: cov[2+nagesqr+k1]=precov[nres][k1];
1.325 brouard 5508: }
1.332 brouard 5509: }/* End of loop on model equation */
5510: /** End of new code */
5511: /** This was old code */
5512: /* for (k=1; k<=nsd;k++){ /\* For single dummy covariates only *\//\* cptcovn error *\/ */
5513: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
5514: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
5515: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/\* Bug valgrind *\/ */
5516: /* /\* printf("hbxij Dummy agexact=%.0f combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov[%d]=%lf codtabm(%d,Tvar[%d])=%d \n",agexact,ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],2+nagesqr+TvarsDind[k],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); *\/ */
5517: /* } */
5518: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
5519: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
5520: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
5521: /* /\* printf("hPxij Quantitative k=%d TvarsQind[%d]=%d, TvarsQ[%d]=V%d,Tqresult[%d][%d]=%f\n",k,k,TvarsQind[k],k,TvarsQ[k],nres,k,Tqresult[nres][k]); *\/ */
5522: /* } */
5523: /* for (k=1; k<=cptcovage;k++){ /\* Should start at cptcovn+1 *\//\* For product with age *\/ */
5524: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age error!!!*\\/ *\/ */
5525: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
5526: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
5527: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
5528: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
5529: /* } */
5530: /* /\* printf("hBxij Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); *\/ */
5531: /* } */
5532: /* for (k=1; k<=cptcovprod;k++){ /\* Useless because included in cptcovn *\/ */
5533: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
5534: /* if(Dummy[Tvard[k][1]]==0){ */
5535: /* if(Dummy[Tvard[k][2]]==0){ */
5536: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]; */
5537: /* }else{ */
5538: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
5539: /* } */
5540: /* }else{ */
5541: /* if(Dummy[Tvard[k][2]]==0){ */
5542: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
5543: /* }else{ */
5544: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
5545: /* } */
5546: /* } */
5547: /* } */
5548: /* /\*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*\/ */
5549: /* /\*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*\/ */
5550: /** End of old code */
5551:
1.218 brouard 5552: /* Careful transposed matrix */
1.266 brouard 5553: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 5554: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 5555: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 5556: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 5557: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 5558: /* if((int)age == 70){ */
5559: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
5560: /* for(i=1; i<=nlstate+ndeath; i++) { */
5561: /* printf("%d pmmij ",i); */
5562: /* for(j=1;j<=nlstate+ndeath;j++) { */
5563: /* printf("%f ",pmmij[i][j]); */
5564: /* } */
5565: /* printf(" oldm "); */
5566: /* for(j=1;j<=nlstate+ndeath;j++) { */
5567: /* printf("%f ",oldm[i][j]); */
5568: /* } */
5569: /* printf("\n"); */
5570: /* } */
5571: /* } */
5572: savm=oldm;
5573: oldm=newm;
5574: }
5575: for(i=1; i<=nlstate+ndeath; i++)
5576: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 5577: po[i][j][h]=newm[i][j];
1.268 brouard 5578: /* if(h==nhstepm) */
5579: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 5580: }
1.268 brouard 5581: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 5582: } /* end h */
1.268 brouard 5583: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 5584: return po;
5585: }
5586:
5587:
1.162 brouard 5588: #ifdef NLOPT
5589: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
5590: double fret;
5591: double *xt;
5592: int j;
5593: myfunc_data *d2 = (myfunc_data *) pd;
5594: /* xt = (p1-1); */
5595: xt=vector(1,n);
5596: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
5597:
5598: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
5599: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
5600: printf("Function = %.12lf ",fret);
5601: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
5602: printf("\n");
5603: free_vector(xt,1,n);
5604: return fret;
5605: }
5606: #endif
1.126 brouard 5607:
5608: /*************** log-likelihood *************/
5609: double func( double *x)
5610: {
1.336 brouard 5611: int i, ii, j, k, mi, d, kk, kf=0;
1.226 brouard 5612: int ioffset=0;
1.339 brouard 5613: int ipos=0,iposold=0,ncovv=0;
5614:
1.340 brouard 5615: double cotvarv, cotvarvold;
1.226 brouard 5616: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
5617: double **out;
5618: double lli; /* Individual log likelihood */
5619: int s1, s2;
1.228 brouard 5620: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
1.336 brouard 5621:
1.226 brouard 5622: double bbh, survp;
5623: double agexact;
1.336 brouard 5624: double agebegin, ageend;
1.226 brouard 5625: /*extern weight */
5626: /* We are differentiating ll according to initial status */
5627: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
5628: /*for(i=1;i<imx;i++)
5629: printf(" %d\n",s[4][i]);
5630: */
1.162 brouard 5631:
1.226 brouard 5632: ++countcallfunc;
1.162 brouard 5633:
1.226 brouard 5634: cov[1]=1.;
1.126 brouard 5635:
1.226 brouard 5636: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 5637: ioffset=0;
1.226 brouard 5638: if(mle==1){
5639: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
5640: /* Computes the values of the ncovmodel covariates of the model
5641: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
5642: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
5643: to be observed in j being in i according to the model.
5644: */
1.243 brouard 5645: ioffset=2+nagesqr ;
1.233 brouard 5646: /* Fixed */
1.345 brouard 5647: for (kf=1; kf<=ncovf;kf++){ /* For each fixed covariate dummy or quant or prod */
1.319 brouard 5648: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
5649: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
5650: /* TvarF[1]=Tvar[6]=2, TvarF[2]=Tvar[7]=7, TvarF[3]=Tvar[9]=1 ID of fixed covariates or product V2, V1*V2, V1 */
1.320 brouard 5651: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.336 brouard 5652: cov[ioffset+TvarFind[kf]]=covar[Tvar[TvarFind[kf]]][i];/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V1 is fixed (TvarFind[1]=6)*/
1.319 brouard 5653: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 5654: }
1.226 brouard 5655: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 5656: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 5657: has been calculated etc */
5658: /* For an individual i, wav[i] gives the number of effective waves */
5659: /* We compute the contribution to Likelihood of each effective transition
5660: mw[mi][i] is real wave of the mi th effectve wave */
5661: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
5662: s2=s[mw[mi+1][i]][i];
1.341 brouard 5663: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i] because now is moved after nvocol+nqv
1.226 brouard 5664: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
5665: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
5666: */
1.336 brouard 5667: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
5668: /* Wave varying (but not age varying) */
1.339 brouard 5669: /* for(k=1; k <= ncovv ; k++){ /\* Varying covariates in the model (single and product but no age )"V5+V4+V3+V4*V3+V5*age+V1*age+V1" +TvarVind 1,2,3,4(V4*V3) Tvar[1]@7{5, 4, 3, 6, 5, 1, 1 ; 6 because the created covar is after V5 and is 6, minus 1+1, 3,2,1,4 positions in cotvar*\/ */
5670: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? *\/ */
5671: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
5672: /* } */
1.340 brouard 5673: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age )*/
5674: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
5675: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
1.345 brouard 5676: if(FixedV[itv]!=0){ /* Not a fixed covariate */
1.341 brouard 5677: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* cotvar[wav][ncovcol+nqv+iv][i] */
1.340 brouard 5678: }else{ /* fixed covariate */
1.345 brouard 5679: cotvarv=covar[itv][i]; /* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model */
1.340 brouard 5680: }
1.339 brouard 5681: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 5682: cotvarvold=cotvarv;
5683: }else{ /* A second product */
5684: cotvarv=cotvarv*cotvarvold;
1.339 brouard 5685: }
5686: iposold=ipos;
1.340 brouard 5687: cov[ioffset+ipos]=cotvarv;
1.234 brouard 5688: }
1.339 brouard 5689: /* for products of time varying to be done */
1.234 brouard 5690: for (ii=1;ii<=nlstate+ndeath;ii++)
5691: for (j=1;j<=nlstate+ndeath;j++){
5692: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
5693: savm[ii][j]=(ii==j ? 1.0 : 0.0);
5694: }
1.336 brouard 5695:
5696: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
5697: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
1.234 brouard 5698: for(d=0; d<dh[mi][i]; d++){
5699: newm=savm;
5700: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
5701: cov[2]=agexact;
5702: if(nagesqr==1)
5703: cov[3]= agexact*agexact; /* Should be changed here */
1.349 brouard 5704: /* for (kk=1; kk<=cptcovage;kk++) { */
5705: /* if(!FixedV[Tvar[Tage[kk]]]) */
5706: /* cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /\* Tage[kk] gives the data-covariate associated with age *\/ */
5707: /* else */
5708: /* cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]*agexact; /\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
5709: /* } */
5710: for(ncovva=1, iposold=0; ncovva <= ncovta ; ncovva++){ /* Time varying covariates with age including individual from products, product is computed dynamically */
5711: itv=TvarAVVA[ncovva]; /* TvarVV={3, 1, 3} gives the name of each varying covariate, exploding product Vn*Vm into Vn and then Vm */
5712: ipos=TvarAVVAind[ncovva]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
5713: if(FixedV[itv]!=0){ /* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv */
5714: cotvarv=cotvar[mw[mi][i]][TvarAVVA[ncovva]][i]; /* because cotvar starts now at first ncovcol+nqv+ntv+nqtv (1 to nqtv) */
5715: }else{ /* fixed covariate */
5716: cotvarv=covar[itv][i]; /* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model */
5717: }
5718: if(ipos!=iposold){ /* Not a product or first of a product */
5719: cotvarvold=cotvarv;
5720: }else{ /* A second product */
5721: cotvarv=cotvarv*cotvarvold;
5722: }
5723: iposold=ipos;
5724: cov[ioffset+ipos]=cotvarv*agexact;
5725: /* For products */
1.234 brouard 5726: }
1.349 brouard 5727:
1.234 brouard 5728: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
5729: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
5730: savm=oldm;
5731: oldm=newm;
5732: } /* end mult */
5733:
5734: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
5735: /* But now since version 0.9 we anticipate for bias at large stepm.
5736: * If stepm is larger than one month (smallest stepm) and if the exact delay
5737: * (in months) between two waves is not a multiple of stepm, we rounded to
5738: * the nearest (and in case of equal distance, to the lowest) interval but now
5739: * we keep into memory the bias bh[mi][i] and also the previous matrix product
5740: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
5741: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 5742: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
5743: * -stepm/2 to stepm/2 .
5744: * For stepm=1 the results are the same as for previous versions of Imach.
5745: * For stepm > 1 the results are less biased than in previous versions.
5746: */
1.234 brouard 5747: s1=s[mw[mi][i]][i];
5748: s2=s[mw[mi+1][i]][i];
5749: bbh=(double)bh[mi][i]/(double)stepm;
5750: /* bias bh is positive if real duration
5751: * is higher than the multiple of stepm and negative otherwise.
5752: */
5753: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
5754: if( s2 > nlstate){
5755: /* i.e. if s2 is a death state and if the date of death is known
5756: then the contribution to the likelihood is the probability to
5757: die between last step unit time and current step unit time,
5758: which is also equal to probability to die before dh
5759: minus probability to die before dh-stepm .
5760: In version up to 0.92 likelihood was computed
5761: as if date of death was unknown. Death was treated as any other
5762: health state: the date of the interview describes the actual state
5763: and not the date of a change in health state. The former idea was
5764: to consider that at each interview the state was recorded
5765: (healthy, disable or death) and IMaCh was corrected; but when we
5766: introduced the exact date of death then we should have modified
5767: the contribution of an exact death to the likelihood. This new
5768: contribution is smaller and very dependent of the step unit
5769: stepm. It is no more the probability to die between last interview
5770: and month of death but the probability to survive from last
5771: interview up to one month before death multiplied by the
5772: probability to die within a month. Thanks to Chris
5773: Jackson for correcting this bug. Former versions increased
5774: mortality artificially. The bad side is that we add another loop
5775: which slows down the processing. The difference can be up to 10%
5776: lower mortality.
5777: */
5778: /* If, at the beginning of the maximization mostly, the
5779: cumulative probability or probability to be dead is
5780: constant (ie = 1) over time d, the difference is equal to
5781: 0. out[s1][3] = savm[s1][3]: probability, being at state
5782: s1 at precedent wave, to be dead a month before current
5783: wave is equal to probability, being at state s1 at
5784: precedent wave, to be dead at mont of the current
5785: wave. Then the observed probability (that this person died)
5786: is null according to current estimated parameter. In fact,
5787: it should be very low but not zero otherwise the log go to
5788: infinity.
5789: */
1.183 brouard 5790: /* #ifdef INFINITYORIGINAL */
5791: /* lli=log(out[s1][s2] - savm[s1][s2]); */
5792: /* #else */
5793: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
5794: /* lli=log(mytinydouble); */
5795: /* else */
5796: /* lli=log(out[s1][s2] - savm[s1][s2]); */
5797: /* #endif */
1.226 brouard 5798: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 5799:
1.226 brouard 5800: } else if ( s2==-1 ) { /* alive */
5801: for (j=1,survp=0. ; j<=nlstate; j++)
5802: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
5803: /*survp += out[s1][j]; */
5804: lli= log(survp);
5805: }
1.336 brouard 5806: /* else if (s2==-4) { */
5807: /* for (j=3,survp=0. ; j<=nlstate; j++) */
5808: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
5809: /* lli= log(survp); */
5810: /* } */
5811: /* else if (s2==-5) { */
5812: /* for (j=1,survp=0. ; j<=2; j++) */
5813: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
5814: /* lli= log(survp); */
5815: /* } */
1.226 brouard 5816: else{
5817: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
5818: /* lli= (savm[s1][s2]>(double)1.e-8 ?log((1.+bbh)*out[s1][s2]- bbh*(savm[s1][s2])):log((1.+bbh)*out[s1][s2]));*/ /* linear interpolation */
5819: }
5820: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
5821: /*if(lli ==000.0)*/
1.340 brouard 5822: /* printf("num[i], i=%d, bbh= %f lli=%f savm=%f out=%f %d\n",bbh,lli,savm[s1][s2], out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]],i); */
1.226 brouard 5823: ipmx +=1;
5824: sw += weight[i];
5825: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
5826: /* if (lli < log(mytinydouble)){ */
5827: /* printf("Close to inf lli = %.10lf < %.10lf i= %d mi= %d, s[%d][i]=%d s1=%d s2=%d\n", lli,log(mytinydouble), i, mi,mw[mi][i], s[mw[mi][i]][i], s1,s2); */
5828: /* fprintf(ficlog,"Close to inf lli = %.10lf i= %d mi= %d, s[mw[mi][i]][i]=%d\n", lli, i, mi,s[mw[mi][i]][i]); */
5829: /* } */
5830: } /* end of wave */
5831: } /* end of individual */
5832: } else if(mle==2){
5833: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 5834: ioffset=2+nagesqr ;
5835: for (k=1; k<=ncovf;k++)
5836: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 5837: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 5838: for(k=1; k <= ncovv ; k++){
1.341 brouard 5839: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.319 brouard 5840: }
1.226 brouard 5841: for (ii=1;ii<=nlstate+ndeath;ii++)
5842: for (j=1;j<=nlstate+ndeath;j++){
5843: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
5844: savm[ii][j]=(ii==j ? 1.0 : 0.0);
5845: }
5846: for(d=0; d<=dh[mi][i]; d++){
5847: newm=savm;
5848: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
5849: cov[2]=agexact;
5850: if(nagesqr==1)
5851: cov[3]= agexact*agexact;
5852: for (kk=1; kk<=cptcovage;kk++) {
5853: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
5854: }
5855: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
5856: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
5857: savm=oldm;
5858: oldm=newm;
5859: } /* end mult */
5860:
5861: s1=s[mw[mi][i]][i];
5862: s2=s[mw[mi+1][i]][i];
5863: bbh=(double)bh[mi][i]/(double)stepm;
5864: lli= (savm[s1][s2]>(double)1.e-8 ?log((1.+bbh)*out[s1][s2]- bbh*(savm[s1][s2])):log((1.+bbh)*out[s1][s2])); /* linear interpolation */
5865: ipmx +=1;
5866: sw += weight[i];
5867: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
5868: } /* end of wave */
5869: } /* end of individual */
5870: } else if(mle==3){ /* exponential inter-extrapolation */
5871: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
5872: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
5873: for(mi=1; mi<= wav[i]-1; mi++){
5874: for (ii=1;ii<=nlstate+ndeath;ii++)
5875: for (j=1;j<=nlstate+ndeath;j++){
5876: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
5877: savm[ii][j]=(ii==j ? 1.0 : 0.0);
5878: }
5879: for(d=0; d<dh[mi][i]; d++){
5880: newm=savm;
5881: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
5882: cov[2]=agexact;
5883: if(nagesqr==1)
5884: cov[3]= agexact*agexact;
5885: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 5886: if(!FixedV[Tvar[Tage[kk]]])
5887: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
5888: else
1.341 brouard 5889: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]*agexact; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.226 brouard 5890: }
5891: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
5892: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
5893: savm=oldm;
5894: oldm=newm;
5895: } /* end mult */
5896:
5897: s1=s[mw[mi][i]][i];
5898: s2=s[mw[mi+1][i]][i];
5899: bbh=(double)bh[mi][i]/(double)stepm;
5900: lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2])); /* exponential inter-extrapolation */
5901: ipmx +=1;
5902: sw += weight[i];
5903: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
5904: } /* end of wave */
5905: } /* end of individual */
5906: }else if (mle==4){ /* ml=4 no inter-extrapolation */
5907: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
5908: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
5909: for(mi=1; mi<= wav[i]-1; mi++){
5910: for (ii=1;ii<=nlstate+ndeath;ii++)
5911: for (j=1;j<=nlstate+ndeath;j++){
5912: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
5913: savm[ii][j]=(ii==j ? 1.0 : 0.0);
5914: }
5915: for(d=0; d<dh[mi][i]; d++){
5916: newm=savm;
5917: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
5918: cov[2]=agexact;
5919: if(nagesqr==1)
5920: cov[3]= agexact*agexact;
5921: for (kk=1; kk<=cptcovage;kk++) {
5922: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
5923: }
1.126 brouard 5924:
1.226 brouard 5925: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
5926: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
5927: savm=oldm;
5928: oldm=newm;
5929: } /* end mult */
5930:
5931: s1=s[mw[mi][i]][i];
5932: s2=s[mw[mi+1][i]][i];
5933: if( s2 > nlstate){
5934: lli=log(out[s1][s2] - savm[s1][s2]);
5935: } else if ( s2==-1 ) { /* alive */
5936: for (j=1,survp=0. ; j<=nlstate; j++)
5937: survp += out[s1][j];
5938: lli= log(survp);
5939: }else{
5940: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
5941: }
5942: ipmx +=1;
5943: sw += weight[i];
5944: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.343 brouard 5945: /* printf("num[i]=%09ld, i=%6d s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",num[i],i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.226 brouard 5946: } /* end of wave */
5947: } /* end of individual */
5948: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
5949: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
5950: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
5951: for(mi=1; mi<= wav[i]-1; mi++){
5952: for (ii=1;ii<=nlstate+ndeath;ii++)
5953: for (j=1;j<=nlstate+ndeath;j++){
5954: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
5955: savm[ii][j]=(ii==j ? 1.0 : 0.0);
5956: }
5957: for(d=0; d<dh[mi][i]; d++){
5958: newm=savm;
5959: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
5960: cov[2]=agexact;
5961: if(nagesqr==1)
5962: cov[3]= agexact*agexact;
5963: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 5964: if(!FixedV[Tvar[Tage[kk]]])
5965: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
5966: else
1.341 brouard 5967: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]*agexact; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.226 brouard 5968: }
1.126 brouard 5969:
1.226 brouard 5970: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
5971: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
5972: savm=oldm;
5973: oldm=newm;
5974: } /* end mult */
5975:
5976: s1=s[mw[mi][i]][i];
5977: s2=s[mw[mi+1][i]][i];
5978: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
5979: ipmx +=1;
5980: sw += weight[i];
5981: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
5982: /*printf("i=%6d s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],out[s1][s2],savm[s1][s2]);*/
5983: } /* end of wave */
5984: } /* end of individual */
5985: } /* End of if */
5986: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
5987: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
5988: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
5989: return -l;
1.126 brouard 5990: }
5991:
5992: /*************** log-likelihood *************/
5993: double funcone( double *x)
5994: {
1.228 brouard 5995: /* Same as func but slower because of a lot of printf and if */
1.359 brouard 5996: int i, ii, j, k, mi, d, kv=0, kf=0;
1.228 brouard 5997: int ioffset=0;
1.339 brouard 5998: int ipos=0,iposold=0,ncovv=0;
5999:
1.340 brouard 6000: double cotvarv, cotvarvold;
1.131 brouard 6001: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 6002: double **out;
6003: double lli; /* Individual log likelihood */
6004: double llt;
6005: int s1, s2;
1.228 brouard 6006: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
6007:
1.126 brouard 6008: double bbh, survp;
1.187 brouard 6009: double agexact;
1.214 brouard 6010: double agebegin, ageend;
1.126 brouard 6011: /*extern weight */
6012: /* We are differentiating ll according to initial status */
6013: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
6014: /*for(i=1;i<imx;i++)
6015: printf(" %d\n",s[4][i]);
6016: */
6017: cov[1]=1.;
6018:
6019: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 6020: ioffset=0;
6021: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.336 brouard 6022: /* Computes the values of the ncovmodel covariates of the model
6023: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
6024: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
6025: to be observed in j being in i according to the model.
6026: */
1.243 brouard 6027: /* ioffset=2+nagesqr+cptcovage; */
6028: ioffset=2+nagesqr;
1.232 brouard 6029: /* Fixed */
1.224 brouard 6030: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 6031: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.349 brouard 6032: for (kf=1; kf<=ncovf;kf++){ /* V2 + V3 + V4 Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
1.339 brouard 6033: /* printf("Debug3 TvarFind[%d]=%d",kf, TvarFind[kf]); */
6034: /* printf(" Tvar[TvarFind[kf]]=%d", Tvar[TvarFind[kf]]); */
6035: /* printf(" i=%d covar[Tvar[TvarFind[kf]]][i]=%f\n",i,covar[Tvar[TvarFind[kf]]][i]); */
1.335 brouard 6036: cov[ioffset+TvarFind[kf]]=covar[Tvar[TvarFind[kf]]][i];/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V1 is fixed (k=6)*/
1.232 brouard 6037: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
6038: /* cov[2+6]=covar[Tvar[6]][i]; */
6039: /* cov[2+6]=covar[2][i]; V2 */
6040: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
6041: /* cov[2+7]=covar[Tvar[7]][i]; */
6042: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
6043: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
6044: /* cov[2+9]=covar[Tvar[9]][i]; */
6045: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 6046: }
1.336 brouard 6047: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
6048: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
6049: has been calculated etc */
6050: /* For an individual i, wav[i] gives the number of effective waves */
6051: /* We compute the contribution to Likelihood of each effective transition
6052: mw[mi][i] is real wave of the mi th effectve wave */
6053: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
6054: s2=s[mw[mi+1][i]][i];
1.341 brouard 6055: And the iv th varying covariate in the DATA is the cotvar[mw[mi+1][i]][ncovcol+nqv+iv][i]
1.336 brouard 6056: */
6057: /* This part may be useless now because everythin should be in covar */
1.232 brouard 6058: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
6059: /* cov[++ioffset]=coqvar[TvarFQ[k]][i];/\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V2 and V1*V2 is fixed (k=6 and 7?)*\/ */
6060: /* } */
1.231 brouard 6061: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
6062: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
6063: /* } */
1.225 brouard 6064:
1.233 brouard 6065:
6066: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.339 brouard 6067: /* Wave varying (but not age varying) *//* V1+V3+age*V1+age*V3+V1*V3 with V4 tv and V5 tvq k= 1 to 5 and extra at V(5+1)=6 for V1*V3 */
6068: /* for(k=1; k <= ncovv ; k++){ /\* Varying covariates (single and product but no age )*\/ */
6069: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; *\/ */
6070: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
6071: /* } */
6072:
6073: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
6074: /* model V1+V3+age*V1+age*V3+V1*V3 */
6075: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
6076: /* TvarVV[1]=V3 (first time varying in the model equation, TvarVV[2]=V1 (in V1*V3) TvarVV[3]=3(V3) */
6077: /* We need the position of the time varying or product in the model */
6078: /* TvarVVind={2,5,5}, for V3 at position 2 and then the product V1*V3 is decomposed into V1 and V3 but at same position 5 */
6079: /* TvarVV gives the variable name */
1.340 brouard 6080: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
6081: * k= 1 2 3 4 5 6 7 8 9
6082: * varying 1 2 3 4 5
6083: * ncovv 1 2 3 4 5 6 7 8
1.343 brouard 6084: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
1.340 brouard 6085: * TvarVVind 2 3 7 7 8 8 9 9
6086: * TvarFind[k] 1 0 0 0 0 0 0 0 0
6087: */
1.345 brouard 6088: /* Other model ncovcol=5 nqv=0 ntv=3 nqtv=0 nlstate=3
1.349 brouard 6089: * V2 V3 V4 are fixed V6 V7 are timevarying so V8 and V5 are not in the model and product column will start at 9 Tvar[(v6*V2)6]=9
1.345 brouard 6090: * FixedV[ncovcol+qv+ntv+nqtv] V5
1.349 brouard 6091: * 3 V1 V2 V3 V4 V5 V6 V7 V8 V3*V2 V7*V2 V6*V3 V7*V3 V6*V4 V7*V4
6092: * 0 0 0 0 0 1 1 1 0, 0, 1,1, 1, 0, 1, 0, 1, 0, 1, 0}
6093: * 3 0 0 0 0 0 1 1 1 0, 1 1 1 1 1}
6094: * model= V2 + V3 + V4 + V6 + V7 + V6*V2 + V7*V2 + V6*V3 + V7*V3 + V6*V4 + V7*V4
6095: * +age*V2 +age*V3 +age*V4 +age*V6 + age*V7
6096: * +age*V6*V2 + age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4
6097: * model2= V2 + V3 + V4 + V6 + V7 + V3*V2 + V7*V2 + V6*V3 + V7*V3 + V6*V4 + V7*V4
6098: * +age*V2 +age*V3 +age*V4 +age*V6 + age*V7
6099: * +age*V3*V2 + age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4
6100: * model3= V2 + V3 + V4 + V6 + V7 + age*V3*V2 + V7*V2 + V6*V3 + V7*V3 + V6*V4 + V7*V4
6101: * +age*V2 +age*V3 +age*V4 +age*V6 + age*V7
6102: * +V3*V2 + age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4
6103: * kmodel 1 2 3 4 5 6 7 8 9 10 11
6104: * 12 13 14 15 16
6105: * 17 18 19 20 21
6106: * Tvar[kmodel] 2 3 4 6 7 9 10 11 12 13 14
6107: * 2 3 4 6 7
6108: * 9 11 12 13 14
6109: * cptcovage=5+5 total of covariates with age
6110: * Tage[cptcovage] age*V2=12 13 14 15 16
6111: *1 17 18 19 20 21 gives the position in model of covariates associated with age
6112: *3 Tage[cptcovage] age*V3*V2=6
6113: *3 age*V2=12 13 14 15 16
6114: *3 age*V6*V3=18 19 20 21
6115: * Tvar[Tage[cptcovage]] Tvar[12]=2 3 4 6 Tvar[16]=7(age*V7)
6116: * Tvar[17]age*V6*V2=9 Tvar[18]age*V6*V3=11 age*V7*V3=12 age*V6*V4=13 Tvar[21]age*V7*V4=14
6117: * 2 Tvar[17]age*V3*V2=9 Tvar[18]age*V6*V3=11 age*V7*V3=12 age*V6*V4=13 Tvar[21]age*V7*V4=14
6118: * 3 Tvar[Tage[cptcovage]] Tvar[6]=9 Tvar[12]=2 3 4 6 Tvar[16]=7(age*V7)
6119: * 3 Tvar[18]age*V6*V3=11 age*V7*V3=12 age*V6*V4=13 Tvar[21]age*V7*V4=14
6120: * 3 Tage[cptcovage] age*V3*V2=6 age*V2=12 age*V3 13 14 15 16
6121: * age*V6*V3=18 19 20 21 gives the position in model of covariates associated with age
6122: * 3 Tvar[17]age*V3*V2=9 Tvar[18]age*V6*V3=11 age*V7*V3=12 age*V6*V4=13 Tvar[21]age*V7*V4=14
6123: * Tvar= {2, 3, 4, 6, 7,
6124: * 9, 10, 11, 12, 13, 14,
6125: * Tvar[12]=2, 3, 4, 6, 7,
6126: * Tvar[17]=9, 11, 12, 13, 14}
6127: * Typevar[1]@21 = {0, 0, 0, 0, 0,
6128: * 2, 2, 2, 2, 2, 2,
6129: * 3 3, 2, 2, 2, 2, 2,
6130: * 1, 1, 1, 1, 1,
6131: * 3, 3, 3, 3, 3}
6132: * 3 2, 3, 3, 3, 3}
6133: * p Tposprod[1]@21 {0, 0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 0, 0, 0, 0, 0, 1, 3, 4, 5, 6} Id of the prod at position k in the model
6134: * p Tprod[1]@21 {6, 7, 8, 9, 10, 11, 0 <repeats 15 times>}
6135: * 3 Tposprod[1]@21 {0, 0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 0, 0, 0, 0, 0, 1, 3, 4, 5, 6}
6136: * 3 Tprod[1]@21 {17, 7, 8, 9, 10, 11, 0 <repeats 15 times>}
6137: * cptcovprod=11 (6+5)
6138: * FixedV[Tvar[Tage[cptcovage]]]] FixedV[2]=0 FixedV[3]=0 0 1 (age*V7)Tvar[16]=1 FixedV[absolute] not [kmodel]
6139: * FixedV[Tvar[17]=FixedV[age*V6*V2]=FixedV[9]=1 1 1 1 1
6140: * 3 FixedV[Tvar[17]=FixedV[age*V3*V2]=FixedV[9]=0 [11]=1 1 1 1
6141: * FixedV[] V1=0 V2=0 V3=0 v4=0 V5=0 V6=1 V7=1 v8=1 OK then model dependent
6142: * 9=1 [V7*V2]=[10]=1 11=1 12=1 13=1 14=1
6143: * 3 9=0 [V7*V2]=[10]=1 11=1 12=1 13=1 14=1
6144: * cptcovdageprod=5 for gnuplot printing
6145: * cptcovprodvage=6
6146: * ncova=15 1 2 3 4 5
6147: * 6 7 8 9 10 11 12 13 14 15
6148: * TvarA 2 3 4 6 7
6149: * 6 2 6 7 7 3 6 4 7 4
6150: * TvaAind 12 12 13 13 14 14 15 15 16 16
1.345 brouard 6151: * ncovf 1 2 3
1.349 brouard 6152: * V6 V7 V6*V2 V7*V2 V6*V3 V7*V3 V6*V4 V7*V4
6153: * ncovvt=14 1 2 3 4 5 6 7 8 9 10 11 12 13 14
6154: * TvarVV[1]@14 = itv {V6=6, 7, V6*V2=6, 2, 7, 2, 6, 3, 7, 3, 6, 4, 7, 4}
6155: * TvarVVind[1]@14= {4, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11}
6156: * 3 ncovvt=12 V6 V7 V7*V2 V6*V3 V7*V3 V6*V4 V7*V4
6157: * 3 TvarVV[1]@12 = itv {6, 7, V7*V2=7, 2, 6, 3, 7, 3, 6, 4, 7, 4}
6158: * 3 1 2 3 4 5 6 7 8 9 10 11 12
6159: * TvarVVind[1]@12= {V6 is in k=4, 5, 7,(4isV2)=7, 8, 8, 9, 9, 10,10, 11,11}TvarVVind[12]=k=11
6160: * TvarV 6, 7, 9, 10, 11, 12, 13, 14
6161: * 3 cptcovprodvage=6
6162: * 3 ncovta=15 +age*V3*V2+age*V2+agev3+ageV4 +age*V6 + age*V7 + age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4
6163: * 3 TvarAVVA[1]@15= itva 3 2 2 3 4 6 7 6 3 7 3 6 4 7 4
6164: * 3 ncovta 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1.354 brouard 6165: *?TvarAVVAind[1]@15= V3 is in k=2 1 1 2 3 4 5 4,2 5,2, 4,3 5 3}TvarVVAind[]
1.349 brouard 6166: * TvarAVVAind[1]@15= V3 is in k=6 6 12 13 14 15 16 18 18 19,19, 20,20 21,21}TvarVVAind[]
6167: * 3 ncovvta=10 +age*V6 + age*V7 + age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4
6168: * 3 we want to compute =cotvar[mw[mi][i]][TvarVVA[ncovva]][i] at position TvarVVAind[ncovva]
6169: * 3 TvarVVA[1]@10= itva 6 7 6 3 7 3 6 4 7 4
6170: * 3 ncovva 1 2 3 4 5 6 7 8 9 10
6171: * TvarVVAind[1]@10= V6 is in k=4 5 8,8 9, 9, 10,10 11 11}TvarVVAind[]
6172: * TvarVVAind[1]@10= 15 16 18,18 19,19, 20,20 21 21}TvarVVAind[]
6173: * TvarVA V3*V2=6 6 , 1, 2, 11, 12, 13, 14
1.345 brouard 6174: * TvarFind[1]@14= {1, 2, 3, 0 <repeats 12 times>}
1.349 brouard 6175: * Tvar[1]@21= {2, 3, 4, 6, 7, 9, 10, 11, 12, 13, 14,
6176: * 2, 3, 4, 6, 7,
6177: * 6, 8, 9, 10, 11}
1.345 brouard 6178: * TvarFind[itv] 0 0 0
6179: * FixedV[itv] 1 1 1 0 1 0 1 0 1 0 0
1.354 brouard 6180: *? FixedV[itv] 1 1 1 0 1 0 1 0 1 0 1 0 1 0
1.345 brouard 6181: * Tvar[TvarFind[ncovf]]=[1]=2 [2]=3 [4]=4
6182: * Tvar[TvarFind[itv]] [0]=? ?ncovv 1 à ncovvt]
6183: * Not a fixed cotvar[mw][itv][i] 6 7 6 2 7, 2, 6, 3, 7, 3, 6, 4, 7, 4}
1.349 brouard 6184: * fixed covar[itv] [6] [7] [6][2]
1.345 brouard 6185: */
6186:
1.349 brouard 6187: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* V6 V7 V7*V2 V6*V3 V7*V3 V6*V4 V7*V4 Time varying covariates (single and extended product but no age) including individual from products, product is computed dynamically */
6188: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate, or fixed covariate of a varying product after exploding product Vn*Vm into Vn and then Vm */
1.340 brouard 6189: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
1.345 brouard 6190: /* if(TvarFind[itv]==0){ /\* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv *\/ */
6191: if(FixedV[itv]!=0){ /* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv */
1.354 brouard 6192: /* printf("DEBUG ncovv=%d, Varying TvarVV[ncovv]=%d\n",ncovv, TvarVV[ncovv]); */
1.345 brouard 6193: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* because cotvar starts now at first ncovcol+nqv+ntv+nqtv (1 to nqtv) */
1.354 brouard 6194: /* printf("DEBUG Varying cov[ioffset+ipos=%d]=%g \n",ioffset+ipos,cotvarv); */
1.340 brouard 6195: }else{ /* fixed covariate */
1.345 brouard 6196: /* cotvarv=covar[Tvar[TvarFind[itv]]][i]; /\* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model *\/ */
1.354 brouard 6197: /* printf("DEBUG ncovv=%d, Fixed TvarVV[ncovv]=%d\n",ncovv, TvarVV[ncovv]); */
1.349 brouard 6198: cotvarv=covar[itv][i]; /* Good: In V6*V3, 3 is fixed at position of the data */
1.354 brouard 6199: /* printf("DEBUG Fixed cov[ioffset+ipos=%d]=%g \n",ioffset+ipos,cotvarv); */
1.340 brouard 6200: }
1.339 brouard 6201: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 6202: cotvarvold=cotvarv;
6203: }else{ /* A second product */
6204: cotvarv=cotvarv*cotvarvold;
1.339 brouard 6205: }
6206: iposold=ipos;
1.340 brouard 6207: cov[ioffset+ipos]=cotvarv;
1.354 brouard 6208: /* printf("DEBUG Product cov[ioffset+ipos=%d] \n",ioffset+ipos); */
1.339 brouard 6209: /* For products */
6210: }
6211: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates single *\/ */
6212: /* iv=TvarVDind[itv]; /\* iv, position in the model equation of time varying covariate itv *\/ */
6213: /* /\* "V1+V3+age*V1+age*V3+V1*V3" with V3 time varying *\/ */
6214: /* /\* 1 2 3 4 5 *\/ */
6215: /* /\*itv 1 *\/ */
6216: /* /\* TvarVInd[1]= 2 *\/ */
6217: /* /\* iv= Tvar[Tmodelind[itv]]-ncovcol-nqv; /\\* Counting the # varying covariate from 1 to ntveff *\\/ *\/ */
6218: /* /\* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; *\/ */
6219: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; *\/ */
6220: /* /\* k=ioffset-2-nagesqr-cptcovage+itv; /\\* position in simple model *\\/ *\/ */
6221: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; *\/ */
6222: /* cov[ioffset+iv]=cotvar[mw[mi][i]][itv][i]; */
6223: /* /\* printf(" i=%d,mi=%d,itv=%d,TmodelInvind[itv]=%d,cotvar[mw[mi][i]][itv][i]=%f\n", i, mi, itv, TvarVDind[itv],cotvar[mw[mi][i]][itv][i]); *\/ */
6224: /* } */
1.232 brouard 6225: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 6226: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
6227: /* /\* printf(" i=%d,mi=%d,iqtv=%d,TmodelInvQind[iqtv]=%d,cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]=%f\n", i, mi, iqtv, TmodelInvQind[iqtv],cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]); *\/ */
6228: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 6229: /* } */
1.126 brouard 6230: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 6231: for (j=1;j<=nlstate+ndeath;j++){
6232: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
6233: savm[ii][j]=(ii==j ? 1.0 : 0.0);
6234: }
1.214 brouard 6235:
6236: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
6237: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
6238: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 6239: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 6240: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
6241: and mw[mi+1][i]. dh depends on stepm.*/
6242: newm=savm;
1.247 brouard 6243: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 6244: cov[2]=agexact;
6245: if(nagesqr==1)
6246: cov[3]= agexact*agexact;
1.349 brouard 6247: for(ncovva=1, iposold=0; ncovva <= ncovta ; ncovva++){ /* Time varying covariates with age including individual from products, product is computed dynamically */
6248: itv=TvarAVVA[ncovva]; /* TvarVV={3, 1, 3} gives the name of each varying covariate, exploding product Vn*Vm into Vn and then Vm */
6249: ipos=TvarAVVAind[ncovva]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
6250: /* if(TvarFind[itv]==0){ /\* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv *\/ */
6251: if(FixedV[itv]!=0){ /* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv */
6252: /* printf("DEBUG ncovva=%d, Varying TvarAVVA[ncovva]=%d\n", ncovva, TvarAVVA[ncovva]); */
6253: cotvarv=cotvar[mw[mi][i]][TvarAVVA[ncovva]][i]; /* because cotvar starts now at first ncovcol+nqv+ntv+nqtv (1 to nqtv) */
6254: }else{ /* fixed covariate */
6255: /* cotvarv=covar[Tvar[TvarFind[itv]]][i]; /\* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model *\/ */
6256: /* printf("DEBUG ncovva=%d, Fixed TvarAVVA[ncovva]=%d\n", ncovva, TvarAVVA[ncovva]); */
6257: cotvarv=covar[itv][i]; /* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model */
6258: }
6259: if(ipos!=iposold){ /* Not a product or first of a product */
6260: cotvarvold=cotvarv;
6261: }else{ /* A second product */
6262: /* printf("DEBUG * \n"); */
6263: cotvarv=cotvarv*cotvarvold;
6264: }
6265: iposold=ipos;
6266: /* printf("DEBUG Product cov[ioffset+ipos=%d] \n",ioffset+ipos); */
6267: cov[ioffset+ipos]=cotvarv*agexact;
6268: /* For products */
1.242 brouard 6269: }
1.349 brouard 6270:
1.242 brouard 6271: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
6272: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
6273: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
6274: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
6275: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
6276: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
6277: savm=oldm;
6278: oldm=newm;
1.126 brouard 6279: } /* end mult */
1.336 brouard 6280: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
6281: /* But now since version 0.9 we anticipate for bias at large stepm.
6282: * If stepm is larger than one month (smallest stepm) and if the exact delay
6283: * (in months) between two waves is not a multiple of stepm, we rounded to
6284: * the nearest (and in case of equal distance, to the lowest) interval but now
6285: * we keep into memory the bias bh[mi][i] and also the previous matrix product
6286: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
6287: * probability in order to take into account the bias as a fraction of the way
6288: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
6289: * -stepm/2 to stepm/2 .
6290: * For stepm=1 the results are the same as for previous versions of Imach.
6291: * For stepm > 1 the results are less biased than in previous versions.
6292: */
1.126 brouard 6293: s1=s[mw[mi][i]][i];
6294: s2=s[mw[mi+1][i]][i];
1.217 brouard 6295: /* if(s2==-1){ */
1.268 brouard 6296: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 6297: /* /\* exit(1); *\/ */
6298: /* } */
1.126 brouard 6299: bbh=(double)bh[mi][i]/(double)stepm;
6300: /* bias is positive if real duration
6301: * is higher than the multiple of stepm and negative otherwise.
6302: */
6303: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 6304: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 6305: } else if ( s2==-1 ) { /* alive */
1.242 brouard 6306: for (j=1,survp=0. ; j<=nlstate; j++)
6307: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
6308: lli= log(survp);
1.126 brouard 6309: }else if (mle==1){
1.242 brouard 6310: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 6311: } else if(mle==2){
1.242 brouard 6312: lli= (savm[s1][s2]>(double)1.e-8 ?log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]):log((1.+bbh)*out[s1][s2])); /* linear interpolation */
1.126 brouard 6313: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 6314: lli= (savm[s1][s2]>(double)1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2])); /* exponential inter-extrapolation */
1.126 brouard 6315: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 6316: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 6317: } else{ /* mle=0 back to 1 */
1.242 brouard 6318: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
6319: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 6320: } /* End of if */
6321: ipmx +=1;
6322: sw += weight[i];
6323: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.342 brouard 6324: /* Printing covariates values for each contribution for checking */
1.343 brouard 6325: /* printf("num[i]=%09ld, i=%6d s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",num[i],i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.126 brouard 6326: if(globpr){
1.246 brouard 6327: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 6328: %11.6f %11.6f %11.6f ", \
1.242 brouard 6329: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
1.268 brouard 6330: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.343 brouard 6331: /* printf("%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\ */
6332: /* %11.6f %11.6f %11.6f ", \ */
6333: /* num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw, */
6334: /* 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.242 brouard 6335: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
6336: llt +=ll[k]*gipmx/gsw;
6337: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
1.335 brouard 6338: /* printf(" %10.6f",-ll[k]*gipmx/gsw); */
1.242 brouard 6339: }
1.343 brouard 6340: fprintf(ficresilk," %10.6f ", -llt);
1.335 brouard 6341: /* printf(" %10.6f\n", -llt); */
1.342 brouard 6342: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
1.343 brouard 6343: /* fprintf(ficresilk,"%09ld ", num[i]); */ /* not necessary */
6344: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
6345: fprintf(ficresilk," %g",covar[Tvar[TvarFind[kf]]][i]);
6346: }
6347: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
6348: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
6349: if(ipos!=iposold){ /* Not a product or first of a product */
6350: fprintf(ficresilk," %g",cov[ioffset+ipos]);
6351: /* printf(" %g",cov[ioffset+ipos]); */
6352: }else{
6353: fprintf(ficresilk,"*");
6354: /* printf("*"); */
1.342 brouard 6355: }
1.343 brouard 6356: iposold=ipos;
6357: }
1.349 brouard 6358: /* for (kk=1; kk<=cptcovage;kk++) { */
6359: /* if(!FixedV[Tvar[Tage[kk]]]){ */
6360: /* fprintf(ficresilk," %g*age",covar[Tvar[Tage[kk]]][i]); */
6361: /* /\* printf(" %g*age",covar[Tvar[Tage[kk]]][i]); *\/ */
6362: /* }else{ */
6363: /* fprintf(ficresilk," %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
6364: /* /\* printf(" %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/\\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\\/ *\/ */
6365: /* } */
6366: /* } */
6367: for(ncovva=1, iposold=0; ncovva <= ncovta ; ncovva++){ /* Time varying covariates with age including individual from products, product is computed dynamically */
6368: itv=TvarAVVA[ncovva]; /* TvarVV={3, 1, 3} gives the name of each varying covariate, exploding product Vn*Vm into Vn and then Vm */
6369: ipos=TvarAVVAind[ncovva]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
6370: /* if(TvarFind[itv]==0){ /\* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv *\/ */
6371: if(FixedV[itv]!=0){ /* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv */
6372: /* printf("DEBUG ncovva=%d, Varying TvarAVVA[ncovva]=%d\n", ncovva, TvarAVVA[ncovva]); */
6373: cotvarv=cotvar[mw[mi][i]][TvarAVVA[ncovva]][i]; /* because cotvar starts now at first ncovcol+nqv+ntv+nqtv (1 to nqtv) */
6374: }else{ /* fixed covariate */
6375: /* cotvarv=covar[Tvar[TvarFind[itv]]][i]; /\* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model *\/ */
6376: /* printf("DEBUG ncovva=%d, Fixed TvarAVVA[ncovva]=%d\n", ncovva, TvarAVVA[ncovva]); */
6377: cotvarv=covar[itv][i]; /* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model */
6378: }
6379: if(ipos!=iposold){ /* Not a product or first of a product */
6380: cotvarvold=cotvarv;
6381: }else{ /* A second product */
6382: /* printf("DEBUG * \n"); */
6383: cotvarv=cotvarv*cotvarvold;
1.342 brouard 6384: }
1.349 brouard 6385: cotvarv=cotvarv*agexact;
6386: fprintf(ficresilk," %g*age",cotvarv);
6387: iposold=ipos;
6388: /* printf("DEBUG Product cov[ioffset+ipos=%d] \n",ioffset+ipos); */
6389: cov[ioffset+ipos]=cotvarv;
6390: /* For products */
1.343 brouard 6391: }
6392: /* printf("\n"); */
1.342 brouard 6393: /* } /\* End debugILK *\/ */
6394: fprintf(ficresilk,"\n");
6395: } /* End if globpr */
1.335 brouard 6396: } /* end of wave */
6397: } /* end of individual */
6398: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
1.232 brouard 6399: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
1.335 brouard 6400: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
6401: if(globpr==0){ /* First time we count the contributions and weights */
6402: gipmx=ipmx;
6403: gsw=sw;
6404: }
1.343 brouard 6405: return -l;
1.126 brouard 6406: }
6407:
6408:
6409: /*************** function likelione ***********/
1.292 brouard 6410: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 6411: {
6412: /* This routine should help understanding what is done with
6413: the selection of individuals/waves and
6414: to check the exact contribution to the likelihood.
6415: Plotting could be done.
1.342 brouard 6416: */
6417: void pstamp(FILE *ficres);
1.343 brouard 6418: int k, kf, kk, kvar, ncovv, iposold, ipos;
1.126 brouard 6419:
6420: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 6421: strcpy(fileresilk,"ILK_");
1.202 brouard 6422: strcat(fileresilk,fileresu);
1.126 brouard 6423: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
6424: printf("Problem with resultfile: %s\n", fileresilk);
6425: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
6426: }
1.342 brouard 6427: pstamp(ficresilk);fprintf(ficresilk,"# model=1+age+%s\n",model);
1.214 brouard 6428: fprintf(ficresilk, "#individual(line's_record) count ageb ageend s1 s2 wave# effective_wave# number_of_matrices_product pij weight weight/gpw -2ln(pij)*weight 0pij_x 0pij_(x-stepm) cumulating_loglikeli_by_health_state(reweighted=-2ll*weightXnumber_of_contribs/sum_of_weights) and_total\n");
6429: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 6430: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
6431: for(k=1; k<=nlstate; k++)
6432: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
1.342 brouard 6433: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total) ");
6434:
6435: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
6436: for(kf=1;kf <= ncovf; kf++){
6437: fprintf(ficresilk,"V%d",Tvar[TvarFind[kf]]);
6438: /* printf("V%d",Tvar[TvarFind[kf]]); */
6439: }
6440: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){
1.343 brouard 6441: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
1.342 brouard 6442: if(ipos!=iposold){ /* Not a product or first of a product */
6443: /* printf(" %d",ipos); */
6444: fprintf(ficresilk," V%d",TvarVV[ncovv]);
6445: }else{
6446: /* printf("*"); */
6447: fprintf(ficresilk,"*");
1.343 brouard 6448: }
1.342 brouard 6449: iposold=ipos;
6450: }
6451: for (kk=1; kk<=cptcovage;kk++) {
6452: if(!FixedV[Tvar[Tage[kk]]]){
6453: /* printf(" %d*age(Fixed)",Tvar[Tage[kk]]); */
6454: fprintf(ficresilk," %d*age(Fixed)",Tvar[Tage[kk]]);
6455: }else{
6456: fprintf(ficresilk," %d*age(Varying)",Tvar[Tage[kk]]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
6457: /* printf(" %d*age(Varying)",Tvar[Tage[kk]]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
6458: }
6459: }
6460: /* } /\* End if debugILK *\/ */
6461: /* printf("\n"); */
6462: fprintf(ficresilk,"\n");
6463: } /* End glogpri */
1.126 brouard 6464:
1.292 brouard 6465: *fretone=(*func)(p);
1.126 brouard 6466: if(*globpri !=0){
6467: fclose(ficresilk);
1.205 brouard 6468: if (mle ==0)
6469: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
6470: else if(mle >=1)
6471: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
6472: fprintf(fichtm," You should at least run with mle >= 1 to get starting values corresponding to the optimized parameters in order to visualize the real contribution of each individual/wave: <a href=\"%s\">%s</a><br>\n",subdirf(fileresilk),subdirf(fileresilk));
1.274 brouard 6473: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 6474:
1.207 brouard 6475: fprintf(fichtm,"<br>- The function drawn is -2Log(L) in Log scale: by state of origin <a href=\"%s-ori.png\">%s-ori.png</a><br> \
1.343 brouard 6476: <img src=\"%s-ori.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 6477: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.343 brouard 6478: <img src=\"%s-dest.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
6479:
6480: for (k=1; k<= nlstate ; k++) {
6481: fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j. Dot's sizes are related to corresponding weight: <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br>\n \
6482: <img src=\"%s-p%dj.png\">\n",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
6483: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
1.350 brouard 6484: kvar=Tvar[TvarFind[kf]]; /* variable */
6485: fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j with colored covariate V%d. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): ",k,k,Tvar[TvarFind[kf]],Tvar[TvarFind[kf]],Tvar[TvarFind[kf]]);
6486: fprintf(fichtm,"<a href=\"%s-p%dj-%d.png\">%s-p%dj-%d.png</a><br>",subdirf2(optionfilefiname,"ILK_"),k,kvar,subdirf2(optionfilefiname,"ILK_"),k,kvar);
6487: fprintf(fichtm,"<img src=\"%s-p%dj-%d.png\">",subdirf2(optionfilefiname,"ILK_"),k,Tvar[TvarFind[kf]]);
1.343 brouard 6488: }
6489: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Loop on the time varying extended covariates (with extension of Vn*Vm */
6490: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
6491: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
6492: /* printf("DebugILK fichtm ncovv=%d, kvar=TvarVV[ncovv]=V%d, ipos=TvarVVind[ncovv]=%d, Dummy[ipos]=%d, Typevar[ipos]=%d\n", ncovv,kvar,ipos,Dummy[ipos],Typevar[ipos]); */
6493: if(ipos!=iposold){ /* Not a product or first of a product */
6494: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
6495: /* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); */
6496: if(Dummy[ipos]==0 && Typevar[ipos]==0){ /* Only if dummy time varying: Dummy(0, 1=quant singor prod without age,2 dummy*age, 3quant*age) Typevar (0 single, 1=*age,2=Vn*vm) */
6497: fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j with colored time varying dummy covariate V%d. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \
6498: <img src=\"%s-p%dj-%d.png\">",k,k,kvar,kvar,kvar,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,kvar);
6499: } /* End only for dummies time varying (single?) */
6500: }else{ /* Useless product */
6501: /* printf("*"); */
6502: /* fprintf(ficresilk,"*"); */
6503: }
6504: iposold=ipos;
6505: } /* For each time varying covariate */
6506: } /* End loop on states */
6507:
6508: /* if(debugILK){ */
6509: /* for(kf=1; kf <= ncovf; kf++){ /\* For each simple dummy covariate of the model *\/ */
6510: /* /\* kvar=Tvar[TvarFind[kf]]; *\/ /\* variable *\/ */
6511: /* for (k=1; k<= nlstate ; k++) { */
6512: /* fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j with colored covariate V%. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \ */
6513: /* <img src=\"%s-p%dj-%d.png\">",k,k,Tvar[TvarFind[kf]],Tvar[TvarFind[kf]],Tvar[TvarFind[kf]],subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,Tvar[TvarFind[kf]]); */
6514: /* } */
6515: /* } */
6516: /* for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /\* Loop on the time varying extended covariates (with extension of Vn*Vm *\/ */
6517: /* ipos=TvarVVind[ncovv]; /\* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate *\/ */
6518: /* kvar=TvarVV[ncovv]; /\* TvarVV={3, 1, 3} gives the name of each varying covariate *\/ */
6519: /* /\* printf("DebugILK fichtm ncovv=%d, kvar=TvarVV[ncovv]=V%d, ipos=TvarVVind[ncovv]=%d, Dummy[ipos]=%d, Typevar[ipos]=%d\n", ncovv,kvar,ipos,Dummy[ipos],Typevar[ipos]); *\/ */
6520: /* if(ipos!=iposold){ /\* Not a product or first of a product *\/ */
6521: /* /\* fprintf(ficresilk," V%d",TvarVV[ncovv]); *\/ */
6522: /* /\* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); *\/ */
6523: /* if(Dummy[ipos]==0 && Typevar[ipos]==0){ /\* Only if dummy time varying: Dummy(0, 1=quant singor prod without age,2 dummy*age, 3quant*age) Typevar (0 single, 1=*age,2=Vn*vm) *\/ */
6524: /* for (k=1; k<= nlstate ; k++) { */
6525: /* fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \ */
6526: /* <img src=\"%s-p%dj-%d.png\">",k,k,kvar,kvar,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,kvar); */
6527: /* } /\* End state *\/ */
6528: /* } /\* End only for dummies time varying (single?) *\/ */
6529: /* }else{ /\* Useless product *\/ */
6530: /* /\* printf("*"); *\/ */
6531: /* /\* fprintf(ficresilk,"*"); *\/ */
6532: /* } */
6533: /* iposold=ipos; */
6534: /* } /\* For each time varying covariate *\/ */
6535: /* }/\* End debugILK *\/ */
1.207 brouard 6536: fflush(fichtm);
1.343 brouard 6537: }/* End globpri */
1.126 brouard 6538: return;
6539: }
6540:
6541:
6542: /*********** Maximum Likelihood Estimation ***************/
6543:
6544: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
6545: {
1.359 brouard 6546: int i,j, jkk=0, iter=0;
1.126 brouard 6547: double **xi;
1.359 brouard 6548: /*double fret;*/
6549: /*double fretone;*/ /* Only one call to likelihood */
1.126 brouard 6550: /* char filerespow[FILENAMELENGTH];*/
1.354 brouard 6551:
1.359 brouard 6552: /*double * p1;*/ /* Shifted parameters from 0 instead of 1 */
1.162 brouard 6553: #ifdef NLOPT
6554: int creturn;
6555: nlopt_opt opt;
6556: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
6557: double *lb;
6558: double minf; /* the minimum objective value, upon return */
1.354 brouard 6559:
1.162 brouard 6560: myfunc_data dinst, *d = &dinst;
6561: #endif
6562:
6563:
1.126 brouard 6564: xi=matrix(1,npar,1,npar);
1.357 brouard 6565: for (i=1;i<=npar;i++) /* Starting with canonical directions j=1,n xi[i=1,n][j] */
1.126 brouard 6566: for (j=1;j<=npar;j++)
6567: xi[i][j]=(i==j ? 1.0 : 0.0);
1.359 brouard 6568: printf("Powell-prax\n"); fprintf(ficlog,"Powell-prax\n");
1.201 brouard 6569: strcpy(filerespow,"POW_");
1.126 brouard 6570: strcat(filerespow,fileres);
6571: if((ficrespow=fopen(filerespow,"w"))==NULL) {
6572: printf("Problem with resultfile: %s\n", filerespow);
6573: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
6574: }
6575: fprintf(ficrespow,"# Powell\n# iter -2*LL");
6576: for (i=1;i<=nlstate;i++)
6577: for(j=1;j<=nlstate+ndeath;j++)
6578: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
6579: fprintf(ficrespow,"\n");
1.162 brouard 6580: #ifdef POWELL
1.319 brouard 6581: #ifdef LINMINORIGINAL
6582: #else /* LINMINORIGINAL */
6583:
6584: flatdir=ivector(1,npar);
6585: for (j=1;j<=npar;j++) flatdir[j]=0;
6586: #endif /*LINMINORIGINAL */
6587:
6588: #ifdef FLATSUP
6589: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
6590: /* reorganizing p by suppressing flat directions */
6591: for(i=1, jk=1; i <=nlstate; i++){
6592: for(k=1; k <=(nlstate+ndeath); k++){
6593: if (k != i) {
6594: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
6595: if(flatdir[jk]==1){
6596: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
6597: }
6598: for(j=1; j <=ncovmodel; j++){
6599: printf("%12.7f ",p[jk]);
6600: jk++;
6601: }
6602: printf("\n");
6603: }
6604: }
6605: }
6606: /* skipping */
6607: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
6608: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
6609: for(k=1; k <=(nlstate+ndeath); k++){
6610: if (k != i) {
6611: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
6612: if(flatdir[jk]==1){
6613: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
6614: for(j=1; j <=ncovmodel; jk++,j++){
6615: printf(" p[%d]=%12.7f",jk, p[jk]);
6616: /*q[jjk]=p[jk];*/
6617: }
6618: }else{
6619: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
6620: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
6621: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
6622: /*q[jjk]=p[jk];*/
6623: }
6624: }
6625: printf("\n");
6626: }
6627: fflush(stdout);
6628: }
6629: }
6630: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
6631: #else /* FLATSUP */
1.359 brouard 6632: /* powell(p,xi,npar,ftol,&iter,&fret,func);*/
6633: /* praxis ( t0, h0, n, prin, x, beale_f ); */
1.364 brouard 6634: int prin=4;
1.362 brouard 6635: /* double h0=0.25; */
6636: /* double macheps; */
6637: /* double fmin; */
1.359 brouard 6638: macheps=pow(16.0,-13.0);
6639: /* #include "praxis.h" */
6640: /* Be careful that praxis start at x[0] and powell start at p[1] */
6641: /* praxis ( ftol, h0, npar, prin, p, func ); */
6642: /* p1= (p+1); */ /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
6643: printf("Praxis Gegenfurtner \n");
6644: fprintf(ficlog, "Praxis Gegenfurtner\n");fflush(ficlog);
6645: /* praxis ( ftol, h0, npar, prin, p1, func ); */
6646: /* fmin = praxis(1.e-5,macheps, h, n, prin, x, func); */
1.362 brouard 6647: ffmin = praxis(ftol,macheps, h0, npar, prin, p, func);
1.359 brouard 6648: printf("End Praxis\n");
1.319 brouard 6649: #endif /* FLATSUP */
6650:
6651: #ifdef LINMINORIGINAL
6652: #else
6653: free_ivector(flatdir,1,npar);
6654: #endif /* LINMINORIGINAL*/
6655: #endif /* POWELL */
1.126 brouard 6656:
1.162 brouard 6657: #ifdef NLOPT
6658: #ifdef NEWUOA
6659: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
6660: #else
6661: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
6662: #endif
6663: lb=vector(0,npar-1);
6664: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
6665: nlopt_set_lower_bounds(opt, lb);
6666: nlopt_set_initial_step1(opt, 0.1);
6667:
6668: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
6669: d->function = func;
6670: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
6671: nlopt_set_min_objective(opt, myfunc, d);
6672: nlopt_set_xtol_rel(opt, ftol);
6673: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
6674: printf("nlopt failed! %d\n",creturn);
6675: }
6676: else {
6677: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
6678: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
6679: iter=1; /* not equal */
6680: }
6681: nlopt_destroy(opt);
6682: #endif
1.319 brouard 6683: #ifdef FLATSUP
6684: /* npared = npar -flatd/ncovmodel; */
6685: /* xired= matrix(1,npared,1,npared); */
6686: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
6687: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
6688: /* free_matrix(xire,1,npared,1,npared); */
6689: #else /* FLATSUP */
6690: #endif /* FLATSUP */
1.126 brouard 6691: free_matrix(xi,1,npar,1,npar);
6692: fclose(ficrespow);
1.203 brouard 6693: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
6694: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 6695: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 6696:
6697: }
6698:
6699: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 6700: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 6701: {
6702: double **a,**y,*x,pd;
1.203 brouard 6703: /* double **hess; */
1.164 brouard 6704: int i, j;
1.126 brouard 6705: int *indx;
6706:
6707: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 6708: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 6709: void lubksb(double **a, int npar, int *indx, double b[]) ;
6710: void ludcmp(double **a, int npar, int *indx, double *d) ;
6711: double gompertz(double p[]);
1.203 brouard 6712: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 6713:
6714: printf("\nCalculation of the hessian matrix. Wait...\n");
6715: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
6716: for (i=1;i<=npar;i++){
1.203 brouard 6717: printf("%d-",i);fflush(stdout);
6718: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 6719:
6720: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
6721:
6722: /* printf(" %f ",p[i]);
6723: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
6724: }
6725:
6726: for (i=1;i<=npar;i++) {
6727: for (j=1;j<=npar;j++) {
6728: if (j>i) {
1.203 brouard 6729: printf(".%d-%d",i,j);fflush(stdout);
6730: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
6731: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 6732:
6733: hess[j][i]=hess[i][j];
6734: /*printf(" %lf ",hess[i][j]);*/
6735: }
6736: }
6737: }
6738: printf("\n");
6739: fprintf(ficlog,"\n");
6740:
6741: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
6742: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
6743:
6744: a=matrix(1,npar,1,npar);
6745: y=matrix(1,npar,1,npar);
6746: x=vector(1,npar);
6747: indx=ivector(1,npar);
6748: for (i=1;i<=npar;i++)
6749: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
6750: ludcmp(a,npar,indx,&pd);
6751:
6752: for (j=1;j<=npar;j++) {
6753: for (i=1;i<=npar;i++) x[i]=0;
6754: x[j]=1;
6755: lubksb(a,npar,indx,x);
6756: for (i=1;i<=npar;i++){
6757: matcov[i][j]=x[i];
6758: }
6759: }
6760:
6761: printf("\n#Hessian matrix#\n");
6762: fprintf(ficlog,"\n#Hessian matrix#\n");
6763: for (i=1;i<=npar;i++) {
6764: for (j=1;j<=npar;j++) {
1.203 brouard 6765: printf("%.6e ",hess[i][j]);
6766: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 6767: }
6768: printf("\n");
6769: fprintf(ficlog,"\n");
6770: }
6771:
1.203 brouard 6772: /* printf("\n#Covariance matrix#\n"); */
6773: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
6774: /* for (i=1;i<=npar;i++) { */
6775: /* for (j=1;j<=npar;j++) { */
6776: /* printf("%.6e ",matcov[i][j]); */
6777: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
6778: /* } */
6779: /* printf("\n"); */
6780: /* fprintf(ficlog,"\n"); */
6781: /* } */
6782:
1.126 brouard 6783: /* Recompute Inverse */
1.203 brouard 6784: /* for (i=1;i<=npar;i++) */
6785: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
6786: /* ludcmp(a,npar,indx,&pd); */
6787:
6788: /* printf("\n#Hessian matrix recomputed#\n"); */
6789:
6790: /* for (j=1;j<=npar;j++) { */
6791: /* for (i=1;i<=npar;i++) x[i]=0; */
6792: /* x[j]=1; */
6793: /* lubksb(a,npar,indx,x); */
6794: /* for (i=1;i<=npar;i++){ */
6795: /* y[i][j]=x[i]; */
6796: /* printf("%.3e ",y[i][j]); */
6797: /* fprintf(ficlog,"%.3e ",y[i][j]); */
6798: /* } */
6799: /* printf("\n"); */
6800: /* fprintf(ficlog,"\n"); */
6801: /* } */
6802:
6803: /* Verifying the inverse matrix */
6804: #ifdef DEBUGHESS
6805: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 6806:
1.203 brouard 6807: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
6808: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 6809:
6810: for (j=1;j<=npar;j++) {
6811: for (i=1;i<=npar;i++){
1.203 brouard 6812: printf("%.2f ",y[i][j]);
6813: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 6814: }
6815: printf("\n");
6816: fprintf(ficlog,"\n");
6817: }
1.203 brouard 6818: #endif
1.126 brouard 6819:
6820: free_matrix(a,1,npar,1,npar);
6821: free_matrix(y,1,npar,1,npar);
6822: free_vector(x,1,npar);
6823: free_ivector(indx,1,npar);
1.203 brouard 6824: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 6825:
6826:
6827: }
6828:
6829: /*************** hessian matrix ****************/
6830: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 6831: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 6832: int i;
6833: int l=1, lmax=20;
1.203 brouard 6834: double k1,k2, res, fx;
1.132 brouard 6835: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 6836: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
6837: int k=0,kmax=10;
6838: double l1;
6839:
6840: fx=func(x);
6841: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 6842: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 6843: l1=pow(10,l);
6844: delts=delt;
6845: for(k=1 ; k <kmax; k=k+1){
6846: delt = delta*(l1*k);
6847: p2[theta]=x[theta] +delt;
1.145 brouard 6848: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 6849: p2[theta]=x[theta]-delt;
6850: k2=func(p2)-fx;
6851: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 6852: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 6853:
1.203 brouard 6854: #ifdef DEBUGHESSII
1.126 brouard 6855: printf("%d %d k1=%.12e k2=%.12e xk1=%.12e xk2=%.12e delt=%.12e res=%.12e l=%d k=%d,fx=%.12e\n",theta,theta,k1,k2,x[theta]+delt,x[theta]-delt,delt,res, l, k,fx);
6856: fprintf(ficlog,"%d %d k1=%.12e k2=%.12e xk1=%.12e xk2=%.12e delt=%.12e res=%.12e l=%d k=%d,fx=%.12e\n",theta,theta,k1,k2,x[theta]+delt,x[theta]-delt,delt,res, l, k,fx);
6857: #endif
6858: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
6859: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
6860: k=kmax;
6861: }
6862: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 6863: k=kmax; l=lmax*10;
1.126 brouard 6864: }
6865: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
6866: delts=delt;
6867: }
1.203 brouard 6868: } /* End loop k */
1.126 brouard 6869: }
6870: delti[theta]=delts;
6871: return res;
6872:
6873: }
6874:
1.203 brouard 6875: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 6876: {
6877: int i;
1.164 brouard 6878: int l=1, lmax=20;
1.126 brouard 6879: double k1,k2,k3,k4,res,fx;
1.132 brouard 6880: double p2[MAXPARM+1];
1.203 brouard 6881: int k, kmax=1;
6882: double v1, v2, cv12, lc1, lc2;
1.208 brouard 6883:
6884: int firstime=0;
1.203 brouard 6885:
1.126 brouard 6886: fx=func(x);
1.203 brouard 6887: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 6888: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 6889: p2[thetai]=x[thetai]+delti[thetai]*k;
6890: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 6891: k1=func(p2)-fx;
6892:
1.203 brouard 6893: p2[thetai]=x[thetai]+delti[thetai]*k;
6894: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 6895: k2=func(p2)-fx;
6896:
1.203 brouard 6897: p2[thetai]=x[thetai]-delti[thetai]*k;
6898: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 6899: k3=func(p2)-fx;
6900:
1.203 brouard 6901: p2[thetai]=x[thetai]-delti[thetai]*k;
6902: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 6903: k4=func(p2)-fx;
1.203 brouard 6904: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
6905: if(k1*k2*k3*k4 <0.){
1.208 brouard 6906: firstime=1;
1.203 brouard 6907: kmax=kmax+10;
1.208 brouard 6908: }
6909: if(kmax >=10 || firstime ==1){
1.354 brouard 6910: /* What are the thetai and thetaj? thetai/ncovmodel thetai=(thetai-thetai%ncovmodel)/ncovmodel +thetai%ncovmodel=(line,pos) */
1.246 brouard 6911: printf("Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you could increase ftol=%.2e\n",thetai,thetaj, ftol);
6912: fprintf(ficlog,"Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you could increase ftol=%.2e\n",thetai,thetaj, ftol);
1.203 brouard 6913: printf("%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti*k=%.12e deltj*k=%.12e, xi-de*k=%.12e xj-de*k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
6914: fprintf(ficlog,"%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti*k=%.12e deltj*k=%.12e, xi-de*k=%.12e xj-de*k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
6915: }
6916: #ifdef DEBUGHESSIJ
6917: v1=hess[thetai][thetai];
6918: v2=hess[thetaj][thetaj];
6919: cv12=res;
6920: /* Computing eigen value of Hessian matrix */
6921: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6922: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6923: if ((lc2 <0) || (lc1 <0) ){
6924: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
6925: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
6926: printf("%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti/k=%.12e deltj/k=%.12e, xi-de/k=%.12e xj-de/k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
6927: fprintf(ficlog,"%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti/k=%.12e deltj/k=%.12e, xi-de/k=%.12e xj-de/k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
6928: }
1.126 brouard 6929: #endif
6930: }
6931: return res;
6932: }
6933:
1.203 brouard 6934: /* Not done yet: Was supposed to fix if not exactly at the maximum */
6935: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
6936: /* { */
6937: /* int i; */
6938: /* int l=1, lmax=20; */
6939: /* double k1,k2,k3,k4,res,fx; */
6940: /* double p2[MAXPARM+1]; */
6941: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
6942: /* int k=0,kmax=10; */
6943: /* double l1; */
6944:
6945: /* fx=func(x); */
6946: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
6947: /* l1=pow(10,l); */
6948: /* delts=delt; */
6949: /* for(k=1 ; k <kmax; k=k+1){ */
6950: /* delt = delti*(l1*k); */
6951: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
6952: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
6953: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
6954: /* k1=func(p2)-fx; */
6955:
6956: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
6957: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
6958: /* k2=func(p2)-fx; */
6959:
6960: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
6961: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
6962: /* k3=func(p2)-fx; */
6963:
6964: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
6965: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
6966: /* k4=func(p2)-fx; */
6967: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
6968: /* #ifdef DEBUGHESSIJ */
6969: /* printf("%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti/k=%.12e deltj/k=%.12e, xi-de/k=%.12e xj-de/k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4); */
6970: /* fprintf(ficlog,"%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti/k=%.12e deltj/k=%.12e, xi-de/k=%.12e xj-de/k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4); */
6971: /* #endif */
6972: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
6973: /* k=kmax; */
6974: /* } */
6975: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
6976: /* k=kmax; l=lmax*10; */
6977: /* } */
6978: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
6979: /* delts=delt; */
6980: /* } */
6981: /* } /\* End loop k *\/ */
6982: /* } */
6983: /* delti[theta]=delts; */
6984: /* return res; */
6985: /* } */
6986:
6987:
1.126 brouard 6988: /************** Inverse of matrix **************/
6989: void ludcmp(double **a, int n, int *indx, double *d)
6990: {
6991: int i,imax,j,k;
6992: double big,dum,sum,temp;
6993: double *vv;
6994:
6995: vv=vector(1,n);
6996: *d=1.0;
6997: for (i=1;i<=n;i++) {
6998: big=0.0;
6999: for (j=1;j<=n;j++)
7000: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 7001: if (big == 0.0){
7002: printf(" Singular Hessian matrix at row %d:\n",i);
7003: for (j=1;j<=n;j++) {
7004: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
7005: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
7006: }
7007: fflush(ficlog);
7008: fclose(ficlog);
7009: nrerror("Singular matrix in routine ludcmp");
7010: }
1.126 brouard 7011: vv[i]=1.0/big;
7012: }
7013: for (j=1;j<=n;j++) {
7014: for (i=1;i<j;i++) {
7015: sum=a[i][j];
7016: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
7017: a[i][j]=sum;
7018: }
7019: big=0.0;
7020: for (i=j;i<=n;i++) {
7021: sum=a[i][j];
7022: for (k=1;k<j;k++)
7023: sum -= a[i][k]*a[k][j];
7024: a[i][j]=sum;
7025: if ( (dum=vv[i]*fabs(sum)) >= big) {
7026: big=dum;
7027: imax=i;
7028: }
7029: }
7030: if (j != imax) {
7031: for (k=1;k<=n;k++) {
7032: dum=a[imax][k];
7033: a[imax][k]=a[j][k];
7034: a[j][k]=dum;
7035: }
7036: *d = -(*d);
7037: vv[imax]=vv[j];
7038: }
7039: indx[j]=imax;
7040: if (a[j][j] == 0.0) a[j][j]=TINY;
7041: if (j != n) {
7042: dum=1.0/(a[j][j]);
7043: for (i=j+1;i<=n;i++) a[i][j] *= dum;
7044: }
7045: }
7046: free_vector(vv,1,n); /* Doesn't work */
7047: ;
7048: }
7049:
7050: void lubksb(double **a, int n, int *indx, double b[])
7051: {
7052: int i,ii=0,ip,j;
7053: double sum;
7054:
7055: for (i=1;i<=n;i++) {
7056: ip=indx[i];
7057: sum=b[ip];
7058: b[ip]=b[i];
7059: if (ii)
7060: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
7061: else if (sum) ii=i;
7062: b[i]=sum;
7063: }
7064: for (i=n;i>=1;i--) {
7065: sum=b[i];
7066: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
7067: b[i]=sum/a[i][i];
7068: }
7069: }
7070:
7071: void pstamp(FILE *fichier)
7072: {
1.196 brouard 7073: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 7074: }
7075:
1.297 brouard 7076: void date2dmy(double date,double *day, double *month, double *year){
7077: double yp=0., yp1=0., yp2=0.;
7078:
7079: yp1=modf(date,&yp);/* extracts integral of date in yp and
7080: fractional in yp1 */
7081: *year=yp;
7082: yp2=modf((yp1*12),&yp);
7083: *month=yp;
7084: yp1=modf((yp2*30.5),&yp);
7085: *day=yp;
7086: if(*day==0) *day=1;
7087: if(*month==0) *month=1;
7088: }
7089:
1.253 brouard 7090:
7091:
1.126 brouard 7092: /************ Frequencies ********************/
1.251 brouard 7093: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 7094: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
7095: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 7096: { /* Some frequencies as well as proposing some starting values */
1.332 brouard 7097: /* Frequencies of any combination of dummy covariate used in the model equation */
1.265 brouard 7098: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 7099: int iind=0, iage=0;
7100: int mi; /* Effective wave */
7101: int first;
7102: double ***freq; /* Frequencies */
1.268 brouard 7103: double *x, *y, a=0.,b=0.,r=1., sa=0., sb=0.; /* for regression, y=b+m*x and r is the correlation coefficient */
7104: int no=0, linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb);
1.284 brouard 7105: double *meanq, *stdq, *idq;
1.226 brouard 7106: double **meanqt;
7107: double *pp, **prop, *posprop, *pospropt;
7108: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
7109: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
7110: double agebegin, ageend;
7111:
7112: pp=vector(1,nlstate);
1.251 brouard 7113: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 7114: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
7115: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
7116: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
7117: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 7118: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 7119: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 7120: meanqt=matrix(1,lastpass,1,nqtveff);
7121: strcpy(fileresp,"P_");
7122: strcat(fileresp,fileresu);
7123: /*strcat(fileresphtm,fileresu);*/
7124: if((ficresp=fopen(fileresp,"w"))==NULL) {
7125: printf("Problem with prevalence resultfile: %s\n", fileresp);
7126: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
7127: exit(0);
7128: }
1.240 brouard 7129:
1.226 brouard 7130: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
7131: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
7132: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
7133: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
7134: fflush(ficlog);
7135: exit(70);
7136: }
7137: else{
7138: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 7139: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 7140: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 7141: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
7142: }
1.319 brouard 7143: fprintf(ficresphtm,"Current page is file <a href=\"%s\">%s</a><br>\n\n<h4>Frequencies (weight=%d) and prevalence by age at begin of transition and dummy covariate value at beginning of transition</h4>\n",fileresphtm, fileresphtm, weightopt);
1.240 brouard 7144:
1.226 brouard 7145: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
7146: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
7147: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
7148: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
7149: fflush(ficlog);
7150: exit(70);
1.240 brouard 7151: } else{
1.226 brouard 7152: fprintf(ficresphtmfr,"<html><head>\n<title>IMaCh PHTM_Frequency table %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.319 brouard 7153: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 7154: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 7155: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
7156: }
1.319 brouard 7157: fprintf(ficresphtmfr,"Current page is file <a href=\"%s\">%s</a><br>\n\n<h4>(weight=%d) frequencies of all effective transitions of the model, by age at begin of transition, and covariate value at the begin of transition (if the covariate is a varying covariate) </h4>Unknown status is -1<br/>\n",fileresphtmfr, fileresphtmfr,weightopt);
1.240 brouard 7158:
1.253 brouard 7159: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
7160: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 7161: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 7162: j1=0;
1.126 brouard 7163:
1.227 brouard 7164: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
1.335 brouard 7165: j=cptcoveff; /* Only simple dummy covariates used in the model */
1.330 brouard 7166: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 7167: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 7168:
7169:
1.226 brouard 7170: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
7171: reference=low_education V1=0,V2=0
7172: med_educ V1=1 V2=0,
7173: high_educ V1=0 V2=1
1.330 brouard 7174: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 7175: */
1.249 brouard 7176: dateintsum=0;
7177: k2cpt=0;
7178:
1.253 brouard 7179: if(cptcoveff == 0 )
1.265 brouard 7180: nl=1; /* Constant and age model only */
1.253 brouard 7181: else
7182: nl=2;
1.265 brouard 7183:
7184: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
7185: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.335 brouard 7186: * Loop on j1(1 to 2**cptcoveff) covariate combination
1.265 brouard 7187: * freq[s1][s2][iage] =0.
7188: * Loop on iind
7189: * ++freq[s1][s2][iage] weighted
7190: * end iind
7191: * if covariate and j!0
7192: * headers Variable on one line
7193: * endif cov j!=0
7194: * header of frequency table by age
7195: * Loop on age
7196: * pp[s1]+=freq[s1][s2][iage] weighted
7197: * pos+=freq[s1][s2][iage] weighted
7198: * Loop on s1 initial state
7199: * fprintf(ficresp
7200: * end s1
7201: * end age
7202: * if j!=0 computes starting values
7203: * end compute starting values
7204: * end j1
7205: * end nl
7206: */
1.253 brouard 7207: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
7208: if(nj==1)
7209: j=0; /* First pass for the constant */
1.265 brouard 7210: else{
1.335 brouard 7211: j=cptcoveff; /* Other passes for the covariate values number of simple covariates in the model V2+V1 =2 (simple dummy fixed or time varying) */
1.265 brouard 7212: }
1.251 brouard 7213: first=1;
1.332 brouard 7214: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on all dummy covariates combination of the model, ie excluding quantitatives, V4=0, V3=0 for example, fixed or varying covariates */
1.251 brouard 7215: posproptt=0.;
1.330 brouard 7216: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 7217: scanf("%d", i);*/
7218: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 7219: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 7220: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 7221: freq[i][s2][m]=0;
1.251 brouard 7222:
7223: for (i=1; i<=nlstate; i++) {
1.240 brouard 7224: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 7225: prop[i][m]=0;
7226: posprop[i]=0;
7227: pospropt[i]=0;
7228: }
1.283 brouard 7229: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 7230: idq[z1]=0.;
7231: meanq[z1]=0.;
7232: stdq[z1]=0.;
1.283 brouard 7233: }
7234: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 7235: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 7236: /* meanqt[m][z1]=0.; */
7237: /* } */
7238: /* } */
1.251 brouard 7239: /* dateintsum=0; */
7240: /* k2cpt=0; */
7241:
1.265 brouard 7242: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 7243: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
7244: bool=1;
7245: if(j !=0){
7246: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.335 brouard 7247: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
7248: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
1.251 brouard 7249: /* if(Tvaraff[z1] ==-20){ */
7250: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
7251: /* }else if(Tvaraff[z1] ==-10){ */
7252: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 brouard 7253: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
1.335 brouard 7254: /* if( iind >=imx-3) printf("Searching error iind=%d Tvaraff[z1]=%d covar[Tvaraff[z1]][iind]=%.f TnsdVar[Tvaraff[z1]]=%d, cptcoveff=%d, cptcovs=%d \n",iind, Tvaraff[z1], covar[Tvaraff[z1]][iind],TnsdVar[Tvaraff[z1]],cptcoveff, cptcovs); */
7255: if(Tvaraff[z1]<1 || Tvaraff[z1]>=NCOVMAX)
1.338 brouard 7256: printf("Error Tvaraff[z1]=%d<1 or >=%d, cptcoveff=%d model=1+age+%s\n",Tvaraff[z1],NCOVMAX, cptcoveff, model);
1.332 brouard 7257: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]){ /* for combination j1 of covariates */
1.265 brouard 7258: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 7259: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.332 brouard 7260: /* printf("bool=%d i=%d, z1=%d, Tvaraff[%d]=%d, covar[Tvarff][%d]=%2f, codtabm(%d,%d)=%d, nbcode[Tvaraff][codtabm(%d,%d)=%d, j1=%d\n", */
7261: /* bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),*/
7262: /* j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
1.251 brouard 7263: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
7264: } /* Onlyf fixed */
7265: } /* end z1 */
1.335 brouard 7266: } /* cptcoveff > 0 */
1.251 brouard 7267: } /* end any */
7268: }/* end j==0 */
1.265 brouard 7269: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 7270: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 7271: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 7272: m=mw[mi][iind];
7273: if(j!=0){
7274: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.335 brouard 7275: for (z1=1; z1<=cptcoveff; z1++) {
1.251 brouard 7276: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 7277: /* iv= Tvar[Tmodelind[z1]]-ncovcol-nqv; /\* Good *\/ */
7278: iv= Tvar[Tmodelind[z1]]; /* Good *//* because cotvar starts now at first at ncovcol+nqv+ntv */
1.332 brouard 7279: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality. If covariate's
1.251 brouard 7280: value is -1, we don't select. It differs from the
7281: constant and age model which counts them. */
7282: bool=0; /* not selected */
7283: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.334 brouard 7284: /* i1=Tvaraff[z1]; */
7285: /* i2=TnsdVar[i1]; */
7286: /* i3=nbcode[i1][i2]; */
7287: /* i4=covar[i1][iind]; */
7288: /* if(i4 != i3){ */
7289: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) { /* Bug valgrind */
1.251 brouard 7290: bool=0;
7291: }
7292: }
7293: }
7294: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
7295: } /* end j==0 */
7296: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 7297: if(bool==1){ /*Selected */
1.251 brouard 7298: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
7299: and mw[mi+1][iind]. dh depends on stepm. */
7300: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
7301: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
7302: if(m >=firstpass && m <=lastpass){
7303: k2=anint[m][iind]+(mint[m][iind]/12.);
7304: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
7305: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
7306: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
7307: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
7308: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
7309: if (m<lastpass) {
7310: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
7311: /* printf(" num=%ld m=%d, iind=%d s1=%d s2=%d agev at m=%d\n", num[iind], m, iind,s[m][iind],s[m+1][iind], (int)agev[m][iind]); */
7312: if(s[m][iind]==-1)
7313: printf(" num=%ld m=%d, iind=%d s1=%d s2=%d agev at m=%d agebegin=%.2f ageend=%.2f, agemed=%d\n", num[iind], m, iind,s[m][iind],s[m+1][iind], (int)agev[m][iind],agebegin, ageend, (int)((agebegin+ageend)/2.));
7314: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
1.311 brouard 7315: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
7316: if(!isnan(covar[ncovcol+z1][iind])){
1.332 brouard 7317: idq[z1]=idq[z1]+weight[iind];
7318: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
7319: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
7320: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
1.311 brouard 7321: }
1.284 brouard 7322: }
1.251 brouard 7323: /* if((int)agev[m][iind] == 55) */
7324: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
7325: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
7326: freq[s[m][iind]][s[m+1][iind]][iagemax+3] += weight[iind]; /* Total is in iagemax+3 *//* At age of beginning of transition, where status is known */
1.234 brouard 7327: }
1.251 brouard 7328: } /* end if between passes */
7329: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
7330: dateintsum=dateintsum+k2; /* on all covariates ?*/
7331: k2cpt++;
7332: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 7333: }
1.251 brouard 7334: }else{
7335: bool=1;
7336: }/* end bool 2 */
7337: } /* end m */
1.284 brouard 7338: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
7339: /* idq[z1]=idq[z1]+weight[iind]; */
7340: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
7341: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
7342: /* } */
1.251 brouard 7343: } /* end bool */
7344: } /* end iind = 1 to imx */
1.319 brouard 7345: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 7346: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
7347:
7348:
7349: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.335 brouard 7350: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
1.265 brouard 7351: pstamp(ficresp);
1.335 brouard 7352: if (cptcoveff>0 && j!=0){
1.265 brouard 7353: pstamp(ficresp);
1.251 brouard 7354: printf( "\n#********** Variable ");
7355: fprintf(ficresp, "\n#********** Variable ");
7356: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
7357: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
7358: fprintf(ficlog, "\n#********** Variable ");
1.340 brouard 7359: for (z1=1; z1<=cptcoveff; z1++){
1.251 brouard 7360: if(!FixedV[Tvaraff[z1]]){
1.330 brouard 7361: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7362: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7363: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7364: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7365: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 7366: }else{
1.330 brouard 7367: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7368: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7369: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7370: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7371: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 7372: }
7373: }
7374: printf( "**********\n#");
7375: fprintf(ficresp, "**********\n#");
7376: fprintf(ficresphtm, "**********</h3>\n");
7377: fprintf(ficresphtmfr, "**********</h3>\n");
7378: fprintf(ficlog, "**********\n");
7379: }
1.284 brouard 7380: /*
7381: Printing means of quantitative variables if any
7382: */
7383: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 7384: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 7385: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 7386: if(weightopt==1){
7387: printf(" Weighted mean and standard deviation of");
7388: fprintf(ficlog," Weighted mean and standard deviation of");
7389: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
7390: }
1.311 brouard 7391: /* mu = \frac{w x}{\sum w}
7392: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
7393: */
7394: printf(" fixed quantitative variable V%d on %.3g (weighted) representatives of the population : %8.5g (%8.5g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt(stdq[z1]/idq[z1]-meanq[z1]*meanq[z1]/idq[z1]/idq[z1]));
7395: fprintf(ficlog," fixed quantitative variable V%d on %.3g (weighted) representatives of the population : %8.5g (%8.5g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt(stdq[z1]/idq[z1]-meanq[z1]*meanq[z1]/idq[z1]/idq[z1]));
7396: fprintf(ficresphtmfr," fixed quantitative variable V%d on %.3g (weighted) representatives of the population : %8.5g (%8.5g)<p>\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt(stdq[z1]/idq[z1]-meanq[z1]*meanq[z1]/idq[z1]/idq[z1]));
1.284 brouard 7397: }
7398: /* for (z1=1; z1<= nqtveff; z1++) { */
7399: /* for(m=1;m<=lastpass;m++){ */
7400: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
7401: /* } */
7402: /* } */
1.283 brouard 7403:
1.251 brouard 7404: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.335 brouard 7405: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 7406: fprintf(ficresp, " Age");
1.335 brouard 7407: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) {
7408: printf(" V%d=%d, z1=%d, Tvaraff[z1]=%d, j1=%d, TnsdVar[Tvaraff[%d]]=%d |",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])], z1, Tvaraff[z1], j1,z1,TnsdVar[Tvaraff[z1]]);
7409: fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7410: }
1.251 brouard 7411: for(i=1; i<=nlstate;i++) {
1.335 brouard 7412: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 7413: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
7414: }
1.335 brouard 7415: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 7416: fprintf(ficresphtm, "\n");
7417:
7418: /* Header of frequency table by age */
7419: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
7420: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 7421: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 7422: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 7423: if(s2!=0 && m!=0)
7424: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 7425: }
1.226 brouard 7426: }
1.251 brouard 7427: fprintf(ficresphtmfr, "\n");
7428:
7429: /* For each age */
7430: for(iage=iagemin; iage <= iagemax+3; iage++){
7431: fprintf(ficresphtm,"<tr>");
7432: if(iage==iagemax+1){
7433: fprintf(ficlog,"1");
7434: fprintf(ficresphtmfr,"<tr><th>0</th> ");
7435: }else if(iage==iagemax+2){
7436: fprintf(ficlog,"0");
7437: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
7438: }else if(iage==iagemax+3){
7439: fprintf(ficlog,"Total");
7440: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
7441: }else{
1.240 brouard 7442: if(first==1){
1.251 brouard 7443: first=0;
7444: printf("See log file for details...\n");
7445: }
7446: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
7447: fprintf(ficlog,"Age %d", iage);
7448: }
1.265 brouard 7449: for(s1=1; s1 <=nlstate ; s1++){
7450: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
7451: pp[s1] += freq[s1][m][iage];
1.251 brouard 7452: }
1.265 brouard 7453: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 7454: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 7455: pos += freq[s1][m][iage];
7456: if(pp[s1]>=1.e-10){
1.251 brouard 7457: if(first==1){
1.265 brouard 7458: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 7459: }
1.265 brouard 7460: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 7461: }else{
7462: if(first==1)
1.265 brouard 7463: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
7464: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 7465: }
7466: }
7467:
1.265 brouard 7468: for(s1=1; s1 <=nlstate ; s1++){
7469: /* posprop[s1]=0; */
7470: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
7471: pp[s1] += freq[s1][m][iage];
7472: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
7473:
7474: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
7475: pos += pp[s1]; /* pos is the total number of transitions until this age */
7476: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
7477: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
7478: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
7479: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
7480: }
7481:
7482: /* Writing ficresp */
1.335 brouard 7483: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 7484: if( iage <= iagemax){
7485: fprintf(ficresp," %d",iage);
7486: }
7487: }else if( nj==2){
7488: if( iage <= iagemax){
7489: fprintf(ficresp," %d",iage);
1.335 brouard 7490: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.265 brouard 7491: }
1.240 brouard 7492: }
1.265 brouard 7493: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 7494: if(pos>=1.e-5){
1.251 brouard 7495: if(first==1)
1.265 brouard 7496: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
7497: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 7498: }else{
7499: if(first==1)
1.265 brouard 7500: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
7501: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 7502: }
7503: if( iage <= iagemax){
7504: if(pos>=1.e-5){
1.335 brouard 7505: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 7506: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
7507: }else if( nj==2){
7508: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
7509: }
7510: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
7511: /*probs[iage][s1][j1]= pp[s1]/pos;*/
7512: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
7513: } else{
1.335 brouard 7514: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 7515: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 7516: }
1.240 brouard 7517: }
1.265 brouard 7518: pospropt[s1] +=posprop[s1];
7519: } /* end loop s1 */
1.251 brouard 7520: /* pospropt=0.; */
1.265 brouard 7521: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 7522: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 7523: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 7524: if(first==1){
1.265 brouard 7525: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 7526: }
1.265 brouard 7527: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
7528: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 7529: }
1.265 brouard 7530: if(s1!=0 && m!=0)
7531: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 7532: }
1.265 brouard 7533: } /* end loop s1 */
1.251 brouard 7534: posproptt=0.;
1.265 brouard 7535: for(s1=1; s1 <=nlstate; s1++){
7536: posproptt += pospropt[s1];
1.251 brouard 7537: }
7538: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 7539: fprintf(ficresphtm,"</tr>\n");
1.335 brouard 7540: if((cptcoveff==0 && nj==1)|| nj==2 ) {
1.265 brouard 7541: if(iage <= iagemax)
7542: fprintf(ficresp,"\n");
1.240 brouard 7543: }
1.251 brouard 7544: if(first==1)
7545: printf("Others in log...\n");
7546: fprintf(ficlog,"\n");
7547: } /* end loop age iage */
1.265 brouard 7548:
1.251 brouard 7549: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 7550: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 7551: if(posproptt < 1.e-5){
1.265 brouard 7552: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 7553: }else{
1.265 brouard 7554: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 7555: }
1.226 brouard 7556: }
1.251 brouard 7557: fprintf(ficresphtm,"</tr>\n");
7558: fprintf(ficresphtm,"</table>\n");
7559: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 7560: if(posproptt < 1.e-5){
1.251 brouard 7561: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
7562: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 7563: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
7564: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 7565: invalidvarcomb[j1]=1;
1.226 brouard 7566: }else{
1.338 brouard 7567: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced (or no resultline).</p>",j1);
1.251 brouard 7568: invalidvarcomb[j1]=0;
1.226 brouard 7569: }
1.251 brouard 7570: fprintf(ficresphtmfr,"</table>\n");
7571: fprintf(ficlog,"\n");
7572: if(j!=0){
7573: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 7574: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 7575: for(k=1; k <=(nlstate+ndeath); k++){
7576: if (k != i) {
1.265 brouard 7577: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 7578: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 7579: if(j1==1){ /* All dummy covariates to zero */
7580: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
7581: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 7582: printf("%d%d ",i,k);
7583: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 7584: printf("%12.7f ln(%.0f/%.0f)= %f, OR=%f sd=%f \n",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]),freq[i][k][iagemax+3]/freq[i][i][iagemax+3], sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]));
7585: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f \n",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
7586: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 7587: }
1.253 brouard 7588: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
7589: for(iage=iagemin; iage <= iagemax+3; iage++){
7590: x[iage]= (double)iage;
7591: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 7592: /* printf("i=%d, k=%d, s1=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,s1,j1,jj, iage, y[iage]); */
1.253 brouard 7593: }
1.268 brouard 7594: /* Some are not finite, but linreg will ignore these ages */
7595: no=0;
1.253 brouard 7596: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 7597: pstart[s1]=b;
7598: pstart[s1-1]=a;
1.252 brouard 7599: }else if( j1!=1 && (j1==2 || (log(j1-1.)/log(2.)-(int)(log(j1-1.)/log(2.))) <0.010) && ( TvarsDind[(int)(log(j1-1.)/log(2.))+1]+2+nagesqr == jj) && Dummy[jj-2-nagesqr]==0){ /* We want only if the position, jj, in model corresponds to unique covariate equal to 1 in j1 combination */
7600: printf("j1=%d, jj=%d, (int)(log(j1-1.)/log(2.))+1=%d, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(int)(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
7601: printf("j1=%d, jj=%d, (log(j1-1.)/log(2.))+1=%f, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
1.265 brouard 7602: pstart[s1]= log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]));
1.252 brouard 7603: printf("%d%d ",i,k);
7604: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 7605: printf("s1=%d,i=%d,k=%d,p[%d]=%12.7f ln((%.0f/%.0f)/(%.0f/%.0f))= %f, OR=%f sd=%f \n",s1,i,k,s1,p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3],freq[i][k][iagemax+4],freq[i][i][iagemax+4], log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4])),(freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]), sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]+1/freq[i][k][iagemax+4]+1/freq[i][i][iagemax+4]));
1.251 brouard 7606: }else{ /* Other cases, like quantitative fixed or varying covariates */
7607: ;
7608: }
7609: /* printf("%12.7f )", param[i][jj][k]); */
7610: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 7611: s1++;
1.251 brouard 7612: } /* end jj */
7613: } /* end k!= i */
7614: } /* end k */
1.265 brouard 7615: } /* end i, s1 */
1.251 brouard 7616: } /* end j !=0 */
7617: } /* end selected combination of covariate j1 */
7618: if(j==0){ /* We can estimate starting values from the occurences in each case */
7619: printf("#Freqsummary: Starting values for the constants:\n");
7620: fprintf(ficlog,"\n");
1.265 brouard 7621: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 7622: for(k=1; k <=(nlstate+ndeath); k++){
7623: if (k != i) {
7624: printf("%d%d ",i,k);
7625: fprintf(ficlog,"%d%d ",i,k);
7626: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 7627: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 7628: if(jj==1){ /* Age has to be done */
1.265 brouard 7629: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
7630: printf("%12.7f ln(%.0f/%.0f)= %12.7f ",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
7631: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f ",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
1.251 brouard 7632: }
7633: /* printf("%12.7f )", param[i][jj][k]); */
7634: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 7635: s1++;
1.250 brouard 7636: }
1.251 brouard 7637: printf("\n");
7638: fprintf(ficlog,"\n");
1.250 brouard 7639: }
7640: }
1.284 brouard 7641: } /* end of state i */
1.251 brouard 7642: printf("#Freqsummary\n");
7643: fprintf(ficlog,"\n");
1.265 brouard 7644: for(s1=-1; s1 <=nlstate+ndeath; s1++){
7645: for(s2=-1; s2 <=nlstate+ndeath; s2++){
7646: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
7647: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
7648: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
7649: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
7650: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
7651: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 7652: /* } */
7653: }
1.265 brouard 7654: } /* end loop s1 */
1.251 brouard 7655:
7656: printf("\n");
7657: fprintf(ficlog,"\n");
7658: } /* end j=0 */
1.249 brouard 7659: } /* end j */
1.252 brouard 7660:
1.253 brouard 7661: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 7662: for(i=1, jk=1; i <=nlstate; i++){
7663: for(j=1; j <=nlstate+ndeath; j++){
7664: if(j!=i){
7665: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7666: printf("%1d%1d",i,j);
7667: fprintf(ficparo,"%1d%1d",i,j);
7668: for(k=1; k<=ncovmodel;k++){
7669: /* printf(" %lf",param[i][j][k]); */
7670: /* fprintf(ficparo," %lf",param[i][j][k]); */
7671: p[jk]=pstart[jk];
7672: printf(" %f ",pstart[jk]);
7673: fprintf(ficparo," %f ",pstart[jk]);
7674: jk++;
7675: }
7676: printf("\n");
7677: fprintf(ficparo,"\n");
7678: }
7679: }
7680: }
7681: } /* end mle=-2 */
1.226 brouard 7682: dateintmean=dateintsum/k2cpt;
1.296 brouard 7683: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 7684:
1.226 brouard 7685: fclose(ficresp);
7686: fclose(ficresphtm);
7687: fclose(ficresphtmfr);
1.283 brouard 7688: free_vector(idq,1,nqfveff);
1.226 brouard 7689: free_vector(meanq,1,nqfveff);
1.284 brouard 7690: free_vector(stdq,1,nqfveff);
1.226 brouard 7691: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 7692: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
7693: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 7694: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 7695: free_vector(pospropt,1,nlstate);
7696: free_vector(posprop,1,nlstate);
1.251 brouard 7697: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 7698: free_vector(pp,1,nlstate);
7699: /* End of freqsummary */
7700: }
1.126 brouard 7701:
1.268 brouard 7702: /* Simple linear regression */
7703: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
7704:
7705: /* y=a+bx regression */
7706: double sumx = 0.0; /* sum of x */
7707: double sumx2 = 0.0; /* sum of x**2 */
7708: double sumxy = 0.0; /* sum of x * y */
7709: double sumy = 0.0; /* sum of y */
7710: double sumy2 = 0.0; /* sum of y**2 */
7711: double sume2 = 0.0; /* sum of square or residuals */
7712: double yhat;
7713:
7714: double denom=0;
7715: int i;
7716: int ne=*no;
7717:
7718: for ( i=ifi, ne=0;i<=ila;i++) {
7719: if(!isfinite(x[i]) || !isfinite(y[i])){
7720: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
7721: continue;
7722: }
7723: ne=ne+1;
7724: sumx += x[i];
7725: sumx2 += x[i]*x[i];
7726: sumxy += x[i] * y[i];
7727: sumy += y[i];
7728: sumy2 += y[i]*y[i];
7729: denom = (ne * sumx2 - sumx*sumx);
7730: /* printf("ne=%d, i=%d,x[%d]=%f, y[%d]=%f sumx=%f, sumx2=%f, sumxy=%f, sumy=%f, sumy2=%f, denom=%f\n",ne,i,i,x[i],i,y[i], sumx, sumx2,sumxy, sumy, sumy2,denom); */
7731: }
7732:
7733: denom = (ne * sumx2 - sumx*sumx);
7734: if (denom == 0) {
7735: // vertical, slope m is infinity
7736: *b = INFINITY;
7737: *a = 0;
7738: if (r) *r = 0;
7739: return 1;
7740: }
7741:
7742: *b = (ne * sumxy - sumx * sumy) / denom;
7743: *a = (sumy * sumx2 - sumx * sumxy) / denom;
7744: if (r!=NULL) {
7745: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
7746: sqrt((sumx2 - sumx*sumx/ne) *
7747: (sumy2 - sumy*sumy/ne));
7748: }
7749: *no=ne;
7750: for ( i=ifi, ne=0;i<=ila;i++) {
7751: if(!isfinite(x[i]) || !isfinite(y[i])){
7752: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
7753: continue;
7754: }
7755: ne=ne+1;
7756: yhat = y[i] - *a -*b* x[i];
7757: sume2 += yhat * yhat ;
7758:
7759: denom = (ne * sumx2 - sumx*sumx);
7760: /* printf("ne=%d, i=%d,x[%d]=%f, y[%d]=%f sumx=%f, sumx2=%f, sumxy=%f, sumy=%f, sumy2=%f, denom=%f\n",ne,i,i,x[i],i,y[i], sumx, sumx2,sumxy, sumy, sumy2,denom); */
7761: }
7762: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
7763: *sa= *sb * sqrt(sumx2/ne);
7764:
7765: return 0;
7766: }
7767:
1.126 brouard 7768: /************ Prevalence ********************/
1.227 brouard 7769: void prevalence(double ***probs, double agemin, double agemax, int **s, double **agev, int nlstate, int imx, int *Tvar, int **nbcode, int *ncodemax,double **mint,double **anint, double dateprev1,double dateprev2, int firstpass, int lastpass)
7770: {
7771: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7772: in each health status at the date of interview (if between dateprev1 and dateprev2).
7773: We still use firstpass and lastpass as another selection.
7774: */
1.126 brouard 7775:
1.227 brouard 7776: int i, m, jk, j1, bool, z1,j, iv;
7777: int mi; /* Effective wave */
7778: int iage;
1.359 brouard 7779: double agebegin; /*, ageend;*/
1.227 brouard 7780:
7781: double **prop;
7782: double posprop;
7783: double y2; /* in fractional years */
7784: int iagemin, iagemax;
7785: int first; /** to stop verbosity which is redirected to log file */
7786:
7787: iagemin= (int) agemin;
7788: iagemax= (int) agemax;
7789: /*pp=vector(1,nlstate);*/
1.251 brouard 7790: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 7791: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
7792: j1=0;
1.222 brouard 7793:
1.227 brouard 7794: /*j=cptcoveff;*/
7795: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 7796:
1.288 brouard 7797: first=0;
1.335 brouard 7798: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of simple dummy covariates */
1.227 brouard 7799: for (i=1; i<=nlstate; i++)
1.251 brouard 7800: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 7801: prop[i][iage]=0.0;
7802: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
7803: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
7804: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
7805:
7806: for (i=1; i<=imx; i++) { /* Each individual */
7807: bool=1;
7808: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
7809: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
7810: m=mw[mi][i];
7811: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
7812: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
7813: for (z1=1; z1<=cptcoveff; z1++){
7814: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 7815: iv= Tvar[Tmodelind[z1]];/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.332 brouard 7816: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality */
1.227 brouard 7817: bool=0;
7818: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.332 brouard 7819: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.227 brouard 7820: bool=0;
7821: }
7822: }
7823: if(bool==1){ /* Otherwise we skip that wave/person */
7824: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
7825: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
7826: if(m >=firstpass && m <=lastpass){
7827: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
7828: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
7829: if(agev[m][i]==0) agev[m][i]=iagemax+1;
7830: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 7831: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 7832: printf("Error on individual # %d agev[m][i]=%f <%d-%d or > %d+3+%d m=%d; either change agemin or agemax or fix data\n",i, agev[m][i],iagemin,AGEMARGE, iagemax,AGEMARGE,m);
7833: exit(1);
7834: }
7835: if (s[m][i]>0 && s[m][i]<=nlstate) {
7836: /*if(i>4620) printf(" i=%d m=%d s[m][i]=%d (int)agev[m][i]=%d weight[i]=%f prop=%f\n",i,m,s[m][i],(int)agev[m][m],weight[i],prop[s[m][i]][(int)agev[m][i]]);*/
7837: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
7838: prop[s[m][i]][iagemax+3] += weight[i];
7839: } /* end valid statuses */
7840: } /* end selection of dates */
7841: } /* end selection of waves */
7842: } /* end bool */
7843: } /* end wave */
7844: } /* end individual */
7845: for(i=iagemin; i <= iagemax+3; i++){
7846: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
7847: posprop += prop[jk][i];
7848: }
7849:
7850: for(jk=1; jk <=nlstate ; jk++){
7851: if( i <= iagemax){
7852: if(posprop>=1.e-5){
7853: probs[i][jk][j1]= prop[jk][i]/posprop;
7854: } else{
1.288 brouard 7855: if(!first){
7856: first=1;
1.266 brouard 7857: printf("Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,jk, j1,probs[i][jk][j1]);
7858: }else{
1.288 brouard 7859: fprintf(ficlog,"Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases.\n",jk,i,jk, j1,probs[i][jk][j1]);
1.227 brouard 7860: }
7861: }
7862: }
7863: }/* end jk */
7864: }/* end i */
1.222 brouard 7865: /*} *//* end i1 */
1.227 brouard 7866: } /* end j1 */
1.222 brouard 7867:
1.227 brouard 7868: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
7869: /*free_vector(pp,1,nlstate);*/
1.251 brouard 7870: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 7871: } /* End of prevalence */
1.126 brouard 7872:
7873: /************* Waves Concatenation ***************/
7874:
7875: void concatwav(int wav[], int **dh, int **bh, int **mw, int **s, double *agedc, double **agev, int firstpass, int lastpass, int imx, int nlstate, int stepm)
7876: {
1.298 brouard 7877: /* Concatenates waves: wav[i] is the number of effective (useful waves in the sense that a non interview is useless) of individual i.
1.126 brouard 7878: Death is a valid wave (if date is known).
7879: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
7880: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 7881: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 7882: */
1.126 brouard 7883:
1.224 brouard 7884: int i=0, mi=0, m=0, mli=0;
1.126 brouard 7885: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
7886: double sum=0., jmean=0.;*/
1.224 brouard 7887: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 7888: int j, k=0,jk, ju, jl;
7889: double sum=0.;
7890: first=0;
1.214 brouard 7891: firstwo=0;
1.217 brouard 7892: firsthree=0;
1.218 brouard 7893: firstfour=0;
1.164 brouard 7894: jmin=100000;
1.126 brouard 7895: jmax=-1;
7896: jmean=0.;
1.224 brouard 7897:
7898: /* Treating live states */
1.214 brouard 7899: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 7900: mi=0; /* First valid wave */
1.227 brouard 7901: mli=0; /* Last valid wave */
1.309 brouard 7902: m=firstpass; /* Loop on waves */
7903: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 7904: if(m >firstpass && s[m][i]==s[m-1][i] && mint[m][i]==mint[m-1][i] && anint[m][i]==anint[m-1][i]){/* Two succesive identical information on wave m */
7905: mli=m-1;/* mw[++mi][i]=m-1; */
7906: }else if(s[m][i]>=1 || s[m][i]==-4 || s[m][i]==-5){ /* Since 0.98r4 if status=-2 vital status is really unknown, wave should be skipped */
1.309 brouard 7907: mw[++mi][i]=m; /* Valid wave: incrementing mi and updating mi; mw[mi] is the wave number of mi_th valid transition */
1.227 brouard 7908: mli=m;
1.224 brouard 7909: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
7910: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 7911: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 7912: }
1.309 brouard 7913: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 7914: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 7915: break;
1.224 brouard 7916: #else
1.317 brouard 7917: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){ /* no death date and known date of interview, case -2 (vital status unknown is warned later */
1.227 brouard 7918: if(firsthree == 0){
1.302 brouard 7919: printf("Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as 1-p_{%d%d} .\nOthers in log file only\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m, s[m][i], nlstate+ndeath);
1.227 brouard 7920: firsthree=1;
1.317 brouard 7921: }else if(firsthree >=1 && firsthree < 10){
7922: fprintf(ficlog,"Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as 1-p_{%d%d} .\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m, s[m][i], nlstate+ndeath);
7923: firsthree++;
7924: }else if(firsthree == 10){
7925: printf("Information, too many Information flags: no more reported to log either\n");
7926: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
7927: firsthree++;
7928: }else{
7929: firsthree++;
1.227 brouard 7930: }
1.309 brouard 7931: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 7932: mli=m;
7933: }
7934: if(s[m][i]==-2){ /* Vital status is really unknown */
7935: nbwarn++;
1.309 brouard 7936: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 7937: printf("Warning! Vital status for individual %ld (line=%d) at last wave %d interviewed at date %d/%d is unknown %d. Please, check if the vital status and the date of death %d/%d are really unknown. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\nOthers in log file only\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], (int) moisdc[i], (int) andc[i], i, m);
7938: fprintf(ficlog,"Warning! Vital status for individual %ld (line=%d) at last wave %d interviewed at date %d/%d is unknown %d. Please, check if the vital status and the date of death %d/%d are really unknown. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], (int) moisdc[i], (int) andc[i], i, m);
7939: }
7940: break;
7941: }
7942: break;
1.224 brouard 7943: #endif
1.227 brouard 7944: }/* End m >= lastpass */
1.126 brouard 7945: }/* end while */
1.224 brouard 7946:
1.227 brouard 7947: /* mi is the last effective wave, m is lastpass, mw[j][i] gives the # of j-th effective wave for individual i */
1.216 brouard 7948: /* After last pass */
1.224 brouard 7949: /* Treating death states */
1.214 brouard 7950: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 7951: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
7952: /* } */
1.126 brouard 7953: mi++; /* Death is another wave */
7954: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 7955: /* Only death is a correct wave */
1.126 brouard 7956: mw[mi][i]=m;
1.257 brouard 7957: } /* else not in a death state */
1.224 brouard 7958: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 7959: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 7960: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 7961: if((andc[i]+moisdc[i]/12.) <=(anint[m][i]+mint[m][i]/12.)){ /* month of death occured before last wave month and status should have been death instead of -1 */
1.227 brouard 7962: nbwarn++;
7963: if(firstfiv==0){
1.309 brouard 7964: printf("Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d, interviewed on %d/%d and should have been coded as death instead of '%d'. This case (%d)/wave (%d) is contributing to likelihood.\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 7965: firstfiv=1;
7966: }else{
1.309 brouard 7967: fprintf(ficlog,"Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d, interviewed on %d/%d and should have been coded as death instead of '%d'. This case (%d)/wave (%d) is contributing to likelihood.\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 7968: }
1.309 brouard 7969: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
7970: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 7971: nberr++;
7972: if(firstwo==0){
1.309 brouard 7973: printf("Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d with status %d. Potential bias if other individuals are still alive on this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood. Please add a new fictitious wave at the date of last vital status scan, with a dead status. See documentation\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 7974: firstwo=1;
7975: }
1.309 brouard 7976: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d with status %d. Potential bias if other individuals are still alive on this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood. Please add a new fictitious wave at the date of last vital status scan, with a dead status. See documentation\n\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 7977: }
1.257 brouard 7978: }else{ /* if date of interview is unknown */
1.227 brouard 7979: /* death is known but not confirmed by death status at any wave */
7980: if(firstfour==0){
1.309 brouard 7981: printf("Error! Death for individual %ld line=%d occurred %d/%d but not confirmed by any death status for any wave, including last wave %d at unknown date %d/%d with status %d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 7982: firstfour=1;
7983: }
1.309 brouard 7984: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred %d/%d but not confirmed by any death status for any wave, including last wave %d at unknown date %d/%d with status %d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.214 brouard 7985: }
1.224 brouard 7986: } /* end if date of death is known */
7987: #endif
1.309 brouard 7988: wav[i]=mi; /* mi should be the last effective wave (or mli), */
7989: /* wav[i]=mw[mi][i]; */
1.126 brouard 7990: if(mi==0){
7991: nbwarn++;
7992: if(first==0){
1.227 brouard 7993: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
7994: first=1;
1.126 brouard 7995: }
7996: if(first==1){
1.227 brouard 7997: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 7998: }
7999: } /* end mi==0 */
8000: } /* End individuals */
1.214 brouard 8001: /* wav and mw are no more changed */
1.223 brouard 8002:
1.317 brouard 8003: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
8004: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
8005:
8006:
1.126 brouard 8007: for(i=1; i<=imx; i++){
8008: for(mi=1; mi<wav[i];mi++){
8009: if (stepm <=0)
1.227 brouard 8010: dh[mi][i]=1;
1.126 brouard 8011: else{
1.260 brouard 8012: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 8013: if (agedc[i] < 2*AGESUP) {
8014: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
8015: if(j==0) j=1; /* Survives at least one month after exam */
8016: else if(j<0){
8017: nberr++;
1.359 brouard 8018: printf("Error! Negative delay (%d to death) between waves %d and %d of individual %ld (around line %d) who is aged %.1f with statuses from %d to %d\n ",j,mw[mi][i],mw[mi+1][i],num[i], i,agev[mw[mi][i]][i],s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]);
1.227 brouard 8019: j=1; /* Temporary Dangerous patch */
8020: printf(" We assumed that the date of interview was correct (and not the date of death) and postponed the death %d month(s) (one stepm) after the interview. You MUST fix the contradiction between dates.\n",stepm);
1.359 brouard 8021: fprintf(ficlog,"Error! Negative delay (%d to death) between waves %d and %d of individual %ld (around line %d) who is aged %.1f with statuses from %d to %d\n ",j,mw[mi][i],mw[mi+1][i],num[i], i,agev[mw[mi][i]][i],s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]);
1.227 brouard 8022: fprintf(ficlog," We assumed that the date of interview was correct (and not the date of death) and postponed the death %d month(s) (one stepm) after the interview. You MUST fix the contradiction between dates.\n",stepm);
8023: }
8024: k=k+1;
8025: if (j >= jmax){
8026: jmax=j;
8027: ijmax=i;
8028: }
8029: if (j <= jmin){
8030: jmin=j;
8031: ijmin=i;
8032: }
8033: sum=sum+j;
8034: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
8035: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
8036: }
8037: }
8038: else{
8039: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 8040: /* if (j<0) printf("%d %lf %lf %d %d %d\n", i,agev[mw[mi+1][i]][i], agev[mw[mi][i]][i],j,s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]); */
1.223 brouard 8041:
1.227 brouard 8042: k=k+1;
8043: if (j >= jmax) {
8044: jmax=j;
8045: ijmax=i;
8046: }
8047: else if (j <= jmin){
8048: jmin=j;
8049: ijmin=i;
8050: }
8051: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
8052: /*printf("%d %lf %d %d %d\n", i,agev[mw[mi][i]][i],j,s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]);*/
8053: if(j<0){
8054: nberr++;
1.359 brouard 8055: printf("Error! Negative delay (%d) between waves %d and %d of individual %ld (around line %d) who is aged %.1f with statuses from %d to %d\n ",j,mw[mi][i],mw[mi+1][i],num[i], i,agev[mw[mi][i]][i],s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]);
8056: fprintf(ficlog,"Error! Negative delay (%d) between waves %d and %d of individual %ld (around line %d) who is aged %.1f with statuses from %d to %d\n ",j,mw[mi][i],mw[mi+1][i],num[i], i,agev[mw[mi][i]][i],s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]);
1.227 brouard 8057: }
8058: sum=sum+j;
8059: }
8060: jk= j/stepm;
8061: jl= j -jk*stepm;
8062: ju= j -(jk+1)*stepm;
8063: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
8064: if(jl==0){
8065: dh[mi][i]=jk;
8066: bh[mi][i]=0;
8067: }else{ /* We want a negative bias in order to only have interpolation ie
8068: * to avoid the price of an extra matrix product in likelihood */
8069: dh[mi][i]=jk+1;
8070: bh[mi][i]=ju;
8071: }
8072: }else{
8073: if(jl <= -ju){
8074: dh[mi][i]=jk;
8075: bh[mi][i]=jl; /* bias is positive if real duration
8076: * is higher than the multiple of stepm and negative otherwise.
8077: */
8078: }
8079: else{
8080: dh[mi][i]=jk+1;
8081: bh[mi][i]=ju;
8082: }
8083: if(dh[mi][i]==0){
8084: dh[mi][i]=1; /* At least one step */
8085: bh[mi][i]=ju; /* At least one step */
8086: /* printf(" bh=%d ju=%d jl=%d dh=%d jk=%d stepm=%d %d\n",bh[mi][i],ju,jl,dh[mi][i],jk,stepm,i);*/
8087: }
8088: } /* end if mle */
1.126 brouard 8089: }
8090: } /* end wave */
8091: }
8092: jmean=sum/k;
8093: printf("Delay (in months) between two waves Min=%d (for indiviudal %ld) Max=%d (%ld) Mean=%f\n\n ",jmin, num[ijmin], jmax, num[ijmax], jmean);
1.141 brouard 8094: fprintf(ficlog,"Delay (in months) between two waves Min=%d (for indiviudal %d) Max=%d (%d) Mean=%f\n\n ",jmin, ijmin, jmax, ijmax, jmean);
1.227 brouard 8095: }
1.126 brouard 8096:
8097: /*********** Tricode ****************************/
1.220 brouard 8098: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 8099: {
8100: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
8101: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
8102: * Boring subroutine which should only output nbcode[Tvar[j]][k]
8103: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
8104: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
8105: */
1.130 brouard 8106:
1.242 brouard 8107: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
8108: int modmaxcovj=0; /* Modality max of covariates j */
8109: int cptcode=0; /* Modality max of covariates j */
8110: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 8111:
8112:
1.242 brouard 8113: /* cptcoveff=0; */
8114: /* *cptcov=0; */
1.126 brouard 8115:
1.242 brouard 8116: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 8117: for (k=1; k <= maxncov; k++)
8118: for(j=1; j<=2; j++)
8119: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 8120:
1.242 brouard 8121: /* Loop on covariates without age and products and no quantitative variable */
1.335 brouard 8122: for (k=1; k<=cptcovt; k++) { /* cptcovt: total number of covariates of the model (2) nbocc(+)+1 = 8 excepting constant and age and age*age */
1.242 brouard 8123: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
1.343 brouard 8124: /* printf("Testing k=%d, cptcovt=%d\n",k, cptcovt); */
1.349 brouard 8125: if(Dummy[k]==0 && Typevar[k] !=1 && Typevar[k] != 3 && Typevar[k] != 2){ /* Dummy covariate and not age product nor fixed product */
1.242 brouard 8126: switch(Fixed[k]) {
8127: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 8128: modmaxcovj=0;
8129: modmincovj=0;
1.242 brouard 8130: for (i=1; i<=imx; i++) { /* Loop on individuals: reads the data file to get the maximum value of the modality of this covariate Vj*/
1.339 brouard 8131: /* printf("Waiting for error tricode Tvar[%d]=%d i=%d (int)(covar[Tvar[k]][i]=%d\n",k,Tvar[k], i, (int)(covar[Tvar[k]][i])); */
1.242 brouard 8132: ij=(int)(covar[Tvar[k]][i]);
8133: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
8134: * If product of Vn*Vm, still boolean *:
8135: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
8136: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
8137: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
8138: modality of the nth covariate of individual i. */
8139: if (ij > modmaxcovj)
8140: modmaxcovj=ij;
8141: else if (ij < modmincovj)
8142: modmincovj=ij;
1.287 brouard 8143: if (ij <0 || ij >1 ){
1.311 brouard 8144: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
8145: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
8146: fflush(ficlog);
8147: exit(1);
1.287 brouard 8148: }
8149: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 8150: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
8151: exit(1);
8152: }else
8153: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
8154: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
8155: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
8156: /* getting the maximum value of the modality of the covariate
8157: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
8158: female ies 1, then modmaxcovj=1.
8159: */
8160: } /* end for loop on individuals i */
8161: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
8162: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
8163: cptcode=modmaxcovj;
8164: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
8165: /*for (i=0; i<=cptcode; i++) {*/
8166: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
8167: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
8168: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
8169: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
8170: if( j != -1){
8171: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
8172: covariate for which somebody answered excluding
8173: undefined. Usually 2: 0 and 1. */
8174: }
8175: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
8176: covariate for which somebody answered including
8177: undefined. Usually 3: -1, 0 and 1. */
8178: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
8179: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
8180: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 8181:
1.242 brouard 8182: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
8183: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
8184: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
8185: /* modmincovj=3; modmaxcovj = 7; */
8186: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
8187: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
8188: /* defining two dummy variables: variables V1_1 and V1_2.*/
8189: /* nbcode[Tvar[j]][ij]=k; */
8190: /* nbcode[Tvar[j]][1]=0; */
8191: /* nbcode[Tvar[j]][2]=1; */
8192: /* nbcode[Tvar[j]][3]=2; */
8193: /* To be continued (not working yet). */
8194: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 8195:
8196: /* for (i=modmincovj; i<=modmaxcovj; i++) { */ /* i= 1 to 2 for dichotomous, or from 1 to 3 or from -1 or 0 to 1 currently*/
8197: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
8198: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
8199: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
8200: /*, could be restored in the future */
8201: for (i=0; i<=1; i++) { /* i= 1 to 2 for dichotomous, or from 1 to 3 or from -1 or 0 to 1 currently*/
1.242 brouard 8202: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
8203: break;
8204: }
8205: ij++;
1.287 brouard 8206: nbcode[Tvar[k]][ij]=i; /* stores the original value of modality i in an array nbcode, ij modality from 1 to last non-nul modality. nbcode[1][1]=0 nbcode[1][2]=1 . Could be -1*/
1.242 brouard 8207: cptcode = ij; /* New max modality for covar j */
8208: } /* end of loop on modality i=-1 to 1 or more */
8209: break;
8210: case 1: /* Testing on varying covariate, could be simple and
8211: * should look at waves or product of fixed *
8212: * varying. No time to test -1, assuming 0 and 1 only */
8213: ij=0;
8214: for(i=0; i<=1;i++){
8215: nbcode[Tvar[k]][++ij]=i;
8216: }
8217: break;
8218: default:
8219: break;
8220: } /* end switch */
8221: } /* end dummy test */
1.349 brouard 8222: if(Dummy[k]==1 && Typevar[k] !=1 && Typevar[k] !=3 && Fixed ==0){ /* Fixed Quantitative covariate and not age product */
1.311 brouard 8223: for (i=1; i<=imx; i++) { /* Loop on individuals: reads the data file to get the maximum value of the modality of this covariate Vj*/
1.335 brouard 8224: if(Tvar[k]<=0 || Tvar[k]>=NCOVMAX){
8225: printf("Error k=%d \n",k);
8226: exit(1);
8227: }
1.311 brouard 8228: if(isnan(covar[Tvar[k]][i])){
8229: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
8230: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
8231: fflush(ficlog);
8232: exit(1);
8233: }
8234: }
1.335 brouard 8235: } /* end Quanti */
1.287 brouard 8236: } /* end of loop on model-covariate k. nbcode[Tvark][1]=-1, nbcode[Tvark][1]=0 and nbcode[Tvark][2]=1 sets the value of covariate k*/
1.242 brouard 8237:
8238: for (k=-1; k< maxncov; k++) Ndum[k]=0;
8239: /* Look at fixed dummy (single or product) covariates to check empty modalities */
8240: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
8241: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
8242: ij=Tvar[i]; /* Tvar 5,4,3,6,5,7,1,4 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V4*age */
8243: Ndum[ij]++; /* Count the # of 1, 2 etc: {1,1,1,2,2,1,1} because V1 once, V2 once, two V4 and V5 in above */
8244: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
8245: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
8246:
8247: ij=0;
8248: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
1.335 brouard 8249: for (k=1; k<= cptcovt; k++) { /* cptcovt: total number of covariates of the model (2) nbocc(+)+1 = 8 excepting constant and age and age*age */
8250: /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
1.242 brouard 8251: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
8252: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
1.335 brouard 8253: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy simple and non empty in the model */
8254: /* Typevar[k] =0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
8255: /* Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product*/
1.242 brouard 8256: /* If product not in single variable we don't print results */
8257: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
1.335 brouard 8258: ++ij;/* V5 + V4 + V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V1, *//* V5 quanti, V2 quanti, V4, V3, V1 dummies */
8259: /* k= 1 2 3 4 5 6 7 8 9 */
8260: /* Tvar[k]= 5 4 3 6 5 2 7 1 1 */
8261: /* ij 1 2 3 */
8262: /* Tvaraff[ij]= 4 3 1 */
8263: /* Tmodelind[ij]=2 3 9 */
8264: /* TmodelInvind[ij]=2 1 1 */
1.242 brouard 8265: Tvaraff[ij]=Tvar[k]; /* For printing combination *//* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, Tvar {5, 4, 3, 6, 5, 2, 7, 1, 1} Tvaraff={4, 3, 1} V4, V3, V1*/
8266: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
8267: TmodelInvind[ij]=Tvar[k]- ncovcol-nqv; /* Inverse TmodelInvind[2=V4]=2 second dummy varying cov (V4)4-1-1 {0, 2, 1, } TmodelInvind[3]=1 */
8268: if(Fixed[k]!=0)
8269: anyvaryingduminmodel=1;
8270: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
8271: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
8272: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
8273: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
8274: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
8275: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
8276: }
8277: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
8278: /* ij--; */
8279: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.335 brouard 8280: *cptcov=ij; /* cptcov= Number of total real effective simple dummies (fixed or time arying) effective (used as cptcoveff in other functions)
1.242 brouard 8281: * because they can be excluded from the model and real
8282: * if in the model but excluded because missing values, but how to get k from ij?*/
8283: for(j=ij+1; j<= cptcovt; j++){
8284: Tvaraff[j]=0;
8285: Tmodelind[j]=0;
8286: }
8287: for(j=ntveff+1; j<= cptcovt; j++){
8288: TmodelInvind[j]=0;
8289: }
8290: /* To be sorted */
8291: ;
8292: }
1.126 brouard 8293:
1.145 brouard 8294:
1.126 brouard 8295: /*********** Health Expectancies ****************/
8296:
1.235 brouard 8297: void evsij(double ***eij, double x[], int nlstate, int stepm, int bage, int fage, double **oldm, double **savm, int cij, int estepm,char strstart[], int nres )
1.126 brouard 8298:
8299: {
8300: /* Health expectancies, no variances */
1.329 brouard 8301: /* cij is the combination in the list of combination of dummy covariates */
8302: /* strstart is a string of time at start of computing */
1.164 brouard 8303: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 8304: int nhstepma, nstepma; /* Decreasing with age */
8305: double age, agelim, hf;
8306: double ***p3mat;
8307: double eip;
8308:
1.238 brouard 8309: /* pstamp(ficreseij); */
1.126 brouard 8310: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
8311: fprintf(ficreseij,"# Age");
8312: for(i=1; i<=nlstate;i++){
8313: for(j=1; j<=nlstate;j++){
8314: fprintf(ficreseij," e%1d%1d ",i,j);
8315: }
8316: fprintf(ficreseij," e%1d. ",i);
8317: }
8318: fprintf(ficreseij,"\n");
8319:
8320:
8321: if(estepm < stepm){
8322: printf ("Problem %d lower than %d\n",estepm, stepm);
8323: }
8324: else hstepm=estepm;
8325: /* We compute the life expectancy from trapezoids spaced every estepm months
8326: * This is mainly to measure the difference between two models: for example
8327: * if stepm=24 months pijx are given only every 2 years and by summing them
8328: * we are calculating an estimate of the Life Expectancy assuming a linear
8329: * progression in between and thus overestimating or underestimating according
8330: * to the curvature of the survival function. If, for the same date, we
8331: * estimate the model with stepm=1 month, we can keep estepm to 24 months
8332: * to compare the new estimate of Life expectancy with the same linear
8333: * hypothesis. A more precise result, taking into account a more precise
8334: * curvature will be obtained if estepm is as small as stepm. */
8335:
8336: /* For example we decided to compute the life expectancy with the smallest unit */
8337: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
8338: nhstepm is the number of hstepm from age to agelim
8339: nstepm is the number of stepm from age to agelin.
1.270 brouard 8340: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 8341: and note for a fixed period like estepm months */
8342: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
8343: survival function given by stepm (the optimization length). Unfortunately it
8344: means that if the survival funtion is printed only each two years of age and if
8345: you sum them up and add 1 year (area under the trapezoids) you won't get the same
8346: results. So we changed our mind and took the option of the best precision.
8347: */
8348: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
8349:
8350: agelim=AGESUP;
8351: /* If stepm=6 months */
8352: /* Computed by stepm unit matrices, product of hstepm matrices, stored
8353: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
8354:
8355: /* nhstepm age range expressed in number of stepm */
8356: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
8357: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
8358: /* if (stepm >= YEARM) hstepm=1;*/
8359: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
8360: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8361:
8362: for (age=bage; age<=fage; age ++){
8363: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
8364: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
8365: /* if (stepm >= YEARM) hstepm=1;*/
8366: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
8367:
8368: /* If stepm=6 months */
8369: /* Computed by stepm unit matrices, product of hstepma matrices, stored
8370: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 brouard 8371: /* printf("HELLO evsij Entering hpxij age=%d cij=%d hstepm=%d x[1]=%f nres=%d\n",(int) age, cij, hstepm, x[1], nres); */
1.235 brouard 8372: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 8373:
8374: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
8375:
8376: printf("%d|",(int)age);fflush(stdout);
8377: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
8378:
8379: /* Computing expectancies */
8380: for(i=1; i<=nlstate;i++)
8381: for(j=1; j<=nlstate;j++)
8382: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
8383: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
8384:
8385: /* if((int)age==70)printf("i=%2d,j=%2d,h=%2d,age=%3d,%9.4f,%9.4f,%9.4f\n",i,j,h,(int)age,p3mat[i][j][h],hf,eij[i][j][(int)age]);*/
8386:
8387: }
8388:
8389: fprintf(ficreseij,"%3.0f",age );
8390: for(i=1; i<=nlstate;i++){
8391: eip=0;
8392: for(j=1; j<=nlstate;j++){
8393: eip +=eij[i][j][(int)age];
8394: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
8395: }
8396: fprintf(ficreseij,"%9.4f", eip );
8397: }
8398: fprintf(ficreseij,"\n");
8399:
8400: }
8401: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8402: printf("\n");
8403: fprintf(ficlog,"\n");
8404:
8405: }
8406:
1.235 brouard 8407: void cvevsij(double ***eij, double x[], int nlstate, int stepm, int bage, int fage, double **oldm, double **savm, int cij, int estepm,double delti[],double **matcov,char strstart[], int nres )
1.126 brouard 8408:
8409: {
8410: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 8411: to initial status i, ei. .
1.126 brouard 8412: */
1.336 brouard 8413: /* Very time consuming function, but already optimized with precov */
1.126 brouard 8414: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
8415: int nhstepma, nstepma; /* Decreasing with age */
8416: double age, agelim, hf;
8417: double ***p3matp, ***p3matm, ***varhe;
8418: double **dnewm,**doldm;
8419: double *xp, *xm;
8420: double **gp, **gm;
8421: double ***gradg, ***trgradg;
8422: int theta;
8423:
8424: double eip, vip;
8425:
8426: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
8427: xp=vector(1,npar);
8428: xm=vector(1,npar);
8429: dnewm=matrix(1,nlstate*nlstate,1,npar);
8430: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
8431:
8432: pstamp(ficresstdeij);
8433: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
8434: fprintf(ficresstdeij,"# Age");
8435: for(i=1; i<=nlstate;i++){
8436: for(j=1; j<=nlstate;j++)
8437: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
8438: fprintf(ficresstdeij," e%1d. ",i);
8439: }
8440: fprintf(ficresstdeij,"\n");
8441:
8442: pstamp(ficrescveij);
8443: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
8444: fprintf(ficrescveij,"# Age");
8445: for(i=1; i<=nlstate;i++)
8446: for(j=1; j<=nlstate;j++){
8447: cptj= (j-1)*nlstate+i;
8448: for(i2=1; i2<=nlstate;i2++)
8449: for(j2=1; j2<=nlstate;j2++){
8450: cptj2= (j2-1)*nlstate+i2;
8451: if(cptj2 <= cptj)
8452: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
8453: }
8454: }
8455: fprintf(ficrescveij,"\n");
8456:
8457: if(estepm < stepm){
8458: printf ("Problem %d lower than %d\n",estepm, stepm);
8459: }
8460: else hstepm=estepm;
8461: /* We compute the life expectancy from trapezoids spaced every estepm months
8462: * This is mainly to measure the difference between two models: for example
8463: * if stepm=24 months pijx are given only every 2 years and by summing them
8464: * we are calculating an estimate of the Life Expectancy assuming a linear
8465: * progression in between and thus overestimating or underestimating according
8466: * to the curvature of the survival function. If, for the same date, we
8467: * estimate the model with stepm=1 month, we can keep estepm to 24 months
8468: * to compare the new estimate of Life expectancy with the same linear
8469: * hypothesis. A more precise result, taking into account a more precise
8470: * curvature will be obtained if estepm is as small as stepm. */
8471:
8472: /* For example we decided to compute the life expectancy with the smallest unit */
8473: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
8474: nhstepm is the number of hstepm from age to agelim
8475: nstepm is the number of stepm from age to agelin.
8476: Look at hpijx to understand the reason of that which relies in memory size
8477: and note for a fixed period like estepm months */
8478: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
8479: survival function given by stepm (the optimization length). Unfortunately it
8480: means that if the survival funtion is printed only each two years of age and if
8481: you sum them up and add 1 year (area under the trapezoids) you won't get the same
8482: results. So we changed our mind and took the option of the best precision.
8483: */
8484: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
8485:
8486: /* If stepm=6 months */
8487: /* nhstepm age range expressed in number of stepm */
8488: agelim=AGESUP;
8489: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
8490: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
8491: /* if (stepm >= YEARM) hstepm=1;*/
8492: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
8493:
8494: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8495: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8496: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
8497: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
8498: gp=matrix(0,nhstepm,1,nlstate*nlstate);
8499: gm=matrix(0,nhstepm,1,nlstate*nlstate);
8500:
8501: for (age=bage; age<=fage; age ++){
8502: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
8503: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
8504: /* if (stepm >= YEARM) hstepm=1;*/
8505: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 8506:
1.126 brouard 8507: /* If stepm=6 months */
8508: /* Computed by stepm unit matrices, product of hstepma matrices, stored
8509: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
8510:
8511: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 8512:
1.126 brouard 8513: /* Computing Variances of health expectancies */
8514: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
8515: decrease memory allocation */
8516: for(theta=1; theta <=npar; theta++){
8517: for(i=1; i<=npar; i++){
1.222 brouard 8518: xp[i] = x[i] + (i==theta ?delti[theta]:0);
8519: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 8520: }
1.235 brouard 8521: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
8522: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 8523:
1.126 brouard 8524: for(j=1; j<= nlstate; j++){
1.222 brouard 8525: for(i=1; i<=nlstate; i++){
8526: for(h=0; h<=nhstepm-1; h++){
8527: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
8528: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
8529: }
8530: }
1.126 brouard 8531: }
1.218 brouard 8532:
1.126 brouard 8533: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 8534: for(h=0; h<=nhstepm-1; h++){
8535: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
8536: }
1.126 brouard 8537: }/* End theta */
8538:
8539:
8540: for(h=0; h<=nhstepm-1; h++)
8541: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 8542: for(theta=1; theta <=npar; theta++)
8543: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 8544:
1.218 brouard 8545:
1.222 brouard 8546: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 8547: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 8548: varhe[ij][ji][(int)age] =0.;
1.218 brouard 8549:
1.222 brouard 8550: printf("%d|",(int)age);fflush(stdout);
8551: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
8552: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 8553: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 8554: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
8555: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
8556: for(ij=1;ij<=nlstate*nlstate;ij++)
8557: for(ji=1;ji<=nlstate*nlstate;ji++)
8558: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 8559: }
8560: }
1.320 brouard 8561: /* if((int)age ==50){ */
8562: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
8563: /* } */
1.126 brouard 8564: /* Computing expectancies */
1.235 brouard 8565: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 8566: for(i=1; i<=nlstate;i++)
8567: for(j=1; j<=nlstate;j++)
1.222 brouard 8568: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
8569: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 8570:
1.222 brouard 8571: /* if((int)age==70)printf("i=%2d,j=%2d,h=%2d,age=%3d,%9.4f,%9.4f,%9.4f\n",i,j,h,(int)age,p3mat[i][j][h],hf,eij[i][j][(int)age]);*/
1.218 brouard 8572:
1.222 brouard 8573: }
1.269 brouard 8574:
8575: /* Standard deviation of expectancies ij */
1.126 brouard 8576: fprintf(ficresstdeij,"%3.0f",age );
8577: for(i=1; i<=nlstate;i++){
8578: eip=0.;
8579: vip=0.;
8580: for(j=1; j<=nlstate;j++){
1.222 brouard 8581: eip += eij[i][j][(int)age];
8582: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
8583: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
8584: fprintf(ficresstdeij," %9.4f (%.4f)", eij[i][j][(int)age], sqrt(varhe[(j-1)*nlstate+i][(j-1)*nlstate+i][(int)age]) );
1.126 brouard 8585: }
8586: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
8587: }
8588: fprintf(ficresstdeij,"\n");
1.218 brouard 8589:
1.269 brouard 8590: /* Variance of expectancies ij */
1.126 brouard 8591: fprintf(ficrescveij,"%3.0f",age );
8592: for(i=1; i<=nlstate;i++)
8593: for(j=1; j<=nlstate;j++){
1.222 brouard 8594: cptj= (j-1)*nlstate+i;
8595: for(i2=1; i2<=nlstate;i2++)
8596: for(j2=1; j2<=nlstate;j2++){
8597: cptj2= (j2-1)*nlstate+i2;
8598: if(cptj2 <= cptj)
8599: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
8600: }
1.126 brouard 8601: }
8602: fprintf(ficrescveij,"\n");
1.218 brouard 8603:
1.126 brouard 8604: }
8605: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
8606: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
8607: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
8608: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
8609: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8610: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8611: printf("\n");
8612: fprintf(ficlog,"\n");
1.218 brouard 8613:
1.126 brouard 8614: free_vector(xm,1,npar);
8615: free_vector(xp,1,npar);
8616: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
8617: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
8618: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
8619: }
1.218 brouard 8620:
1.126 brouard 8621: /************ Variance ******************/
1.235 brouard 8622: void varevsij(char optionfilefiname[], double ***vareij, double **matcov, double x[], double delti[], int nlstate, int stepm, double bage, double fage, double **oldm, double **savm, double **prlim, double ftolpl, int *ncvyearp, int ij, int estepm, int cptcov, int cptcod, int popbased, int mobilav, char strstart[], int nres)
1.218 brouard 8623: {
1.361 brouard 8624: /** Computes the matrix of variance covariance of health expectancies e.j= sum_i w_i e_ij where w_i depends of popbased,
8625: * either cross-sectional or implied.
8626: * return vareij[i][j][(int)age]=cov(e.i,e.j)=sum_h sum_k trgrad(h_p.i) V(theta) grad(k_p.k) Equation 20
1.279 brouard 8627: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
8628: * double **newm;
8629: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
8630: */
1.218 brouard 8631:
8632: /* int movingaverage(); */
8633: double **dnewm,**doldm;
8634: double **dnewmp,**doldmp;
8635: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 8636: int first=0;
1.218 brouard 8637: int k;
8638: double *xp;
1.279 brouard 8639: double **gp, **gm; /**< for var eij */
8640: double ***gradg, ***trgradg; /**< for var eij */
8641: double **gradgp, **trgradgp; /**< for var p point j */
8642: double *gpp, *gmp; /**< for var p point j */
1.362 brouard 8643: double **varppt; /**< for var p.3 p.death nlstate+1 to nlstate+ndeath */
1.218 brouard 8644: double ***p3mat;
8645: double age,agelim, hf;
8646: /* double ***mobaverage; */
8647: int theta;
8648: char digit[4];
8649: char digitp[25];
8650:
8651: char fileresprobmorprev[FILENAMELENGTH];
8652:
8653: if(popbased==1){
8654: if(mobilav!=0)
8655: strcpy(digitp,"-POPULBASED-MOBILAV_");
8656: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
8657: }
8658: else
8659: strcpy(digitp,"-STABLBASED_");
1.126 brouard 8660:
1.218 brouard 8661: /* if (mobilav!=0) { */
8662: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8663: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
8664: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
8665: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
8666: /* } */
8667: /* } */
8668:
8669: strcpy(fileresprobmorprev,"PRMORPREV-");
8670: sprintf(digit,"%-d",ij);
8671: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
8672: strcat(fileresprobmorprev,digit); /* Tvar to be done */
8673: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
8674: strcat(fileresprobmorprev,fileresu);
8675: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
8676: printf("Problem with resultfile: %s\n", fileresprobmorprev);
8677: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
8678: }
8679: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
8680: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
8681: pstamp(ficresprobmorprev);
8682: fprintf(ficresprobmorprev,"# probabilities of dying before estepm=%d months for people of exact age and weighted probabilities w1*p1j+w2*p2j+... stand dev in()\n",estepm);
1.238 brouard 8683: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
1.337 brouard 8684:
8685: /* We use TinvDoQresult[nres][resultmodel[nres][j] we sort according to the equation model and the resultline: it is a choice */
8686: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ /\* To be done*\/ */
8687: /* fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
8688: /* } */
8689: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */ /* To be done*/
1.344 brouard 8690: /* fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); */
1.337 brouard 8691: fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 8692: }
1.337 brouard 8693: /* for(j=1;j<=cptcoveff;j++) */
8694: /* fprintf(ficresprobmorprev," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,TnsdVar[Tvaraff[j]])]); */
1.238 brouard 8695: fprintf(ficresprobmorprev,"\n");
8696:
1.218 brouard 8697: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
8698: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
8699: fprintf(ficresprobmorprev," p.%-d SE",j);
8700: for(i=1; i<=nlstate;i++)
8701: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
8702: }
8703: fprintf(ficresprobmorprev,"\n");
8704:
8705: fprintf(ficgp,"\n# Routine varevsij");
8706: fprintf(ficgp,"\nunset title \n");
8707: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
8708: fprintf(fichtm,"\n<li><h4> Computing probabilities of dying over estepm months as a weighted average (i.e global mortality independent of initial healh state)</h4></li>\n");
8709: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 8710:
1.361 brouard 8711: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath); /* In fact, currently a double */
1.218 brouard 8712: pstamp(ficresvij);
8713: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
8714: if(popbased==1)
8715: fprintf(ficresvij,"the age specific prevalence observed (cross-sectionally) in the population i.e cross-sectionally\n in each health state (popbased=1) (mobilav=%d\n",mobilav);
8716: else
8717: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
8718: fprintf(ficresvij,"# Age");
8719: for(i=1; i<=nlstate;i++)
8720: for(j=1; j<=nlstate;j++)
8721: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
8722: fprintf(ficresvij,"\n");
8723:
8724: xp=vector(1,npar);
8725: dnewm=matrix(1,nlstate,1,npar);
8726: doldm=matrix(1,nlstate,1,nlstate);
8727: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
8728: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
8729:
8730: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
8731: gpp=vector(nlstate+1,nlstate+ndeath);
8732: gmp=vector(nlstate+1,nlstate+ndeath);
8733: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 8734:
1.218 brouard 8735: if(estepm < stepm){
8736: printf ("Problem %d lower than %d\n",estepm, stepm);
8737: }
8738: else hstepm=estepm;
8739: /* For example we decided to compute the life expectancy with the smallest unit */
8740: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
8741: nhstepm is the number of hstepm from age to agelim
8742: nstepm is the number of stepm from age to agelim.
8743: Look at function hpijx to understand why because of memory size limitations,
8744: we decided (b) to get a life expectancy respecting the most precise curvature of the
8745: survival function given by stepm (the optimization length). Unfortunately it
8746: means that if the survival funtion is printed every two years of age and if
8747: you sum them up and add 1 year (area under the trapezoids) you won't get the same
8748: results. So we changed our mind and took the option of the best precision.
8749: */
8750: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
8751: agelim = AGESUP;
8752: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
8753: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
8754: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
8755: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8756: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
8757: gp=matrix(0,nhstepm,1,nlstate);
8758: gm=matrix(0,nhstepm,1,nlstate);
8759:
8760:
8761: for(theta=1; theta <=npar; theta++){
8762: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
8763: xp[i] = x[i] + (i==theta ?delti[theta]:0);
8764: }
1.279 brouard 8765: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
8766: * returns into prlim .
1.288 brouard 8767: */
1.242 brouard 8768: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 8769:
8770: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 8771: if (popbased==1) {
8772: if(mobilav ==0){
8773: for(i=1; i<=nlstate;i++)
8774: prlim[i][i]=probs[(int)age][i][ij];
8775: }else{ /* mobilav */
8776: for(i=1; i<=nlstate;i++)
8777: prlim[i][i]=mobaverage[(int)age][i][ij];
8778: }
8779: }
1.361 brouard 8780: /**< Computes the shifted plus (gp) transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 8781: */
8782: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres); /* Returns p3mat[i][j][h] for h=0 to nhstepm */
1.292 brouard 8783: /**< And for each alive state j, sums over i \f$ w^i_x {}{h}_p^{ij}x\f$, which are the probability
1.279 brouard 8784: * at horizon h in state j including mortality.
8785: */
1.218 brouard 8786: for(j=1; j<= nlstate; j++){
8787: for(h=0; h<=nhstepm; h++){
8788: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
1.361 brouard 8789: gp[h][j] += prlim[i][i]*p3mat[i][j][h]; /* gp[h][j]= w_i h_pij */
1.218 brouard 8790: }
8791: }
1.279 brouard 8792: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 8793: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 8794: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 8795: */
1.361 brouard 8796: for(j=nlstate+1;j<=nlstate+ndeath;j++){ /* Currently only once for theta plus p.3(age) Sum_i wi pi3*/
1.218 brouard 8797: for(i=1,gpp[j]=0.; i<= nlstate; i++)
8798: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 8799: }
8800:
8801: /* Again with minus shift */
1.218 brouard 8802:
8803: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
8804: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 8805:
1.242 brouard 8806: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 8807:
8808: if (popbased==1) {
8809: if(mobilav ==0){
8810: for(i=1; i<=nlstate;i++)
8811: prlim[i][i]=probs[(int)age][i][ij];
8812: }else{ /* mobilav */
8813: for(i=1; i<=nlstate;i++)
8814: prlim[i][i]=mobaverage[(int)age][i][ij];
8815: }
8816: }
8817:
1.361 brouard 8818: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres); /* Still minus */
1.218 brouard 8819:
1.361 brouard 8820: for(j=1; j<= nlstate; j++){ /* gm[h][j]= Sum_i of wi * pij = h_p.j */
1.218 brouard 8821: for(h=0; h<=nhstepm; h++){
8822: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
8823: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
8824: }
8825: }
8826: /* This for computing probability of death (h=1 means
8827: computed over hstepm matrices product = hstepm*stepm months)
1.361 brouard 8828: as a weighted average of prlim. j is death. gmp[3]=sum_i w_i*p_i3=p.3 minus theta
1.218 brouard 8829: */
1.361 brouard 8830: for(j=nlstate+1;j<=nlstate+ndeath;j++){ /* Currently only once theta_minus p.3=Sum_i wi pi3*/
1.218 brouard 8831: for(i=1,gmp[j]=0.; i<= nlstate; i++)
8832: gmp[j] += prlim[i][i]*p3mat[i][j][1];
8833: }
1.279 brouard 8834: /* end shifting computations */
8835:
1.361 brouard 8836: /**< Computing gradient of p.j matrix at horizon h and still for one parameter of vector theta
8837: * equation 31 and 32
1.279 brouard 8838: */
1.361 brouard 8839: for(j=1; j<= nlstate; j++) /* computes grad p.j(x, over each h) where p.j is Sum_i w_i*pij(x over h)
8840: * equation 24 */
1.218 brouard 8841: for(h=0; h<=nhstepm; h++){
8842: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
8843: }
1.361 brouard 8844: /**< Gradient of overall mortality p.3 (or p.death)
1.279 brouard 8845: */
1.361 brouard 8846: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* computes grad of p.3 from wi+pi3 grad p.3 (theta) */
1.218 brouard 8847: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
8848: }
8849:
8850: } /* End theta */
1.279 brouard 8851:
1.361 brouard 8852: /* We got the gradient matrix for each theta and each state j of gradg(h]theta][j)=grad(_hp.j(theta) */
8853: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar);
1.218 brouard 8854:
1.361 brouard 8855: for(h=0; h<=nhstepm; h++) /* veij */ /* computes the transposed of grad (_hp.j(theta)*/
1.218 brouard 8856: for(j=1; j<=nlstate;j++)
8857: for(theta=1; theta <=npar; theta++)
8858: trgradg[h][j][theta]=gradg[h][theta][j];
8859:
1.361 brouard 8860: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* computes transposed of grad p.3 (theta)*/
1.218 brouard 8861: for(theta=1; theta <=npar; theta++)
8862: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 8863: /**< as well as its transposed matrix
8864: */
1.218 brouard 8865:
8866: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
8867: for(i=1;i<=nlstate;i++)
8868: for(j=1;j<=nlstate;j++)
8869: vareij[i][j][(int)age] =0.;
1.279 brouard 8870:
8871: /* Computing trgradg by matcov by gradg at age and summing over h
1.361 brouard 8872: * and k (nhstepm) formula 32 of article
8873: * Lievre-Brouard-Heathcote so that for each j, computes the cov(e.j,e.k) (formula 31).
8874: * for given h and k computes trgradg[h](i,j) matcov (theta) gradg(k)(i,j) into vareij[i][j] which is
8875: cov(e.i,e.j) and sums on h and k
8876: * including the covariances.
1.279 brouard 8877: */
8878:
1.218 brouard 8879: for(h=0;h<=nhstepm;h++){
8880: for(k=0;k<=nhstepm;k++){
8881: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
8882: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
8883: for(i=1;i<=nlstate;i++)
8884: for(j=1;j<=nlstate;j++)
1.361 brouard 8885: vareij[i][j][(int)age] += doldm[i][j]*hf*hf; /* This is vareij=sum_h sum_k trgrad(h_pij) V(theta) grad(k_pij)
8886: including the covariances of e.j */
1.218 brouard 8887: }
8888: }
8889:
1.361 brouard 8890: /* Mortality: pptj is p.3 or p.death = trgradgp by cov by gradgp, variance of
8891: * p.3=1-p..=1-sum i p.i overall mortality computed directly because
1.279 brouard 8892: * we compute the grad (wix pijx) instead of grad (pijx),even if
1.361 brouard 8893: * wix is independent of theta.
1.279 brouard 8894: */
1.218 brouard 8895: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
8896: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
8897: for(j=nlstate+1;j<=nlstate+ndeath;j++)
8898: for(i=nlstate+1;i<=nlstate+ndeath;i++)
1.361 brouard 8899: varppt[j][i]=doldmp[j][i]; /* This is the variance of p.3 */
1.218 brouard 8900: /* end ppptj */
8901: /* x centered again */
8902:
1.242 brouard 8903: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 8904:
8905: if (popbased==1) {
8906: if(mobilav ==0){
8907: for(i=1; i<=nlstate;i++)
8908: prlim[i][i]=probs[(int)age][i][ij];
8909: }else{ /* mobilav */
8910: for(i=1; i<=nlstate;i++)
8911: prlim[i][i]=mobaverage[(int)age][i][ij];
8912: }
8913: }
8914:
8915: /* This for computing probability of death (h=1 means
8916: computed over hstepm (estepm) matrices product = hstepm*stepm months)
8917: as a weighted average of prlim.
8918: */
1.235 brouard 8919: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 8920: for(j=nlstate+1;j<=nlstate+ndeath;j++){
8921: for(i=1,gmp[j]=0.;i<= nlstate; i++)
1.361 brouard 8922: gmp[j] += prlim[i][i]*p3mat[i][j][1]; /* gmp[j] is p.3 */
1.218 brouard 8923: }
8924: /* end probability of death */
8925:
8926: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
8927: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
1.361 brouard 8928: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));/* p.3 (STD p.3) */
1.218 brouard 8929: for(i=1; i<=nlstate;i++){
1.361 brouard 8930: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]); /* wi, pi3 */
1.218 brouard 8931: }
8932: }
8933: fprintf(ficresprobmorprev,"\n");
8934:
8935: fprintf(ficresvij,"%.0f ",age );
8936: for(i=1; i<=nlstate;i++)
8937: for(j=1; j<=nlstate;j++){
8938: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
8939: }
8940: fprintf(ficresvij,"\n");
8941: free_matrix(gp,0,nhstepm,1,nlstate);
8942: free_matrix(gm,0,nhstepm,1,nlstate);
8943: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
8944: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
8945: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8946: } /* End age */
8947: free_vector(gpp,nlstate+1,nlstate+ndeath);
8948: free_vector(gmp,nlstate+1,nlstate+ndeath);
8949: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
8950: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
8951: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
8952: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
8953: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
8954: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
8955: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
8956: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
8957: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
8958: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
8959: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
8960: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
8961: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
8962: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
8963: fprintf(fichtm,"\n<br> Probability is computed over estepm=%d months. <br> <img src=\"%s%s.svg\"> <br>\n", estepm,subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
8964: /* fprintf(fichtm,"\n<br> Probability is computed over estepm=%d months and then divided by estepm and multiplied by %.0f in order to have the probability to die over a year <br> <img src=\"varmuptjgr%s%s.svg\"> <br>\n", stepm,YEARM,digitp,digit);
1.126 brouard 8965: */
1.218 brouard 8966: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
8967: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 8968:
1.218 brouard 8969: free_vector(xp,1,npar);
8970: free_matrix(doldm,1,nlstate,1,nlstate);
8971: free_matrix(dnewm,1,nlstate,1,npar);
8972: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
8973: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
8974: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
8975: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8976: fclose(ficresprobmorprev);
8977: fflush(ficgp);
8978: fflush(fichtm);
8979: } /* end varevsij */
1.126 brouard 8980:
8981: /************ Variance of prevlim ******************/
1.269 brouard 8982: void varprevlim(char fileresvpl[], FILE *ficresvpl, double **varpl, double **matcov, double x[], double delti[], int nlstate, int stepm, double bage, double fage, double **oldm, double **savm, double **prlim, double ftolpl, int *ncvyearp, int ij, char strstart[], int nres)
1.126 brouard 8983: {
1.205 brouard 8984: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 8985: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 8986:
1.268 brouard 8987: double **dnewmpar,**doldm;
1.126 brouard 8988: int i, j, nhstepm, hstepm;
8989: double *xp;
8990: double *gp, *gm;
8991: double **gradg, **trgradg;
1.208 brouard 8992: double **mgm, **mgp;
1.126 brouard 8993: double age,agelim;
8994: int theta;
8995:
8996: pstamp(ficresvpl);
1.288 brouard 8997: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 8998: fprintf(ficresvpl,"# Age ");
8999: if(nresult >=1)
9000: fprintf(ficresvpl," Result# ");
1.126 brouard 9001: for(i=1; i<=nlstate;i++)
9002: fprintf(ficresvpl," %1d-%1d",i,i);
9003: fprintf(ficresvpl,"\n");
9004:
9005: xp=vector(1,npar);
1.268 brouard 9006: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 9007: doldm=matrix(1,nlstate,1,nlstate);
9008:
9009: hstepm=1*YEARM; /* Every year of age */
9010: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
9011: agelim = AGESUP;
9012: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
9013: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9014: if (stepm >= YEARM) hstepm=1;
9015: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9016: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 9017: mgp=matrix(1,npar,1,nlstate);
9018: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 9019: gp=vector(1,nlstate);
9020: gm=vector(1,nlstate);
9021:
9022: for(theta=1; theta <=npar; theta++){
9023: for(i=1; i<=npar; i++){ /* Computes gradient */
9024: xp[i] = x[i] + (i==theta ?delti[theta]:0);
9025: }
1.288 brouard 9026: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
9027: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
9028: /* else */
9029: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 9030: for(i=1;i<=nlstate;i++){
1.126 brouard 9031: gp[i] = prlim[i][i];
1.208 brouard 9032: mgp[theta][i] = prlim[i][i];
9033: }
1.126 brouard 9034: for(i=1; i<=npar; i++) /* Computes gradient */
9035: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 9036: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
9037: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
9038: /* else */
9039: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 9040: for(i=1;i<=nlstate;i++){
1.126 brouard 9041: gm[i] = prlim[i][i];
1.208 brouard 9042: mgm[theta][i] = prlim[i][i];
9043: }
1.126 brouard 9044: for(i=1;i<=nlstate;i++)
9045: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 9046: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 9047: } /* End theta */
9048:
9049: trgradg =matrix(1,nlstate,1,npar);
9050:
9051: for(j=1; j<=nlstate;j++)
9052: for(theta=1; theta <=npar; theta++)
9053: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 9054: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
9055: /* printf("\nmgm mgp %d ",(int)age); */
9056: /* for(j=1; j<=nlstate;j++){ */
9057: /* printf(" %d ",j); */
9058: /* for(theta=1; theta <=npar; theta++) */
9059: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
9060: /* printf("\n "); */
9061: /* } */
9062: /* } */
9063: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
9064: /* printf("\n gradg %d ",(int)age); */
9065: /* for(j=1; j<=nlstate;j++){ */
9066: /* printf("%d ",j); */
9067: /* for(theta=1; theta <=npar; theta++) */
9068: /* printf("%d %lf ",theta,gradg[theta][j]); */
9069: /* printf("\n "); */
9070: /* } */
9071: /* } */
1.126 brouard 9072:
9073: for(i=1;i<=nlstate;i++)
9074: varpl[i][(int)age] =0.;
1.209 brouard 9075: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 9076: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
9077: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 9078: }else{
1.268 brouard 9079: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
9080: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 9081: }
1.126 brouard 9082: for(i=1;i<=nlstate;i++)
9083: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
9084:
9085: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 9086: if(nresult >=1)
9087: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 9088: for(i=1; i<=nlstate;i++){
1.126 brouard 9089: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 9090: /* for(j=1;j<=nlstate;j++) */
9091: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
9092: }
1.126 brouard 9093: fprintf(ficresvpl,"\n");
9094: free_vector(gp,1,nlstate);
9095: free_vector(gm,1,nlstate);
1.208 brouard 9096: free_matrix(mgm,1,npar,1,nlstate);
9097: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 9098: free_matrix(gradg,1,npar,1,nlstate);
9099: free_matrix(trgradg,1,nlstate,1,npar);
9100: } /* End age */
9101:
9102: free_vector(xp,1,npar);
9103: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 9104: free_matrix(dnewmpar,1,nlstate,1,nlstate);
9105:
9106: }
9107:
9108:
9109: /************ Variance of backprevalence limit ******************/
1.269 brouard 9110: void varbrevlim(char fileresvbl[], FILE *ficresvbl, double **varbpl, double **matcov, double x[], double delti[], int nlstate, int stepm, double bage, double fage, double **oldm, double **savm, double **bprlim, double ftolpl, int mobilavproj, int *ncvyearp, int ij, char strstart[], int nres)
1.268 brouard 9111: {
9112: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
9113: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
9114:
9115: double **dnewmpar,**doldm;
9116: int i, j, nhstepm, hstepm;
9117: double *xp;
9118: double *gp, *gm;
9119: double **gradg, **trgradg;
9120: double **mgm, **mgp;
9121: double age,agelim;
9122: int theta;
9123:
9124: pstamp(ficresvbl);
9125: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
9126: fprintf(ficresvbl,"# Age ");
9127: if(nresult >=1)
9128: fprintf(ficresvbl," Result# ");
9129: for(i=1; i<=nlstate;i++)
9130: fprintf(ficresvbl," %1d-%1d",i,i);
9131: fprintf(ficresvbl,"\n");
9132:
9133: xp=vector(1,npar);
9134: dnewmpar=matrix(1,nlstate,1,npar);
9135: doldm=matrix(1,nlstate,1,nlstate);
9136:
9137: hstepm=1*YEARM; /* Every year of age */
9138: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
9139: agelim = AGEINF;
9140: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
9141: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9142: if (stepm >= YEARM) hstepm=1;
9143: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9144: gradg=matrix(1,npar,1,nlstate);
9145: mgp=matrix(1,npar,1,nlstate);
9146: mgm=matrix(1,npar,1,nlstate);
9147: gp=vector(1,nlstate);
9148: gm=vector(1,nlstate);
9149:
9150: for(theta=1; theta <=npar; theta++){
9151: for(i=1; i<=npar; i++){ /* Computes gradient */
9152: xp[i] = x[i] + (i==theta ?delti[theta]:0);
9153: }
9154: if(mobilavproj > 0 )
9155: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
9156: else
9157: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
9158: for(i=1;i<=nlstate;i++){
9159: gp[i] = bprlim[i][i];
9160: mgp[theta][i] = bprlim[i][i];
9161: }
9162: for(i=1; i<=npar; i++) /* Computes gradient */
9163: xp[i] = x[i] - (i==theta ?delti[theta]:0);
9164: if(mobilavproj > 0 )
9165: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
9166: else
9167: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
9168: for(i=1;i<=nlstate;i++){
9169: gm[i] = bprlim[i][i];
9170: mgm[theta][i] = bprlim[i][i];
9171: }
9172: for(i=1;i<=nlstate;i++)
9173: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
9174: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
9175: } /* End theta */
9176:
9177: trgradg =matrix(1,nlstate,1,npar);
9178:
9179: for(j=1; j<=nlstate;j++)
9180: for(theta=1; theta <=npar; theta++)
9181: trgradg[j][theta]=gradg[theta][j];
9182: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
9183: /* printf("\nmgm mgp %d ",(int)age); */
9184: /* for(j=1; j<=nlstate;j++){ */
9185: /* printf(" %d ",j); */
9186: /* for(theta=1; theta <=npar; theta++) */
9187: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
9188: /* printf("\n "); */
9189: /* } */
9190: /* } */
9191: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
9192: /* printf("\n gradg %d ",(int)age); */
9193: /* for(j=1; j<=nlstate;j++){ */
9194: /* printf("%d ",j); */
9195: /* for(theta=1; theta <=npar; theta++) */
9196: /* printf("%d %lf ",theta,gradg[theta][j]); */
9197: /* printf("\n "); */
9198: /* } */
9199: /* } */
9200:
9201: for(i=1;i<=nlstate;i++)
9202: varbpl[i][(int)age] =0.;
9203: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
9204: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
9205: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
9206: }else{
9207: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
9208: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
9209: }
9210: for(i=1;i<=nlstate;i++)
9211: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
9212:
9213: fprintf(ficresvbl,"%.0f ",age );
9214: if(nresult >=1)
9215: fprintf(ficresvbl,"%d ",nres );
9216: for(i=1; i<=nlstate;i++)
9217: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
9218: fprintf(ficresvbl,"\n");
9219: free_vector(gp,1,nlstate);
9220: free_vector(gm,1,nlstate);
9221: free_matrix(mgm,1,npar,1,nlstate);
9222: free_matrix(mgp,1,npar,1,nlstate);
9223: free_matrix(gradg,1,npar,1,nlstate);
9224: free_matrix(trgradg,1,nlstate,1,npar);
9225: } /* End age */
9226:
9227: free_vector(xp,1,npar);
9228: free_matrix(doldm,1,nlstate,1,npar);
9229: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 9230:
9231: }
9232:
9233: /************ Variance of one-step probabilities ******************/
9234: void varprob(char optionfilefiname[], double **matcov, double x[], double delti[], int nlstate, double bage, double fage, int ij, int *Tvar, int **nbcode, int *ncodemax, char strstart[])
1.222 brouard 9235: {
9236: int i, j=0, k1, l1, tj;
9237: int k2, l2, j1, z1;
9238: int k=0, l;
9239: int first=1, first1, first2;
1.326 brouard 9240: int nres=0; /* New */
1.222 brouard 9241: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
9242: double **dnewm,**doldm;
9243: double *xp;
9244: double *gp, *gm;
9245: double **gradg, **trgradg;
9246: double **mu;
9247: double age, cov[NCOVMAX+1];
9248: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
9249: int theta;
9250: char fileresprob[FILENAMELENGTH];
9251: char fileresprobcov[FILENAMELENGTH];
9252: char fileresprobcor[FILENAMELENGTH];
9253: double ***varpij;
9254:
9255: strcpy(fileresprob,"PROB_");
1.356 brouard 9256: strcat(fileresprob,fileresu);
1.222 brouard 9257: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
9258: printf("Problem with resultfile: %s\n", fileresprob);
9259: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
9260: }
9261: strcpy(fileresprobcov,"PROBCOV_");
9262: strcat(fileresprobcov,fileresu);
9263: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
9264: printf("Problem with resultfile: %s\n", fileresprobcov);
9265: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
9266: }
9267: strcpy(fileresprobcor,"PROBCOR_");
9268: strcat(fileresprobcor,fileresu);
9269: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
9270: printf("Problem with resultfile: %s\n", fileresprobcor);
9271: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
9272: }
9273: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
9274: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
9275: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
9276: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
9277: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
9278: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
9279: pstamp(ficresprob);
9280: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
9281: fprintf(ficresprob,"# Age");
9282: pstamp(ficresprobcov);
9283: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
9284: fprintf(ficresprobcov,"# Age");
9285: pstamp(ficresprobcor);
9286: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
9287: fprintf(ficresprobcor,"# Age");
1.126 brouard 9288:
9289:
1.222 brouard 9290: for(i=1; i<=nlstate;i++)
9291: for(j=1; j<=(nlstate+ndeath);j++){
9292: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
9293: fprintf(ficresprobcov," p%1d-%1d ",i,j);
9294: fprintf(ficresprobcor," p%1d-%1d ",i,j);
9295: }
9296: /* fprintf(ficresprob,"\n");
9297: fprintf(ficresprobcov,"\n");
9298: fprintf(ficresprobcor,"\n");
9299: */
9300: xp=vector(1,npar);
9301: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
9302: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
9303: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
9304: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
9305: first=1;
9306: fprintf(ficgp,"\n# Routine varprob");
9307: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
9308: fprintf(fichtm,"\n");
9309:
1.288 brouard 9310: fprintf(fichtm,"\n<li><h4> <a href=\"%s\">Matrix of variance-covariance of one-step probabilities (drawings)</a></h4> this page is important in order to visualize confidence intervals and especially correlation between disability and recovery, or more generally, way in and way back. File %s</li>\n",optionfilehtmcov,optionfilehtmcov);
1.222 brouard 9311: fprintf(fichtmcov,"Current page is file <a href=\"%s\">%s</a><br>\n\n<h4>Matrix of variance-covariance of pairs of step probabilities</h4>\n",optionfilehtmcov, optionfilehtmcov);
9312: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 9313: and drawn. It helps understanding how is the covariance between two incidences.\
9314: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 9315: fprintf(fichtmcov,"\n<br> Contour plot corresponding to x'cov<sup>-1</sup>x = 4 (where x is the column vector (pij,pkl)) are drawn. \
1.126 brouard 9316: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
9317: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
9318: standard deviations wide on each axis. <br>\
9319: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
9320: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
9321: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
9322:
1.222 brouard 9323: cov[1]=1;
9324: /* tj=cptcoveff; */
1.225 brouard 9325: tj = (int) pow(2,cptcoveff);
1.222 brouard 9326: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
9327: j1=0;
1.332 brouard 9328:
9329: for(nres=1;nres <=nresult; nres++){ /* For each resultline */
9330: for(j1=1; j1<=tj;j1++){ /* For any combination of dummy covariates, fixed and varying */
1.342 brouard 9331: /* printf("Varprob TKresult[nres]=%d j1=%d, nres=%d, cptcovn=%d, cptcoveff=%d tj=%d cptcovs=%d\n", TKresult[nres], j1, nres, cptcovn, cptcoveff, tj, cptcovs); */
1.332 brouard 9332: if(tj != 1 && TKresult[nres]!= j1)
9333: continue;
9334:
9335: /* for(j1=1; j1<=tj;j1++){ /\* For each valid combination of covariates or only once*\/ */
9336: /* for(nres=1;nres <=1; nres++){ /\* For each resultline *\/ */
9337: /* /\* for(nres=1;nres <=nresult; nres++){ /\\* For each resultline *\\/ *\/ */
1.222 brouard 9338: if (cptcovn>0) {
1.334 brouard 9339: fprintf(ficresprob, "\n#********** Variable ");
9340: fprintf(ficresprobcov, "\n#********** Variable ");
9341: fprintf(ficgp, "\n#********** Variable ");
9342: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
9343: fprintf(ficresprobcor, "\n#********** Variable ");
9344:
9345: /* Including quantitative variables of the resultline to be done */
9346: for (z1=1; z1<=cptcovs; z1++){ /* Loop on each variable of this resultline */
1.343 brouard 9347: /* printf("Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model); */
1.338 brouard 9348: fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model);
9349: /* fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=1+age+%s resultline[%d]=%s \n",nres, z1, modelresult[nres][z1], model, nres, resultline[nres]); */
1.334 brouard 9350: if(Dummy[modelresult[nres][z1]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to z1 in resultline */
9351: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed referenced to model equation */
9352: fprintf(ficresprob,"V%d=%d ",Tvresult[nres][z1],Tresult[nres][z1]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
9353: fprintf(ficresprobcov,"V%d=%d ",Tvresult[nres][z1],Tresult[nres][z1]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
9354: fprintf(ficgp,"V%d=%d ",Tvresult[nres][z1],Tresult[nres][z1]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
9355: fprintf(fichtmcov,"V%d=%d ",Tvresult[nres][z1],Tresult[nres][z1]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
9356: fprintf(ficresprobcor,"V%d=%d ",Tvresult[nres][z1],Tresult[nres][z1]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
9357: fprintf(ficresprob,"fixed ");
9358: fprintf(ficresprobcov,"fixed ");
9359: fprintf(ficgp,"fixed ");
9360: fprintf(fichtmcov,"fixed ");
9361: fprintf(ficresprobcor,"fixed ");
9362: }else{
9363: fprintf(ficresprob,"varyi ");
9364: fprintf(ficresprobcov,"varyi ");
9365: fprintf(ficgp,"varyi ");
9366: fprintf(fichtmcov,"varyi ");
9367: fprintf(ficresprobcor,"varyi ");
9368: }
9369: }else if(Dummy[modelresult[nres][z1]]==1){ /* Quanti variable */
9370: /* For each selected (single) quantitative value */
1.337 brouard 9371: fprintf(ficresprob," V%d=%lg ",Tvqresult[nres][z1],Tqresult[nres][z1]);
1.334 brouard 9372: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed */
9373: fprintf(ficresprob,"fixed ");
9374: fprintf(ficresprobcov,"fixed ");
9375: fprintf(ficgp,"fixed ");
9376: fprintf(fichtmcov,"fixed ");
9377: fprintf(ficresprobcor,"fixed ");
9378: }else{
9379: fprintf(ficresprob,"varyi ");
9380: fprintf(ficresprobcov,"varyi ");
9381: fprintf(ficgp,"varyi ");
9382: fprintf(fichtmcov,"varyi ");
9383: fprintf(ficresprobcor,"varyi ");
9384: }
9385: }else{
9386: printf("Error in varprob() Dummy[modelresult[%d][%d]]=%d, modelresult[%d][%d]=V%d cptcovs=%d, cptcoveff=%d \n", nres, z1, Dummy[modelresult[nres][z1]],nres,z1,modelresult[nres][z1],cptcovs, cptcoveff); /* end if dummy or quanti */
9387: fprintf(ficlog,"Error in varprob() Dummy[modelresult[%d][%d]]=%d, modelresult[%d][%d]=V%d cptcovs=%d, cptcoveff=%d \n", nres, z1, Dummy[modelresult[nres][z1]],nres,z1,modelresult[nres][z1],cptcovs, cptcoveff); /* end if dummy or quanti */
9388: exit(1);
9389: }
9390: } /* End loop on variable of this resultline */
9391: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]); */
1.222 brouard 9392: fprintf(ficresprob, "**********\n#\n");
9393: fprintf(ficresprobcov, "**********\n#\n");
9394: fprintf(ficgp, "**********\n#\n");
9395: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
9396: fprintf(ficresprobcor, "**********\n#");
9397: if(invalidvarcomb[j1]){
9398: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
9399: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
9400: continue;
9401: }
9402: }
9403: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
9404: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
9405: gp=vector(1,(nlstate)*(nlstate+ndeath));
9406: gm=vector(1,(nlstate)*(nlstate+ndeath));
1.334 brouard 9407: for (age=bage; age<=fage; age ++){ /* Fo each age we feed the model equation with covariates, using precov as in hpxij() ? */
1.222 brouard 9408: cov[2]=age;
9409: if(nagesqr==1)
9410: cov[3]= age*age;
1.334 brouard 9411: /* New code end of combination but for each resultline */
9412: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.349 brouard 9413: if(Typevar[k1]==1 || Typevar[k1] ==3){ /* A product with age */
1.334 brouard 9414: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.326 brouard 9415: }else{
1.334 brouard 9416: cov[2+nagesqr+k1]=precov[nres][k1];
1.326 brouard 9417: }
1.334 brouard 9418: }/* End of loop on model equation */
9419: /* Old code */
9420: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
9421: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; *\/ */
9422: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
9423: /* /\* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates *\/ */
9424: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TnsdVar[TvarsD[k]])]; */
9425: /* /\*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*\//\* j1 1 2 3 4 */
9426: /* * 1 1 1 1 1 */
9427: /* * 2 2 1 1 1 */
9428: /* * 3 1 2 1 1 */
9429: /* *\/ */
9430: /* /\* nbcode[1][1]=0 nbcode[1][2]=1;*\/ */
9431: /* } */
9432: /* /\* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
9433: /* /\* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] *\/ */
9434: /* /\*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; *\/ */
9435: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
9436: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
9437: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,TnsdVar[Tvar[Tage[k]]])]*cov[2]; */
9438: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
9439: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
9440: /* printf("Internal IMaCh error, don't know which value for quantitative covariate with age, Tage[k]%d, k=%d, Tvar[Tage[k]]=V%d, age=%d\n",Tage[k],k ,Tvar[Tage[k]], (int)cov[2]); */
9441: /* /\* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\\* Using the mean of quantitative variable Tvar[Tage[k]] /\\* Tqresult[nres][k]; *\\/ *\/ */
9442: /* /\* exit(1); *\/ */
9443: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
9444: /* } */
9445: /* /\* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
9446: /* } */
9447: /* for (k=1; k<=cptcovprod;k++){/\* For product without age *\/ */
9448: /* if(Dummy[Tvard[k][1]]==0){ */
9449: /* if(Dummy[Tvard[k][2]]==0){ */
9450: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])]; */
9451: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
9452: /* }else{ /\* Should we use the mean of the quantitative variables? *\/ */
9453: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * Tqresult[nres][resultmodel[nres][k]]; */
9454: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
9455: /* } */
9456: /* }else{ */
9457: /* if(Dummy[Tvard[k][2]]==0){ */
9458: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][TnsdVar[Tvard[k][1]]]; */
9459: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
9460: /* }else{ */
9461: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][TnsdVar[Tvard[k][1]]]* Tqinvresult[nres][TnsdVar[Tvard[k][2]]]; */
9462: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
9463: /* } */
9464: /* } */
9465: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
9466: /* } */
1.326 brouard 9467: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 9468: for(theta=1; theta <=npar; theta++){
9469: for(i=1; i<=npar; i++)
9470: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 9471:
1.222 brouard 9472: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 9473:
1.222 brouard 9474: k=0;
9475: for(i=1; i<= (nlstate); i++){
9476: for(j=1; j<=(nlstate+ndeath);j++){
9477: k=k+1;
9478: gp[k]=pmmij[i][j];
9479: }
9480: }
1.220 brouard 9481:
1.222 brouard 9482: for(i=1; i<=npar; i++)
9483: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 9484:
1.222 brouard 9485: pmij(pmmij,cov,ncovmodel,xp,nlstate);
9486: k=0;
9487: for(i=1; i<=(nlstate); i++){
9488: for(j=1; j<=(nlstate+ndeath);j++){
9489: k=k+1;
9490: gm[k]=pmmij[i][j];
9491: }
9492: }
1.220 brouard 9493:
1.222 brouard 9494: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
9495: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
9496: }
1.126 brouard 9497:
1.222 brouard 9498: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
9499: for(theta=1; theta <=npar; theta++)
9500: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 9501:
1.222 brouard 9502: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
9503: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 9504:
1.222 brouard 9505: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 9506:
1.222 brouard 9507: k=0;
9508: for(i=1; i<=(nlstate); i++){
9509: for(j=1; j<=(nlstate+ndeath);j++){
9510: k=k+1;
9511: mu[k][(int) age]=pmmij[i][j];
9512: }
9513: }
9514: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
9515: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
9516: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 9517:
1.222 brouard 9518: /*printf("\n%d ",(int)age);
9519: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
9520: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
9521: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
9522: }*/
1.220 brouard 9523:
1.222 brouard 9524: fprintf(ficresprob,"\n%d ",(int)age);
9525: fprintf(ficresprobcov,"\n%d ",(int)age);
9526: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 9527:
1.222 brouard 9528: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
9529: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
9530: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
9531: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
9532: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
9533: }
9534: i=0;
9535: for (k=1; k<=(nlstate);k++){
9536: for (l=1; l<=(nlstate+ndeath);l++){
9537: i++;
9538: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
9539: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
9540: for (j=1; j<=i;j++){
9541: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
9542: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
9543: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
9544: }
9545: }
9546: }/* end of loop for state */
9547: } /* end of loop for age */
9548: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
9549: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
9550: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
9551: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
9552:
9553: /* Confidence intervalle of pij */
9554: /*
9555: fprintf(ficgp,"\nunset parametric;unset label");
9556: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
9557: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
9558: fprintf(fichtm,"\n<br>Probability with confidence intervals expressed in year<sup>-1</sup> :<a href=\"pijgr%s.png\">pijgr%s.png</A>, ",optionfilefiname,optionfilefiname);
9559: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
9560: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
9561: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
9562: */
9563:
9564: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
9565: first1=1;first2=2;
9566: for (k2=1; k2<=(nlstate);k2++){
9567: for (l2=1; l2<=(nlstate+ndeath);l2++){
9568: if(l2==k2) continue;
9569: j=(k2-1)*(nlstate+ndeath)+l2;
9570: for (k1=1; k1<=(nlstate);k1++){
9571: for (l1=1; l1<=(nlstate+ndeath);l1++){
9572: if(l1==k1) continue;
9573: i=(k1-1)*(nlstate+ndeath)+l1;
9574: if(i<=j) continue;
9575: for (age=bage; age<=fage; age ++){
9576: if ((int)age %5==0){
9577: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
9578: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
9579: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
9580: mu1=mu[i][(int) age]/stepm*YEARM ;
9581: mu2=mu[j][(int) age]/stepm*YEARM;
9582: c12=cv12/sqrt(v1*v2);
9583: /* Computing eigen value of matrix of covariance */
9584: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
9585: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
9586: if ((lc2 <0) || (lc1 <0) ){
9587: if(first2==1){
9588: first1=0;
9589: printf("Strange: j1=%d One eigen value of 2x2 matrix of covariance is negative, lc1=%11.3e, lc2=%11.3e, v1=%11.3e, v2=%11.3e, cv12=%11.3e.\n It means that the matrix was not well estimated (varpij), for i=%2d, j=%2d, age=%4d .\n See files %s and %s. Probably WRONG RESULTS. See log file for details...\n", j1, lc1, lc2, v1, v2, cv12, i, j, (int)age,fileresprobcov, fileresprobcor);
9590: }
9591: fprintf(ficlog,"Strange: j1=%d One eigen value of 2x2 matrix of covariance is negative, lc1=%11.3e, lc2=%11.3e, v1=%11.3e, v2=%11.3e, cv12=%11.3e.\n It means that the matrix was not well estimated (varpij), for i=%2d, j=%2d, age=%4d .\n See files %s and %s. Probably WRONG RESULTS.\n", j1, lc1, lc2, v1, v2, cv12, i, j, (int)age,fileresprobcov, fileresprobcor);fflush(ficlog);
9592: /* lc1=fabs(lc1); */ /* If we want to have them positive */
9593: /* lc2=fabs(lc2); */
9594: }
1.220 brouard 9595:
1.222 brouard 9596: /* Eigen vectors */
1.280 brouard 9597: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
9598: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
9599: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
9600: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
9601: }else
9602: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 9603: /*v21=sqrt(1.-v11*v11); *//* error */
9604: v21=(lc1-v1)/cv12*v11;
9605: v12=-v21;
9606: v22=v11;
9607: tnalp=v21/v11;
9608: if(first1==1){
9609: first1=0;
9610: printf("%d %d%d-%d%d mu %.4e %.4e Var %.4e %.4e cor %.3f cov %.4e Eig %.3e %.3e 1stv %.3f %.3f tang %.3f\nOthers in log...\n",(int) age,k1,l1,k2,l2,mu1,mu2,v1,v2,c12,cv12,lc1,lc2,v11,v21,tnalp);
9611: }
9612: fprintf(ficlog,"%d %d%d-%d%d mu %.4e %.4e Var %.4e %.4e cor %.3f cov %.4e Eig %.3e %.3e 1stv %.3f %.3f tan %.3f\n",(int) age,k1,l1,k2,l2,mu1,mu2,v1,v2,c12,cv12,lc1,lc2,v11,v21,tnalp);
9613: /*printf(fignu*/
9614: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
9615: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
9616: if(first==1){
9617: first=0;
9618: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
9619: fprintf(ficgp,"\nset parametric;unset label");
9620: fprintf(ficgp,"\nset log y;set log x; set xlabel \"p%1d%1d (year-1)\";set ylabel \"p%1d%1d (year-1)\"",k1,l1,k2,l2);
9621: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 9622: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 9623: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 9624: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 9625: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
9626: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
9627: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
9628: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
9629: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
9630: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
9631: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
9632: fprintf(ficgp,"\nplot [-pi:pi] %11.3e+ %.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)), %11.3e +%.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)) not", \
1.280 brouard 9633: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
9634: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 9635: }else{
9636: first=0;
9637: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
9638: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
9639: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
9640: fprintf(ficgp,"\nreplot %11.3e+ %.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)), %11.3e +%.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)) not", \
1.266 brouard 9641: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
9642: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 9643: }/* if first */
9644: } /* age mod 5 */
9645: } /* end loop age */
9646: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
9647: first=1;
9648: } /*l12 */
9649: } /* k12 */
9650: } /*l1 */
9651: }/* k1 */
1.332 brouard 9652: } /* loop on combination of covariates j1 */
1.326 brouard 9653: } /* loop on nres */
1.222 brouard 9654: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
9655: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
9656: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
9657: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
9658: free_vector(xp,1,npar);
9659: fclose(ficresprob);
9660: fclose(ficresprobcov);
9661: fclose(ficresprobcor);
9662: fflush(ficgp);
9663: fflush(fichtmcov);
9664: }
1.126 brouard 9665:
9666:
9667: /******************* Printing html file ***********/
1.201 brouard 9668: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9669: int lastpass, int stepm, int weightopt, char model[],\
9670: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 9671: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
9672: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
9673: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.359 brouard 9674: int jj1, k1, cpt, nres;
1.319 brouard 9675: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 9676: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
9677: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
9678: </ul>");
1.319 brouard 9679: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
9680: /* </ul>", model); */
1.214 brouard 9681: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
9682: fprintf(fichtm,"<li>- Observed frequency between two states (during the period defined between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf): <a href=\"%s\">%s</a> (html file)<br/>\n",
9683: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
1.332 brouard 9684: fprintf(fichtm,"<li> - Observed prevalence (cross-sectional prevalence) in each state (during the period defined between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf): <a href=\"%s\">%s</a> (html file) ",
1.213 brouard 9685: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
9686: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 9687: fprintf(fichtm,"\
9688: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 9689: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 9690: fprintf(fichtm,"\
1.217 brouard 9691: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
9692: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
9693: fprintf(fichtm,"\
1.288 brouard 9694: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 9695: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 9696: fprintf(fichtm,"\
1.288 brouard 9697: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 9698: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
9699: fprintf(fichtm,"\
1.211 brouard 9700: - (a) Life expectancies by health status at initial age, e<sub>i.</sub> (b) health expectancies by health status at initial age, e<sub>ij</sub> . If one or more covariates are included, specific tables for each value of the covariate are output in sequences within the same file (estepm=%2d months): \
1.126 brouard 9701: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 9702: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 9703: if(prevfcast==1){
9704: fprintf(fichtm,"\
9705: - Prevalence projections by age and states: \
1.201 brouard 9706: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 9707: }
1.126 brouard 9708:
9709:
1.225 brouard 9710: m=pow(2,cptcoveff);
1.222 brouard 9711: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 9712:
1.317 brouard 9713: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 9714:
9715: jj1=0;
9716:
9717: fprintf(fichtm," \n<ul>");
1.337 brouard 9718: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9719: /* k1=nres; */
1.338 brouard 9720: k1=TKresult[nres];
9721: if(TKresult[nres]==0)k1=1; /* To be checked for no result */
1.337 brouard 9722: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
9723: /* if(m != 1 && TKresult[nres]!= k1) */
9724: /* continue; */
1.264 brouard 9725: jj1++;
9726: if (cptcovn > 0) {
9727: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
1.337 brouard 9728: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
9729: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 9730: }
1.337 brouard 9731: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
9732: /* fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
9733: /* } */
9734: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
9735: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
9736: /* } */
1.264 brouard 9737: fprintf(fichtm,"\">");
9738:
9739: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
9740: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 9741: for (cpt=1; cpt<=cptcovs;cpt++){
9742: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 9743: }
1.337 brouard 9744: /* fprintf(fichtm,"************ Results for covariates"); */
9745: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
9746: /* fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
9747: /* } */
9748: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
9749: /* fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
9750: /* } */
1.264 brouard 9751: if(invalidvarcomb[k1]){
9752: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
9753: continue;
9754: }
9755: fprintf(fichtm,"</a></li>");
9756: } /* cptcovn >0 */
9757: }
1.317 brouard 9758: fprintf(fichtm," \n</ul>");
1.264 brouard 9759:
1.222 brouard 9760: jj1=0;
1.237 brouard 9761:
1.337 brouard 9762: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9763: /* k1=nres; */
1.338 brouard 9764: k1=TKresult[nres];
9765: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 9766: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
9767: /* if(m != 1 && TKresult[nres]!= k1) */
9768: /* continue; */
1.220 brouard 9769:
1.222 brouard 9770: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
9771: jj1++;
9772: if (cptcovn > 0) {
1.264 brouard 9773: fprintf(fichtm,"\n<p><a name=\"rescov");
1.337 brouard 9774: for (cpt=1; cpt<=cptcovs;cpt++){
9775: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 9776: }
1.337 brouard 9777: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
9778: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
9779: /* } */
1.264 brouard 9780: fprintf(fichtm,"\"</a>");
9781:
1.222 brouard 9782: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 9783: for (cpt=1; cpt<=cptcovs;cpt++){
9784: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
9785: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 9786: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
9787: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 9788: }
1.230 brouard 9789: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.338 brouard 9790: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 9791: if(invalidvarcomb[k1]){
9792: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
9793: printf("\nCombination (%d) ignored because no cases \n",k1);
9794: continue;
9795: }
9796: }
9797: /* aij, bij */
1.259 brouard 9798: fprintf(fichtm,"<br>- Logit model (yours is: logit(pij)=log(pij/pii)= aij+ bij age+%s) as a function of age: <a href=\"%s_%d-1-%d.svg\">%s_%d-1-%d.svg</a><br> \
1.241 brouard 9799: <img src=\"%s_%d-1-%d.svg\">",model,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres);
1.222 brouard 9800: /* Pij */
1.241 brouard 9801: fprintf(fichtm,"<br>\n- P<sub>ij</sub> or conditional probabilities to be observed in state j being in state i, %d (stepm) months before: <a href=\"%s_%d-2-%d.svg\">%s_%d-2-%d.svg</a><br> \
9802: <img src=\"%s_%d-2-%d.svg\">",stepm,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres);
1.222 brouard 9803: /* Quasi-incidences */
9804: fprintf(fichtm,"<br>\n- I<sub>ij</sub> or Conditional probabilities to be observed in state j being in state i %d (stepm) months\
1.220 brouard 9805: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 9806: incidence (rates) are the limit when h tends to zero of the ratio of the probability <sub>h</sub>P<sub>ij</sub> \
1.241 brouard 9807: divided by h: <sub>h</sub>P<sub>ij</sub>/h : <a href=\"%s_%d-3-%d.svg\">%s_%d-3-%d.svg</a><br> \
9808: <img src=\"%s_%d-3-%d.svg\">",stepm,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres);
1.222 brouard 9809: /* Survival functions (period) in state j */
9810: for(cpt=1; cpt<=nlstate;cpt++){
1.359 brouard 9811: fprintf(fichtm,"<br>\n- Survival functions in state %d. And probability to be observed in state %d being in state (1 to %d) at different ages. Mean times spent in state (or Life Expectancy or Health Expectancy etc.) are the areas under each curve. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, cpt, nlstate, subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.329 brouard 9812: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
9813: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 9814: }
9815: /* State specific survival functions (period) */
9816: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 9817: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
1.359 brouard 9818: And probability to be observed in various states (up to %d) being in state %d at different ages. Mean times spent in state (or Life Expectancy or Health Expectancy etc.) are the areas under each curve. \
1.329 brouard 9819: <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> ", cpt, nlstate, cpt, subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
9820: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
9821: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 9822: }
1.288 brouard 9823: /* Period (forward stable) prevalence in each health state */
1.222 brouard 9824: for(cpt=1; cpt<=nlstate;cpt++){
1.359 brouard 9825: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability for a person being in state (1 to %d) at different ages, to be alive in state %d some years after. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, nlstate, cpt, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.338 brouard 9826: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
1.329 brouard 9827: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 9828: }
1.296 brouard 9829: if(prevbcast==1){
1.288 brouard 9830: /* Backward prevalence in each health state */
1.222 brouard 9831: for(cpt=1; cpt<=nlstate;cpt++){
1.338 brouard 9832: fprintf(fichtm,"<br>\n- Convergence to mixed (stable) back prevalence in state %d. Or probability for a person to be in state %d at a younger age, knowing that she/he was in state (1 to %d) at different older ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, cpt, nlstate, subdirf2(optionfilefiname,"PB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
9833: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJB_"),subdirf2(optionfilefiname,"PIJB_"));
9834: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.222 brouard 9835: }
1.217 brouard 9836: }
1.222 brouard 9837: if(prevfcast==1){
1.288 brouard 9838: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 9839: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 9840: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), from year %.1f up to year %.1f tending to period (stable) forward prevalence in state %d. Or probability to be in state %d being in an observed weighted state (from 1 to %d). <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a>", dateprev1, dateprev2, mobilavproj, dateprojd, dateprojf, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
9841: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
9842: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
9843: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 9844: }
9845: }
1.296 brouard 9846: if(prevbcast==1){
1.268 brouard 9847: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
9848: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 9849: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
1.359 brouard 9850: from year %.1f up to year %.1f (probably close to stable [mixed] back prevalence in state %d). Randomness in cross-sectional prevalence is not taken into \
9851: account but can visually be appreciated. Or probability to have been in an state %d, knowing that the person was in either state (1 or %d) \
1.314 brouard 9852: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a>", dateprev1, dateprev2, mobilavproj, dateback1, dateback2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
9853: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
9854: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 9855: }
9856: }
1.220 brouard 9857:
1.222 brouard 9858: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 9859: fprintf(fichtm,"\n<br>- Life expectancy by health state (%d) at initial age and its decomposition into health expectancies in each alive state (1 to %d) (or area under each survival functions): <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a>",cpt,nlstate,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
9860: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
9861: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 9862: }
9863: /* } /\* end i1 *\/ */
1.337 brouard 9864: }/* End k1=nres */
1.222 brouard 9865: fprintf(fichtm,"</ul>");
1.126 brouard 9866:
1.222 brouard 9867: fprintf(fichtm,"\
1.126 brouard 9868: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 9869: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 9870: - 95%% confidence intervals and Wald tests of the estimated parameters are in the log file if optimization has been done (mle != 0).<br> \
1.197 brouard 9871: But because parameters are usually highly correlated (a higher incidence of disability \
9872: and a higher incidence of recovery can give very close observed transition) it might \
9873: be very useful to look not only at linear confidence intervals estimated from the \
9874: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
9875: (parameters) of the logistic regression, it might be more meaningful to visualize the \
9876: covariance matrix of the one-step probabilities. \
9877: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 9878:
1.222 brouard 9879: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
9880: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
9881: fprintf(fichtm,"\
1.126 brouard 9882: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 9883: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 9884:
1.222 brouard 9885: fprintf(fichtm,"\
1.126 brouard 9886: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 9887: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
9888: fprintf(fichtm,"\
1.126 brouard 9889: - Variances and covariances of health expectancies by age and <b>initial health status</b> (cov(e<sup>ij</sup>,e<sup>kl</sup>)(estepm=%2d months): \
9890: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 9891: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 9892: fprintf(fichtm,"\
1.126 brouard 9893: - (a) Health expectancies by health status at initial age (e<sup>ij</sup>) and standard errors (in parentheses) (b) life expectancies and standard errors (e<sup>i.</sup>=e<sup>i1</sup>+e<sup>i2</sup>+...)(estepm=%2d months): \
9894: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 9895: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 9896: fprintf(fichtm,"\
1.288 brouard 9897: - Variances and covariances of health expectancies by age. Status (i) based health expectancies (in state j), e<sup>ij</sup> are weighted by the forward (period) prevalences in each state i (if popbased=1, an additional computation is done using the cross-sectional prevalences, i.e population based) (estepm=%d months): <a href=\"%s\">%s</a><br>\n",
1.222 brouard 9898: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
9899: fprintf(fichtm,"\
1.128 brouard 9900: - Total life expectancy and total health expectancies to be spent in each health state e<sup>.j</sup> with their standard errors (if popbased=1, an additional computation is done using the cross-sectional prevalences, i.e population based) (estepm=%d months): <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 9901: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
9902: fprintf(fichtm,"\
1.288 brouard 9903: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 9904: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 9905:
9906: /* if(popforecast==1) fprintf(fichtm,"\n */
9907: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
9908: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
9909: /* <br>",fileres,fileres,fileres,fileres); */
9910: /* else */
1.338 brouard 9911: /* fprintf(fichtm,"\n No population forecast: popforecast = %d (instead of 1) or stepm = %d (instead of 1) or model=1+age+%s (instead of .)<br><br></li>\n",popforecast, stepm, model); */
1.222 brouard 9912: fflush(fichtm);
1.126 brouard 9913:
1.225 brouard 9914: m=pow(2,cptcoveff);
1.222 brouard 9915: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 9916:
1.317 brouard 9917: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
9918:
9919: jj1=0;
9920:
9921: fprintf(fichtm," \n<ul>");
1.337 brouard 9922: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9923: /* k1=nres; */
1.338 brouard 9924: k1=TKresult[nres];
1.337 brouard 9925: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
9926: /* if(m != 1 && TKresult[nres]!= k1) */
9927: /* continue; */
1.317 brouard 9928: jj1++;
9929: if (cptcovn > 0) {
9930: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
1.337 brouard 9931: for (cpt=1; cpt<=cptcovs;cpt++){
9932: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 9933: }
9934: fprintf(fichtm,"\">");
9935:
9936: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
9937: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 9938: for (cpt=1; cpt<=cptcovs;cpt++){
9939: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 9940: }
9941: if(invalidvarcomb[k1]){
9942: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
9943: continue;
9944: }
9945: fprintf(fichtm,"</a></li>");
9946: } /* cptcovn >0 */
1.337 brouard 9947: } /* End nres */
1.317 brouard 9948: fprintf(fichtm," \n</ul>");
9949:
1.222 brouard 9950: jj1=0;
1.237 brouard 9951:
1.241 brouard 9952: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 9953: /* k1=nres; */
1.338 brouard 9954: k1=TKresult[nres];
9955: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 9956: /* for(k1=1; k1<=m;k1++){ */
9957: /* if(m != 1 && TKresult[nres]!= k1) */
9958: /* continue; */
1.222 brouard 9959: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
9960: jj1++;
1.126 brouard 9961: if (cptcovn > 0) {
1.317 brouard 9962: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
1.337 brouard 9963: for (cpt=1; cpt<=cptcovs;cpt++){
9964: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 9965: }
9966: fprintf(fichtm,"\"</a>");
9967:
1.126 brouard 9968: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 9969: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcoveff number of variables */
9970: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
9971: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 9972: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 9973: }
1.237 brouard 9974:
1.338 brouard 9975: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 9976:
1.222 brouard 9977: if(invalidvarcomb[k1]){
9978: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
9979: continue;
9980: }
1.337 brouard 9981: } /* If cptcovn >0 */
1.126 brouard 9982: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 9983: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 9984: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>",mobilav,cpt,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
9985: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
9986: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 9987: }
9988: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.360 brouard 9989: health expectancies in each live state (1 to %d) with confidence intervals \
9990: on left y-scale as well as proportions of time spent in each live state \
9991: (with confidence intervals) on right y-scale 0 to 100%%.\
9992: If popbased=1 the smooth (due to the model) \
1.128 brouard 9993: true period expectancies (those weighted with period prevalences are also\
9994: drawn in addition to the population based expectancies computed using\
1.314 brouard 9995: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>",nlstate, subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
9996: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
9997: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 9998: /* } /\* end i1 *\/ */
1.241 brouard 9999: }/* End nres */
1.222 brouard 10000: fprintf(fichtm,"</ul>");
10001: fflush(fichtm);
1.126 brouard 10002: }
10003:
10004: /******************* Gnuplot file **************/
1.296 brouard 10005: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double bage, double fage , int prevfcast, int prevbcast, char pathc[], double p[], int offyear, int offbyear){
1.126 brouard 10006:
1.354 brouard 10007: char dirfileres[256],optfileres[256];
10008: char gplotcondition[256], gplotlabel[256];
1.343 brouard 10009: int cpt=0,k1=0,i=0,k=0,j=0,jk=0,k2=0,k3=0,k4=0,kf=0,kvar=0,kk=0,ipos=0,iposold=0,ij=0, ijp=0, l=0;
1.365 brouard 10010: /* int lv=0, vlv=0, kl=0; */
10011: int lv=0, kl=0;
10012: double vlv=0;
1.130 brouard 10013: int ng=0;
1.201 brouard 10014: int vpopbased;
1.223 brouard 10015: int ioffset; /* variable offset for columns */
1.270 brouard 10016: int iyearc=1; /* variable column for year of projection */
10017: int iagec=1; /* variable column for age of projection */
1.235 brouard 10018: int nres=0; /* Index of resultline */
1.266 brouard 10019: int istart=1; /* For starting graphs in projections */
1.219 brouard 10020:
1.126 brouard 10021: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
10022: /* printf("Problem with file %s",optionfilegnuplot); */
10023: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
10024: /* } */
10025:
10026: /*#ifdef windows */
10027: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 10028: /*#endif */
1.225 brouard 10029: m=pow(2,cptcoveff);
1.126 brouard 10030:
1.274 brouard 10031: /* diagram of the model */
10032: fprintf(ficgp,"\n#Diagram of the model \n");
10033: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
10034: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
10035: fprintf(ficgp,"\n#Peripheral arrows\nset for [i=1:%d] for [j=1:%d] arrow i*10+j from cos(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d))-(i!=j?(i-j)/abs(i-j)*delta:0), yoff +sin(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) rto -0.95*(cos(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d))+(i!=j?(i-j)/abs(i-j)*delta:0) - cos(pi*((1-(%d/2)*2./%d)/2+(j-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta2:0)), -0.95*(sin(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) - sin(pi*((1-(%d/2)*2./%d)/2+(j-1)*2./%d))+( i!=j?(i-j)/abs(i-j)*delta2:0)) ls (i < j? 1:2)\n",nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
10036:
1.343 brouard 10037: fprintf(ficgp,"\n#Centripete arrows (turning in other direction (1-i) instead of (i-1)) \nset for [i=1:%d] for [j=1:%d] arrow (%d+1)*10+i from cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))-(i!=j?(i-j)/abs(i-j)*delta:0), yoff +sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) rto -0.80*(cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))+(i!=j?(i-j)/abs(i-j)*delta:0) ), -0.80*(sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) + yoff ) ls 4\n",nlstate, nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
1.274 brouard 10038: fprintf(ficgp,"\n#show arrow\nunset label\n");
10039: fprintf(ficgp,"\n#States labels, starting from 2 (2-i) instead of (1-i), was (i-1)\nset for [i=1:%d] label i sprintf(\"State %%d\",i) center at cos(pi*((1-(%d/2)*2./%d)/2+(2-i)*2./%d)), yoff+sin(pi*((1-(%d/2)*2./%d)/2+(2-i)*2./%d)) font \"helvetica, 16\" tc rgbcolor \"blue\"\n",nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
10040: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
10041: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
10042: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
10043: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
10044:
1.202 brouard 10045: /* Contribution to likelihood */
10046: /* Plot the probability implied in the likelihood */
1.223 brouard 10047: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
10048: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
10049: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
10050: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 10051: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 10052: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
10053: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 10054: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
10055: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
10056: fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$13):6 t \"All sample, transitions colored by destination\" with dots lc variable; set out;\n",subdirf(fileresilk));
10057: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
10058: fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$13):5 t \"All sample, transitions colored by origin\" with dots lc variable; set out;\n\n",subdirf(fileresilk));
10059: for (i=1; i<= nlstate ; i ++) {
10060: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
10061: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
10062: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($12/4.):6 t \"p%d%d\" with points pointtype 7 ps variable lc variable \\\n",i,1,i,1);
10063: for (j=2; j<= nlstate+ndeath ; j ++) {
10064: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($12/4.):6 t \"p%d%d\" with points pointtype 7 ps variable lc variable ",i,j,i,j);
10065: }
10066: fprintf(ficgp,";\nset out; unset ylabel;\n");
10067: }
10068: /* unset log; plot "rrtest1_sorted_4/ILK_rrtest1_sorted_4.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with points lc variable */
10069: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
10070: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
10071: fprintf(ficgp,"\nset out;unset log\n");
10072: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 10073:
1.343 brouard 10074: /* Plot the probability implied in the likelihood by covariate value */
10075: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
10076: /* if(debugILK==1){ */
10077: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
1.347 brouard 10078: kvar=Tvar[TvarFind[kf]]; /* variable name */
10079: /* k=18+Tvar[TvarFind[kf]];/\*offset because there are 18 columns in the ILK_ file but could be placed else where *\/ */
1.350 brouard 10080: /* k=18+kf;/\*offset because there are 18 columns in the ILK_ file *\/ */
1.356 brouard 10081: /* k=19+kf;/\*offset because there are 19 columns in the ILK_ file *\/ */
1.355 brouard 10082: k=16+nlstate+kf;/*offset because there are 19 columns in the ILK_ file, first cov Vn on col 21 with 4 living states */
1.343 brouard 10083: for (i=1; i<= nlstate ; i ++) {
10084: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
10085: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
1.348 brouard 10086: if(gnuplotversion >=5.2){ /* Former gnuplot versions do not have variable pointsize!! */
10087: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable \\\n",i,1,k,k,i,1,kvar);
10088: for (j=2; j<= nlstate+ndeath ; j ++) {
10089: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable ",i,j,k,k,i,j,kvar);
10090: }
10091: }else{
10092: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt 7 ps 0.4 lc variable \\\n",i,1,k,i,1,kvar);
10093: for (j=2; j<= nlstate+ndeath ; j ++) {
10094: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt 7 ps 0.4 lc variable ",i,j,k,i,j,kvar);
10095: }
1.343 brouard 10096: }
10097: fprintf(ficgp,";\nset out; unset ylabel;\n");
10098: }
10099: } /* End of each covariate dummy */
10100: for(ncovv=1, iposold=0, kk=0; ncovv <= ncovvt ; ncovv++){
10101: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
10102: * kmodel = 1 2 3 4 5 6 7 8 9
10103: * varying 1 2 3 4 5
10104: * ncovv 1 2 3 4 5 6 7 8
10105: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
10106: * TvarVVind[ncovv]=kmodel 2 3 7 7 8 8 9 9
10107: * TvarFind[kmodel] 1 0 0 0 0 0 0 0 0
10108: * kdata ncovcol=[V1 V2] nqv=0 ntv=[V3 V4] nqtv=V5
10109: * Dummy[kmodel] 0 0 1 2 2 3 1 1 1
10110: */
10111: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
10112: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
10113: /* printf("DebugILK ficgp ncovv=%d, kvar=TvarVV[ncovv]=%d, ipos=TvarVVind[ncovv]=%d, Dummy[ipos]=%d, Typevar[ipos]=%d\n", ncovv,kvar,ipos,Dummy[ipos],Typevar[ipos]); */
10114: if(ipos!=iposold){ /* Not a product or first of a product */
10115: /* printf(" %d",ipos); */
10116: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
10117: /* printf(" DebugILK ficgp suite ipos=%d != iposold=%d\n", ipos, iposold); */
10118: kk++; /* Position of the ncovv column in ILK_ */
10119: k=18+ncovf+kk; /*offset because there are 18 columns in the ILK_ file plus ncovf fixed covariate */
10120: if(Dummy[ipos]==0 && Typevar[ipos]==0){ /* Only if dummy time varying: Dummy(0, 1=quant singor prod without age,2 dummy*age, 3quant*age) Typevar (0 single, 1=*age,2=Vn*vm) */
10121: for (i=1; i<= nlstate ; i ++) {
10122: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
10123: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
10124:
1.348 brouard 10125: /* printf("Before DebugILK gnuplotversion=%g >=5.2\n",gnuplotversion); */
1.343 brouard 10126: if(gnuplotversion >=5.2){ /* Former gnuplot versions do not have variable pointsize!! */
10127: /* printf("DebugILK gnuplotversion=%g >=5.2\n",gnuplotversion); */
10128: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable \\\n",i,1,k,k,i,1,kvar);
10129: for (j=2; j<= nlstate+ndeath ; j ++) {
10130: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable ",i,j,k,k,i,j,kvar);
10131: }
10132: }else{
10133: /* printf("DebugILK gnuplotversion=%g <5.2\n",gnuplotversion); */
10134: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt 7 ps 0.4 lc variable \\\n",i,1,k,i,1,kvar);
10135: for (j=2; j<= nlstate+ndeath ; j ++) {
10136: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt 7 ps 0.4 lc variable ",i,j,k,i,j,kvar);
10137: }
10138: }
10139: fprintf(ficgp,";\nset out; unset ylabel;\n");
10140: }
10141: }/* End if dummy varying */
10142: }else{ /*Product */
10143: /* printf("*"); */
10144: /* fprintf(ficresilk,"*"); */
10145: }
10146: iposold=ipos;
10147: } /* For each time varying covariate */
10148: /* } /\* debugILK==1 *\/ */
10149: /* unset log; plot "rrtest1_sorted_4/ILK_rrtest1_sorted_4.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with points lc variable */
10150: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
10151: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
10152: fprintf(ficgp,"\nset out;unset log\n");
10153: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
10154:
10155:
10156:
1.126 brouard 10157: strcpy(dirfileres,optionfilefiname);
10158: strcpy(optfileres,"vpl");
1.223 brouard 10159: /* 1eme*/
1.238 brouard 10160: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
1.337 brouard 10161: /* for (k1=1; k1<= m ; k1 ++){ /\* For each valid combination of covariate *\/ */
1.236 brouard 10162: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10163: k1=TKresult[nres];
1.338 brouard 10164: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.238 brouard 10165: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.337 brouard 10166: /* if(m != 1 && TKresult[nres]!= k1) */
10167: /* continue; */
1.238 brouard 10168: /* We are interested in selected combination by the resultline */
1.246 brouard 10169: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 10170: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 10171: strcpy(gplotlabel,"(");
1.337 brouard 10172: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10173: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10174: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10175:
10176: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate k get corresponding value lv for combination k1 *\/ */
10177: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the value of the covariate corresponding to k1 combination *\\/ *\/ */
10178: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10179: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10180: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10181: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10182: /* vlv= nbcode[Tvaraff[k]][lv]; /\* vlv is the value of the covariate lv, 0 or 1 *\/ */
10183: /* /\* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv *\/ */
10184: /* /\* printf(" V%d=%d ",Tvaraff[k],vlv); *\/ */
10185: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10186: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10187: /* } */
10188: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10189: /* /\* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); *\/ */
10190: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
10191: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.264 brouard 10192: }
10193: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 10194: /* printf("\n#\n"); */
1.238 brouard 10195: fprintf(ficgp,"\n#\n");
10196: if(invalidvarcomb[k1]){
1.260 brouard 10197: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 10198: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10199: continue;
10200: }
1.235 brouard 10201:
1.241 brouard 10202: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
10203: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 10204: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
1.338 brouard 10205: fprintf(ficgp,"set title \"Alive state %d %s model=1+age+%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.367 ! brouard 10206: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \nset ter svg size 640, 480\nplot [%.f:%.f] [0:1] \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",ageminpar,fage,subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.260 brouard 10207: /* fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \nset ter svg size 640, 480\nplot [%.f:%.f] \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",ageminpar,fage,subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres); */
10208: /* k1-1 error should be nres-1*/
1.238 brouard 10209: for (i=1; i<= nlstate ; i ++) {
10210: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10211: else fprintf(ficgp," %%*lf (%%*lf)");
10212: }
1.288 brouard 10213: fprintf(ficgp,"\" t\"Forward prevalence\" w l lt 0,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 10214: for (i=1; i<= nlstate ; i ++) {
10215: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10216: else fprintf(ficgp," %%*lf (%%*lf)");
10217: }
1.260 brouard 10218: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 10219: for (i=1; i<= nlstate ; i ++) {
10220: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10221: else fprintf(ficgp," %%*lf (%%*lf)");
10222: }
1.265 brouard 10223: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
10224:
10225: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
10226: if(cptcoveff ==0){
1.271 brouard 10227: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 10228: }else{
10229: kl=0;
10230: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 10231: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
10232: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.265 brouard 10233: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
10234: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
10235: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
10236: vlv= nbcode[Tvaraff[k]][lv];
10237: kl++;
10238: /* kl=6+(cpt-1)*(nlstate+1)+1+(i-1); /\* 6+(1-1)*(2+1)+1+(1-1)=7, 6+(2-1)(2+1)+1+(1-1)=10 *\/ */
10239: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
10240: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
10241: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
10242: if(k==cptcoveff){
10243: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
10244: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
10245: }else{
10246: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
10247: kl++;
10248: }
10249: } /* end covariate */
10250: } /* end if no covariate */
10251:
1.296 brouard 10252: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 10253: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 10254: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 10255: if(cptcoveff ==0){
1.245 brouard 10256: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 10257: }else{
10258: kl=0;
10259: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 10260: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
10261: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 10262: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
10263: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
10264: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 10265: /* vlv= nbcode[Tvaraff[k]][lv]; */
10266: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 10267: kl++;
1.238 brouard 10268: /* kl=6+(cpt-1)*(nlstate+1)+1+(i-1); /\* 6+(1-1)*(2+1)+1+(1-1)=7, 6+(2-1)(2+1)+1+(1-1)=10 *\/ */
10269: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
10270: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
10271: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
10272: if(k==cptcoveff){
1.245 brouard 10273: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 10274: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 10275: }else{
1.332 brouard 10276: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]);
1.238 brouard 10277: kl++;
10278: }
10279: } /* end covariate */
10280: } /* end if no covariate */
1.296 brouard 10281: if(prevbcast == 1){
1.268 brouard 10282: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
10283: /* k1-1 error should be nres-1*/
10284: for (i=1; i<= nlstate ; i ++) {
10285: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10286: else fprintf(ficgp," %%*lf (%%*lf)");
10287: }
1.271 brouard 10288: fprintf(ficgp,"\" t\"Backward (stable) prevalence\" w l lt 6 dt 3,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
1.268 brouard 10289: for (i=1; i<= nlstate ; i ++) {
10290: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10291: else fprintf(ficgp," %%*lf (%%*lf)");
10292: }
1.276 brouard 10293: fprintf(ficgp,"\" t\"95%% CI\" w l lt 4,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
1.268 brouard 10294: for (i=1; i<= nlstate ; i ++) {
10295: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10296: else fprintf(ficgp," %%*lf (%%*lf)");
10297: }
1.274 brouard 10298: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 10299: } /* end if backprojcast */
1.296 brouard 10300: } /* end if prevbcast */
1.276 brouard 10301: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
10302: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 10303: } /* nres */
1.337 brouard 10304: /* } /\* k1 *\/ */
1.201 brouard 10305: } /* cpt */
1.235 brouard 10306:
10307:
1.126 brouard 10308: /*2 eme*/
1.337 brouard 10309: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 10310: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10311: k1=TKresult[nres];
1.338 brouard 10312: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10313: /* if(m != 1 && TKresult[nres]!= k1) */
10314: /* continue; */
1.238 brouard 10315: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 10316: strcpy(gplotlabel,"(");
1.337 brouard 10317: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10318: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10319: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10320: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10321: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10322: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10323: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10324: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10325: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10326: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10327: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10328: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10329: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10330: /* } */
10331: /* /\* for(k=1; k <= ncovds; k++){ *\/ */
10332: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10333: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
10334: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
10335: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 10336: }
1.264 brouard 10337: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 10338: fprintf(ficgp,"\n#\n");
1.223 brouard 10339: if(invalidvarcomb[k1]){
10340: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10341: continue;
10342: }
1.219 brouard 10343:
1.241 brouard 10344: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 10345: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 10346: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
10347: if(vpopbased==0){
1.360 brouard 10348: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nunset ytics; unset y2tics; set ytics nomirror; set y2tics 0,10,100;set y2range [0:100];\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 10349: }else
1.238 brouard 10350: fprintf(ficgp,"\nreplot ");
1.360 brouard 10351: for (i=1; i<= nlstate+1 ; i ++) { /* For state i-1=0 is LE, while i-1=1 to nlstate are origin state */
1.238 brouard 10352: k=2*i;
1.360 brouard 10353: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ?$4 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1, vpopbased); /* for fixed variables age, popbased, mobilav */
10354: for (j=1; j<= nlstate+1 ; j ++) { /* e.. e.1 e.2 again j-1 is the state of end, wlim_i eij*/
10355: if (j==i) fprintf(ficgp," %%lf (%%lf)"); /* We want to read e.. i=1,j=1, e.1 i=2,j=2, e.2 i=3,j=3 */
10356: else fprintf(ficgp," %%*lf (%%*lf)"); /* skipping that field with a star */
1.238 brouard 10357: }
10358: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
1.360 brouard 10359: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1); /* state=i-1=1 to nlstate */
1.261 brouard 10360: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4-$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
1.238 brouard 10361: for (j=1; j<= nlstate+1 ; j ++) {
10362: if (j==i) fprintf(ficgp," %%lf (%%lf)");
10363: else fprintf(ficgp," %%*lf (%%*lf)");
10364: }
10365: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 10366: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4+$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
1.238 brouard 10367: for (j=1; j<= nlstate+1 ; j ++) {
10368: if (j==i) fprintf(ficgp," %%lf (%%lf)");
10369: else fprintf(ficgp," %%*lf (%%*lf)");
10370: }
1.360 brouard 10371: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0,\\\n"); /* ,\\\n added for th percentage graphs */
1.238 brouard 10372: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
10373: } /* state */
1.360 brouard 10374: /* again for the percentag spent in state i-1=1 to i-1=nlstate */
10375: for (i=2; i<= nlstate+1 ; i ++) { /* For state i-1=0 is LE, while i-1=1 to nlstate are origin state */
10376: k=2*i;
10377: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && ($4)<=1 && ($4)>=0 ?($4)*100. : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1, vpopbased); /* for fixed variables age, popbased, mobilav */
10378: for (j=1; j<= nlstate ; j ++)
10379: fprintf(ficgp," %%*lf (%%*lf)"); /* Skipping TLE and LE to read %LE only */
10380: for (j=1; j<= nlstate+1 ; j ++) { /* e.. e.1 e.2 again j-1 is the state of end, wlim_i eij*/
10381: if (j==i) fprintf(ficgp," %%lf (%%lf)"); /* We want to read e.. i=1,j=1, e.1 i=2,j=2, e.2 i=3,j=3 */
10382: else fprintf(ficgp," %%*lf (%%*lf)"); /* skipping that field with a star */
10383: }
10384: if (i== 1) fprintf(ficgp,"\" t\"%%TLE\" w l lt %d axis x1y2, \\\n",i); /* Not used */
10385: else fprintf(ficgp,"\" t\"%%LE in state (%d)\" w l lw 2 lt %d axis x1y2, \\\n",i-1,i+1); /* state=i-1=1 to nlstate */
10386: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && ($4-$5*2)<=1 && ($4-$5*2)>=0? ($4-$5*2)*100. : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
10387: for (j=1; j<= nlstate ; j ++)
10388: fprintf(ficgp," %%*lf (%%*lf)"); /* Skipping TLE and LE to read %LE only */
10389: for (j=1; j<= nlstate+1 ; j ++) {
10390: if (j==i) fprintf(ficgp," %%lf (%%lf)");
10391: else fprintf(ficgp," %%*lf (%%*lf)");
10392: }
10393: fprintf(ficgp,"\" t\"\" w l lt 0 axis x1y2,");
10394: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && ($4+$5*2)<=1 && ($4+$5*2)>=0 ? ($4+$5*2)*100. : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
10395: for (j=1; j<= nlstate ; j ++)
10396: fprintf(ficgp," %%*lf (%%*lf)"); /* Skipping TLE and LE to read %LE only */
10397: for (j=1; j<= nlstate+1 ; j ++) {
10398: if (j==i) fprintf(ficgp," %%lf (%%lf)");
10399: else fprintf(ficgp," %%*lf (%%*lf)");
10400: }
10401: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0 axis x1y2");
10402: else fprintf(ficgp,"\" t\"\" w l lt 0 axis x1y2,\\\n");
10403: } /* state for percent */
1.238 brouard 10404: } /* vpopbased */
1.264 brouard 10405: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; unset label;\n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 10406: } /* end nres */
1.337 brouard 10407: /* } /\* k1 end 2 eme*\/ */
1.238 brouard 10408:
10409:
10410: /*3eme*/
1.337 brouard 10411: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 10412: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10413: k1=TKresult[nres];
1.338 brouard 10414: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10415: /* if(m != 1 && TKresult[nres]!= k1) */
10416: /* continue; */
1.238 brouard 10417:
1.332 brouard 10418: for (cpt=1; cpt<= nlstate ; cpt ++) { /* Fragile no verification of covariate values */
1.261 brouard 10419: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 10420: strcpy(gplotlabel,"(");
1.337 brouard 10421: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10422: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10423: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10424: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10425: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10426: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
10427: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10428: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10429: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10430: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10431: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10432: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10433: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10434: /* } */
10435: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10436: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
10437: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
10438: }
1.264 brouard 10439: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 10440: fprintf(ficgp,"\n#\n");
10441: if(invalidvarcomb[k1]){
10442: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10443: continue;
10444: }
10445:
10446: /* k=2+nlstate*(2*cpt-2); */
10447: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 10448: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 10449: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 10450: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 10451: plot [%.f:%.f] \"%s\" every :::%d::%d u 1:%d t \"e%d1\" w l",ageminpar,fage,subdirf2(fileresu,"E_"),nres-1,nres-1,k,cpt);
1.238 brouard 10452: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
10453: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
10454: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
10455: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
10456: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
10457: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 10458:
1.238 brouard 10459: */
10460: for (i=1; i< nlstate ; i ++) {
1.261 brouard 10461: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d%d\" w l",subdirf2(fileresu,"E_"),nres-1,nres-1,k+i,cpt,i+1);
1.238 brouard 10462: /* fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d%d\" w l",subdirf2(fileres,"e"),k1-1,k1-1,k+2*i,cpt,i+1);*/
1.219 brouard 10463:
1.238 brouard 10464: }
1.261 brouard 10465: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),nres-1,nres-1,k+nlstate,cpt);
1.238 brouard 10466: }
1.264 brouard 10467: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 10468: } /* end nres */
1.337 brouard 10469: /* } /\* end kl 3eme *\/ */
1.126 brouard 10470:
1.223 brouard 10471: /* 4eme */
1.201 brouard 10472: /* Survival functions (period) from state i in state j by initial state i */
1.337 brouard 10473: /* for (k1=1; k1<=m; k1++){ /\* For each covariate and each value *\/ */
1.238 brouard 10474: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10475: k1=TKresult[nres];
1.338 brouard 10476: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10477: /* if(m != 1 && TKresult[nres]!= k1) */
10478: /* continue; */
1.238 brouard 10479: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 10480: strcpy(gplotlabel,"(");
1.337 brouard 10481: fprintf(ficgp,"\n#\n#\n# Survival functions in state %d : 'LIJ_' files, cov=%d state=%d", cpt, k1, cpt);
10482: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10483: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10484: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10485: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10486: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10487: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10488: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10489: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10490: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10491: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10492: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10493: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10494: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10495: /* } */
10496: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10497: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10498: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 10499: }
1.264 brouard 10500: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 10501: fprintf(ficgp,"\n#\n");
10502: if(invalidvarcomb[k1]){
10503: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10504: continue;
1.223 brouard 10505: }
1.238 brouard 10506:
1.241 brouard 10507: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 10508: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.238 brouard 10509: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
10510: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
10511: k=3;
10512: for (i=1; i<= nlstate ; i ++){
10513: if(i==1){
10514: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
10515: }else{
10516: fprintf(ficgp,", '' ");
10517: }
10518: l=(nlstate+ndeath)*(i-1)+1;
10519: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
10520: for (j=2; j<= nlstate+ndeath ; j ++)
10521: fprintf(ficgp,"+$%d",k+l+j-1);
10522: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
10523: } /* nlstate */
1.264 brouard 10524: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 10525: } /* end cpt state*/
10526: } /* end nres */
1.337 brouard 10527: /* } /\* end covariate k1 *\/ */
1.238 brouard 10528:
1.220 brouard 10529: /* 5eme */
1.201 brouard 10530: /* Survival functions (period) from state i in state j by final state j */
1.337 brouard 10531: /* for (k1=1; k1<= m ; k1++){ /\* For each covariate combination if any *\/ */
1.238 brouard 10532: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10533: k1=TKresult[nres];
1.338 brouard 10534: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10535: /* if(m != 1 && TKresult[nres]!= k1) */
10536: /* continue; */
1.238 brouard 10537: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 10538: strcpy(gplotlabel,"(");
1.238 brouard 10539: fprintf(ficgp,"\n#\n#\n# Survival functions in state j and all livestates from state i by final state j: 'lij' files, cov=%d state=%d",k1, cpt);
1.337 brouard 10540: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10541: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10542: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10543: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10544: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10545: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10546: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10547: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10548: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10549: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10550: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10551: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10552: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10553: /* } */
10554: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10555: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10556: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 10557: }
1.264 brouard 10558: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 10559: fprintf(ficgp,"\n#\n");
10560: if(invalidvarcomb[k1]){
10561: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10562: continue;
10563: }
1.227 brouard 10564:
1.241 brouard 10565: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 10566: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.238 brouard 10567: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
10568: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
10569: k=3;
10570: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
10571: if(j==1)
10572: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
10573: else
10574: fprintf(ficgp,", '' ");
10575: l=(nlstate+ndeath)*(cpt-1) +j;
10576: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
10577: /* for (i=2; i<= nlstate+ndeath ; i ++) */
10578: /* fprintf(ficgp,"+$%d",k+l+i-1); */
10579: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
10580: } /* nlstate */
10581: fprintf(ficgp,", '' ");
10582: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
10583: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
10584: l=(nlstate+ndeath)*(cpt-1) +j;
10585: if(j < nlstate)
10586: fprintf(ficgp,"$%d +",k+l);
10587: else
10588: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
10589: }
1.264 brouard 10590: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 10591: } /* end cpt state*/
1.337 brouard 10592: /* } /\* end covariate *\/ */
1.238 brouard 10593: } /* end nres */
1.227 brouard 10594:
1.220 brouard 10595: /* 6eme */
1.202 brouard 10596: /* CV preval stable (period) for each covariate */
1.337 brouard 10597: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 10598: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10599: k1=TKresult[nres];
1.338 brouard 10600: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10601: /* if(m != 1 && TKresult[nres]!= k1) */
10602: /* continue; */
1.255 brouard 10603: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 10604: strcpy(gplotlabel,"(");
1.288 brouard 10605: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 10606: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10607: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10608: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10609: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10610: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10611: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10612: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10613: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10614: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10615: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10616: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10617: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10618: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10619: /* } */
10620: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10621: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10622: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 10623: }
1.264 brouard 10624: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 10625: fprintf(ficgp,"\n#\n");
1.223 brouard 10626: if(invalidvarcomb[k1]){
1.227 brouard 10627: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10628: continue;
1.223 brouard 10629: }
1.227 brouard 10630:
1.241 brouard 10631: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 10632: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.126 brouard 10633: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 10634: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 10635: k=3; /* Offset */
1.255 brouard 10636: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 10637: if(i==1)
10638: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
10639: else
10640: fprintf(ficgp,", '' ");
1.255 brouard 10641: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 10642: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
10643: for (j=2; j<= nlstate ; j ++)
10644: fprintf(ficgp,"+$%d",k+l+j-1);
10645: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 10646: } /* nlstate */
1.264 brouard 10647: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 10648: } /* end cpt state*/
10649: } /* end covariate */
1.227 brouard 10650:
10651:
1.220 brouard 10652: /* 7eme */
1.296 brouard 10653: if(prevbcast == 1){
1.288 brouard 10654: /* CV backward prevalence for each covariate */
1.337 brouard 10655: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 10656: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10657: k1=TKresult[nres];
1.338 brouard 10658: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10659: /* if(m != 1 && TKresult[nres]!= k1) */
10660: /* continue; */
1.268 brouard 10661: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 10662: strcpy(gplotlabel,"(");
1.288 brouard 10663: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 10664: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10665: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10666: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10667: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10668: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10669: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10670: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10671: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10672: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10673: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10674: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10675: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10676: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10677: /* } */
10678: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10679: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10680: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 10681: }
1.264 brouard 10682: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 10683: fprintf(ficgp,"\n#\n");
10684: if(invalidvarcomb[k1]){
10685: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10686: continue;
10687: }
10688:
1.241 brouard 10689: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 10690: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.227 brouard 10691: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 10692: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 10693: k=3; /* Offset */
1.268 brouard 10694: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 10695: if(i==1)
10696: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
10697: else
10698: fprintf(ficgp,", '' ");
10699: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 10700: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 10701: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
10702: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
1.255 brouard 10703: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 10704: /* for (j=2; j<= nlstate ; j ++) */
10705: /* fprintf(ficgp,"+$%d",k+l+j-1); */
10706: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 10707: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 10708: } /* nlstate */
1.264 brouard 10709: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 10710: } /* end cpt state*/
10711: } /* end covariate */
1.296 brouard 10712: } /* End if prevbcast */
1.218 brouard 10713:
1.223 brouard 10714: /* 8eme */
1.218 brouard 10715: if(prevfcast==1){
1.288 brouard 10716: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 10717:
1.337 brouard 10718: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 10719: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10720: k1=TKresult[nres];
1.338 brouard 10721: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10722: /* if(m != 1 && TKresult[nres]!= k1) */
10723: /* continue; */
1.211 brouard 10724: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 10725: strcpy(gplotlabel,"(");
1.288 brouard 10726: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 10727: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10728: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10729: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10730: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
10731: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
10732: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10733: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10734: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10735: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10736: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10737: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10738: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10739: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10740: /* } */
10741: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10742: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10743: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 10744: }
1.264 brouard 10745: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 10746: fprintf(ficgp,"\n#\n");
10747: if(invalidvarcomb[k1]){
10748: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10749: continue;
10750: }
10751:
10752: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 10753: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 10754: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.227 brouard 10755: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 10756: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 10757:
10758: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
10759: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
10760: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
10761: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 10762: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10763: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
10764: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10765: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 10766: if(i==istart){
1.227 brouard 10767: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
10768: }else{
10769: fprintf(ficgp,",\\\n '' ");
10770: }
1.367 ! brouard 10771: /* if(cptcoveff ==0){ /\* No covariate *\/ */
! 10772: if(cptcovs ==0){ /* No covariate */
1.227 brouard 10773: ioffset=2; /* Age is in 2 */
10774: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
10775: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
10776: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
10777: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
1.367 ! brouard 10778: /*# V1 = 1 yearproj age age*p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
! 10779: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
1.227 brouard 10780: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 10781: if(i==nlstate+1){
1.270 brouard 10782: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 10783: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
10784: fprintf(ficgp,",\\\n '' ");
10785: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 10786: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 10787: offyear, \
1.268 brouard 10788: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 10789: }else
1.227 brouard 10790: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
10791: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
10792: }else{ /* more than 2 covariates */
1.367 ! brouard 10793: /* ioffset=2*cptcoveff+2; */ /* Age is in 4 or 6 or etc.*/
! 10794: ioffset=2*cptcovs+2; /* Age is in 4 or 6 or etc.*/
1.270 brouard 10795: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10796: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.367 ! brouard 10797: /* # Forecasting at date 3/1/2003 */
! 10798: /* V1=0 V2=1 V3=0 V6=2.47 yearproj age */
! 10799: /* # 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 */
! 10800: /* # p11 p21 p31 wp.1 p12 p22 p32 wp.2 p13 p23 p33 wp.3 p14 p24 p34 wp.4 */
! 10801: /* 1 0 2 1 3 0 6 2.47 2003 100 1.000 0.000 0.000 0.297 0.000 1.000 0.000 0.207 0.000 0.000 1.000 0.497 0.000 0.000 0.000 0.000 */
1.270 brouard 10802: iyearc=ioffset-1;
10803: iagec=ioffset;
1.227 brouard 10804: fprintf(ficgp," u %d:(",ioffset);
10805: kl=0;
10806: strcpy(gplotcondition,"(");
1.351 brouard 10807: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate writing the chain of conditions *\/ */
1.332 brouard 10808: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
1.351 brouard 10809: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10810: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10811: lv=Tvresult[nres][k];
10812: vlv=TinvDoQresult[nres][Tvresult[nres][k]];
1.227 brouard 10813: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
10814: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
10815: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 10816: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
1.351 brouard 10817: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
1.227 brouard 10818: kl++;
1.364 brouard 10819: /* Problem with quantitative variables TinvDoQresult[nres] */
1.351 brouard 10820: /* sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]); */
1.364 brouard 10821: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%lg " ,kl,lv, kl+1, vlv );/* Solved but quantitative must be shifted */
1.227 brouard 10822: kl++;
1.351 brouard 10823: if(k <cptcovs && cptcovs>1)
1.227 brouard 10824: sprintf(gplotcondition+strlen(gplotcondition)," && ");
10825: }
10826: strcpy(gplotcondition+strlen(gplotcondition),")");
10827: /* kl=6+(cpt-1)*(nlstate+1)+1+(i-1); /\* 6+(1-1)*(2+1)+1+(1-1)=7, 6+(2-1)(2+1)+1+(1-1)=10 *\/ */
10828: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
10829: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
10830: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
10831: if(i==nlstate+1){
1.270 brouard 10832: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
10833: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 10834: fprintf(ficgp,",\\\n '' ");
1.364 brouard 10835: fprintf(ficgp," u %d:(",iagec); /* Below iyearc should be increades if quantitative variable in the reult line */
10836: /* $7==6 && $8==2.47 ) && (($9-$10) == 1953 ) ? $12/(1.-$24) : 1/0):7 with labels center not */
10837: /* but was && $7==6 && $8==2 ) && (($7-$8) == 1953 ) ? $12/(1.-$24) : 1/0):7 with labels center not */
1.270 brouard 10838: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
10839: iyearc, iagec, offyear, \
10840: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 10841: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
1.227 brouard 10842: }else{
10843: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
10844: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
10845: }
10846: } /* end if covariate */
10847: } /* nlstate */
1.264 brouard 10848: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 10849: } /* end cpt state*/
10850: } /* end covariate */
10851: } /* End if prevfcast */
1.227 brouard 10852:
1.296 brouard 10853: if(prevbcast==1){
1.268 brouard 10854: /* Back projection from cross-sectional to stable (mixed) for each covariate */
10855:
1.337 brouard 10856: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.268 brouard 10857: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10858: k1=TKresult[nres];
1.338 brouard 10859: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10860: /* if(m != 1 && TKresult[nres]!= k1) */
10861: /* continue; */
1.268 brouard 10862: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
10863: strcpy(gplotlabel,"(");
10864: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
1.337 brouard 10865: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10866: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10867: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10868: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
10869: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
10870: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
10871: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10872: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10873: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10874: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10875: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10876: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10877: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10878: /* } */
10879: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10880: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10881: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.268 brouard 10882: }
10883: strcpy(gplotlabel+strlen(gplotlabel),")");
10884: fprintf(ficgp,"\n#\n");
10885: if(invalidvarcomb[k1]){
10886: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10887: continue;
10888: }
10889:
10890: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
10891: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
10892: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
10893: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
10894: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
10895:
10896: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
10897: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
10898: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
10899: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
10900: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10901: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
10902: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10903: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
10904: if(i==istart){
10905: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
10906: }else{
10907: fprintf(ficgp,",\\\n '' ");
10908: }
1.351 brouard 10909: /* if(cptcoveff ==0){ /\* No covariate *\/ */
10910: if(cptcovs ==0){ /* No covariate */
1.268 brouard 10911: ioffset=2; /* Age is in 2 */
10912: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
10913: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
10914: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
10915: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
10916: fprintf(ficgp," u %d:(", ioffset);
10917: if(i==nlstate+1){
1.367 ! brouard 10918: fprintf(ficgp," $%d):1 t 'bw%d' with line lc variable ", \
! 10919: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),cpt );
! 10920: /* fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \ */
! 10921: /* ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt ); */
1.268 brouard 10922: fprintf(ficgp,",\\\n '' ");
10923: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 10924: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 10925: offbyear, \
10926: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
1.367 ! brouard 10927: }else /* not sure divided by 1- to be checked */
! 10928: fprintf(ficgp," $%d) t 'b%d%d' with line ", \
! 10929: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),cpt,i );
! 10930: /* fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \ */
! 10931: /* ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i ); */
1.268 brouard 10932: }else{ /* more than 2 covariates */
1.367 ! brouard 10933: /* ioffset=2*cptcoveff+2; /\* Age is in 4 or 6 or etc.*\/ */
! 10934: ioffset=2*cptcovs+2; /* Age is in 4 or 6 or etc.*/
1.270 brouard 10935: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10936: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.367 ! brouard 10937: /* #****** hbijx=probability over h years, hb.jx is weighted by observed prev */
! 10938: /* # V1=0 V2=1 V3=0 V6=2.47 */
! 10939: /* yearbproj age b11 b21 b31 b.1 b12 b22 b32 b.2 b13 b23 b33 b.3 b14 b24 b34 b.4 */
! 10940: /* # Back Forecasting at date 3/1/2003 */
! 10941: /* 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 */
! 10942: /* 1 0 2 1 3 0 6 2.47 2003 50 1.000 0.000 0.000 0.714 0.000 1.000 0.000 0.286 0.000 0.000 1.000 0.000 0.000 0.000 0.000 0.000 */
1.270 brouard 10943: iyearc=ioffset-1;
10944: iagec=ioffset;
1.268 brouard 10945: fprintf(ficgp," u %d:(",ioffset);
10946: kl=0;
10947: strcpy(gplotcondition,"(");
1.337 brouard 10948: for (k=1; k<=cptcovs; k++){ /* For each covariate k of the resultline, get corresponding value lv for combination k1 */
1.367 ! brouard 10949: /* if(Dummy[modelresult[nres][k]]==0){ /\* To be verified *\/ */
1.337 brouard 10950: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate writing the chain of conditions *\/ */
10951: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
10952: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
10953: lv=Tvresult[nres][k];
10954: vlv=TinvDoQresult[nres][Tvresult[nres][k]];
10955: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
10956: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
10957: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
10958: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
10959: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10960: kl++;
10961: /* sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]); */
10962: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%lg " ,kl,Tvresult[nres][k], kl+1,TinvDoQresult[nres][Tvresult[nres][k]]);
10963: kl++;
1.338 brouard 10964: if(k <cptcovs && cptcovs>1)
1.337 brouard 10965: sprintf(gplotcondition+strlen(gplotcondition)," && ");
1.367 ! brouard 10966: /* } */ /* end dummy */
1.268 brouard 10967: }
10968: strcpy(gplotcondition+strlen(gplotcondition),")");
10969: /* kl=6+(cpt-1)*(nlstate+1)+1+(i-1); /\* 6+(1-1)*(2+1)+1+(1-1)=7, 6+(2-1)(2+1)+1+(1-1)=10 *\/ */
10970: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
10971: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
10972: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
10973: if(i==nlstate+1){
1.270 brouard 10974: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
10975: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 10976: fprintf(ficgp,",\\\n '' ");
1.270 brouard 10977: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 10978: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 10979: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
10980: iyearc,iagec,offbyear, \
10981: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 10982: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
10983: }else{
10984: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
10985: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
10986: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
10987: }
10988: } /* end if covariate */
10989: } /* nlstate */
10990: fprintf(ficgp,"\nset out; unset label;\n");
10991: } /* end cpt state*/
10992: } /* end covariate */
1.296 brouard 10993: } /* End if prevbcast */
1.268 brouard 10994:
1.227 brouard 10995:
1.238 brouard 10996: /* 9eme writing MLE parameters */
10997: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 10998: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 10999: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 11000: for(k=1; k <=(nlstate+ndeath); k++){
11001: if (k != i) {
1.227 brouard 11002: fprintf(ficgp,"# current state %d\n",k);
11003: for(j=1; j <=ncovmodel; j++){
11004: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
11005: jk++;
11006: }
11007: fprintf(ficgp,"\n");
1.126 brouard 11008: }
11009: }
1.223 brouard 11010: }
1.187 brouard 11011: fprintf(ficgp,"##############\n#\n");
1.227 brouard 11012:
1.145 brouard 11013: /*goto avoid;*/
1.238 brouard 11014: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
11015: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 11016: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
11017: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
11018: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
11019: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
11020: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
11021: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
11022: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
11023: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
11024: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
11025: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
11026: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
11027: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
11028: fprintf(ficgp,"#\n");
1.223 brouard 11029: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 11030: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.338 brouard 11031: fprintf(ficgp,"#model=1+age+%s \n",model);
1.238 brouard 11032: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.351 brouard 11033: /* fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/\* to be checked *\/ */
11034: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcovs,m);/* to be checked */
1.337 brouard 11035: /* for(k1=1; k1 <=m; k1++) /\* For each combination of covariate *\/ */
1.237 brouard 11036: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 11037: /* k1=nres; */
1.338 brouard 11038: k1=TKresult[nres];
11039: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 11040: fprintf(ficgp,"\n\n# Resultline k1=%d ",k1);
1.264 brouard 11041: strcpy(gplotlabel,"(");
1.276 brouard 11042: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.337 brouard 11043: for (k=1; k<=cptcovs; k++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
11044: /* for each resultline nres, and position k, Tvresult[nres][k] gives the name of the variable and
11045: TinvDoQresult[nres][Tvresult[nres][k]] gives its value double or integer) */
11046: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
11047: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
11048: }
11049: /* if(m != 1 && TKresult[nres]!= k1) */
11050: /* continue; */
11051: /* fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1); */
11052: /* strcpy(gplotlabel,"("); */
11053: /* /\*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*\/ */
11054: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
11055: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
11056: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
11057: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
11058: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
11059: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
11060: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
11061: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
11062: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
11063: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
11064: /* } */
11065: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
11066: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
11067: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
11068: /* } */
1.264 brouard 11069: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 11070: fprintf(ficgp,"\n#\n");
1.264 brouard 11071: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 11072: fprintf(ficgp,"\nset key outside ");
11073: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
11074: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 11075: fprintf(ficgp,"\nset ter svg size 640, 480 ");
11076: if (ng==1){
11077: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
11078: fprintf(ficgp,"\nunset log y");
11079: }else if (ng==2){
11080: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
11081: fprintf(ficgp,"\nset log y");
11082: }else if (ng==3){
11083: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
11084: fprintf(ficgp,"\nset log y");
11085: }else
11086: fprintf(ficgp,"\nunset title ");
11087: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
11088: i=1;
11089: for(k2=1; k2<=nlstate; k2++) {
11090: k3=i;
11091: for(k=1; k<=(nlstate+ndeath); k++) {
11092: if (k != k2){
11093: switch( ng) {
11094: case 1:
11095: if(nagesqr==0)
11096: fprintf(ficgp," p%d+p%d*x",i,i+1);
11097: else /* nagesqr =1 */
11098: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
11099: break;
11100: case 2: /* ng=2 */
11101: if(nagesqr==0)
11102: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
11103: else /* nagesqr =1 */
11104: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
11105: break;
11106: case 3:
11107: if(nagesqr==0)
11108: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
11109: else /* nagesqr =1 */
11110: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
11111: break;
11112: }
11113: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 11114: ijp=1; /* product no age */
11115: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
11116: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 11117: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 11118: switch(Typevar[j]){
11119: case 1:
11120: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
11121: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
11122: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
11123: if(DummyV[j]==0){/* Bug valgrind */
11124: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
11125: }else{ /* quantitative */
11126: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
11127: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
11128: }
11129: ij++;
1.268 brouard 11130: }
1.237 brouard 11131: }
1.329 brouard 11132: }
11133: break;
11134: case 2:
11135: if(cptcovprod >0){
11136: if(j==Tprod[ijp]) { /* */
11137: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
11138: if(ijp <=cptcovprod) { /* Product */
11139: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
11140: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
11141: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],nbcode[Tvard[ijp][2]][codtabm(k1,j)]); */
11142: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
11143: }else{ /* Vn is dummy and Vm is quanti */
11144: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
11145: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
11146: }
11147: }else{ /* Vn*Vm Vn is quanti */
11148: if(DummyV[Tvard[ijp][2]]==0){
11149: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
11150: }else{ /* Both quanti */
11151: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
11152: }
1.268 brouard 11153: }
1.329 brouard 11154: ijp++;
1.237 brouard 11155: }
1.329 brouard 11156: } /* end Tprod */
11157: }
11158: break;
1.349 brouard 11159: case 3:
11160: if(cptcovdageprod >0){
11161: /* if(j==Tprod[ijp]) { */ /* not necessary */
11162: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
1.350 brouard 11163: if(ijp <=cptcovprod) { /* Product Vn*Vm and age*VN*Vm*/
11164: if(DummyV[Tvardk[ijp][1]]==0){/* Vn is dummy */
11165: if(DummyV[Tvardk[ijp][2]]==0){/* Vn and Vm are dummy */
1.349 brouard 11166: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],nbcode[Tvard[ijp][2]][codtabm(k1,j)]); */
11167: fprintf(ficgp,"+p%d*%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
11168: }else{ /* Vn is dummy and Vm is quanti */
11169: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
1.350 brouard 11170: fprintf(ficgp,"+p%d*%d*%f*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvardk[ijp][1]],Tqinvresult[nres][Tvardk[ijp][2]]);
1.349 brouard 11171: }
1.350 brouard 11172: }else{ /* age* Vn*Vm Vn is quanti HERE */
1.349 brouard 11173: if(DummyV[Tvard[ijp][2]]==0){
1.350 brouard 11174: fprintf(ficgp,"+p%d*%d*%f*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvardk[ijp][2]],Tqinvresult[nres][Tvardk[ijp][1]]);
1.349 brouard 11175: }else{ /* Both quanti */
1.350 brouard 11176: fprintf(ficgp,"+p%d*%f*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvardk[ijp][1]],Tqinvresult[nres][Tvardk[ijp][2]]);
1.349 brouard 11177: }
11178: }
11179: ijp++;
11180: }
11181: /* } */ /* end Tprod */
11182: }
11183: break;
1.329 brouard 11184: case 0:
11185: /* simple covariate */
1.264 brouard 11186: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 11187: if(Dummy[j]==0){
11188: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
11189: }else{ /* quantitative */
11190: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 11191: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 11192: }
1.329 brouard 11193: /* end simple */
11194: break;
11195: default:
11196: break;
11197: } /* end switch */
1.237 brouard 11198: } /* end j */
1.329 brouard 11199: }else{ /* k=k2 */
11200: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
11201: fprintf(ficgp," (1.");i=i-ncovmodel;
11202: }else
11203: i=i-ncovmodel;
1.223 brouard 11204: }
1.227 brouard 11205:
1.223 brouard 11206: if(ng != 1){
11207: fprintf(ficgp,")/(1");
1.227 brouard 11208:
1.264 brouard 11209: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 11210: if(nagesqr==0)
1.264 brouard 11211: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 11212: else /* nagesqr =1 */
1.264 brouard 11213: fprintf(ficgp,"+exp(p%d+p%d*x+p%d*x*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1,k3+(cpt-1)*ncovmodel+1+nagesqr);
1.217 brouard 11214:
1.223 brouard 11215: ij=1;
1.329 brouard 11216: ijp=1;
11217: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
11218: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
11219: switch(Typevar[j]){
11220: case 1:
11221: if(cptcovage >0){
11222: if(j==Tage[ij]) { /* Bug valgrind */
11223: if(ij <=cptcovage) { /* Bug valgrind */
11224: if(DummyV[j]==0){/* Bug valgrind */
11225: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
11226: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
11227: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
11228: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
11229: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
11230: }else{ /* quantitative */
11231: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
11232: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
11233: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
11234: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
11235: }
11236: ij++;
11237: }
11238: }
11239: }
11240: break;
11241: case 2:
11242: if(cptcovprod >0){
11243: if(j==Tprod[ijp]) { /* */
11244: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
11245: if(ijp <=cptcovprod) { /* Product */
11246: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
11247: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
11248: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],nbcode[Tvard[ijp][2]][codtabm(k1,j)]); */
11249: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
11250: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
11251: }else{ /* Vn is dummy and Vm is quanti */
11252: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
11253: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
11254: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
11255: }
11256: }else{ /* Vn*Vm Vn is quanti */
11257: if(DummyV[Tvard[ijp][2]]==0){
11258: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
11259: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
11260: }else{ /* Both quanti */
11261: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
11262: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
11263: }
11264: }
11265: ijp++;
11266: }
11267: } /* end Tprod */
11268: } /* end if */
11269: break;
1.349 brouard 11270: case 3:
11271: if(cptcovdageprod >0){
11272: /* if(j==Tprod[ijp]) { /\* *\/ */
11273: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
11274: if(ijp <=cptcovprod) { /* Product */
1.350 brouard 11275: if(DummyV[Tvardk[ijp][1]]==0){/* Vn is dummy */
11276: if(DummyV[Tvardk[ijp][2]]==0){/* Vn and Vm are dummy */
1.349 brouard 11277: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],nbcode[Tvard[ijp][2]][codtabm(k1,j)]); */
1.350 brouard 11278: fprintf(ficgp,"+p%d*%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvardk[ijp][1]],Tinvresult[nres][Tvardk[ijp][2]]);
1.349 brouard 11279: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
11280: }else{ /* Vn is dummy and Vm is quanti */
11281: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
1.350 brouard 11282: fprintf(ficgp,"+p%d*%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvardk[ijp][1]],Tqinvresult[nres][Tvardk[ijp][2]]);
1.349 brouard 11283: /* fprintf(ficgp,"+p%d*%d*%f*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
11284: }
11285: }else{ /* Vn*Vm Vn is quanti */
1.350 brouard 11286: if(DummyV[Tvardk[ijp][2]]==0){
11287: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvardk[ijp][2]],Tqinvresult[nres][Tvardk[ijp][1]]);
1.349 brouard 11288: /* fprintf(ficgp,"+p%d*%d*%f*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
11289: }else{ /* Both quanti */
1.350 brouard 11290: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvardk[ijp][1]],Tqinvresult[nres][Tvardk[ijp][2]]);
1.349 brouard 11291: /* fprintf(ficgp,"+p%d*%f*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
11292: }
11293: }
11294: ijp++;
11295: }
11296: /* } /\* end Tprod *\/ */
11297: } /* end if */
11298: break;
1.329 brouard 11299: case 0:
11300: /* simple covariate */
11301: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
11302: if(Dummy[j]==0){
11303: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
11304: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
11305: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
11306: }else{ /* quantitative */
11307: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
11308: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
11309: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
11310: }
11311: /* end simple */
11312: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
11313: break;
11314: default:
11315: break;
11316: } /* end switch */
1.223 brouard 11317: }
11318: fprintf(ficgp,")");
11319: }
11320: fprintf(ficgp,")");
11321: if(ng ==2)
1.276 brouard 11322: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"p%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 11323: else /* ng= 3 */
1.276 brouard 11324: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"i%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.329 brouard 11325: }else{ /* end ng <> 1 */
1.223 brouard 11326: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 11327: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"logit(p%d%d)\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 11328: }
11329: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
11330: fprintf(ficgp,",");
11331: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
11332: fprintf(ficgp,",");
11333: i=i+ncovmodel;
11334: } /* end k */
11335: } /* end k2 */
1.276 brouard 11336: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
11337: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.337 brouard 11338: } /* end resultline */
1.223 brouard 11339: } /* end ng */
11340: /* avoid: */
11341: fflush(ficgp);
1.126 brouard 11342: } /* end gnuplot */
11343:
11344:
11345: /*************** Moving average **************/
1.219 brouard 11346: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 11347: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 11348:
1.222 brouard 11349: int i, cpt, cptcod;
11350: int modcovmax =1;
11351: int mobilavrange, mob;
11352: int iage=0;
1.288 brouard 11353: int firstA1=0, firstA2=0;
1.222 brouard 11354:
1.266 brouard 11355: double sum=0., sumr=0.;
1.222 brouard 11356: double age;
1.266 brouard 11357: double *sumnewp, *sumnewm, *sumnewmr;
11358: double *agemingood, *agemaxgood;
11359: double *agemingoodr, *agemaxgoodr;
1.222 brouard 11360:
11361:
1.278 brouard 11362: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
11363: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 11364:
11365: sumnewp = vector(1,ncovcombmax);
11366: sumnewm = vector(1,ncovcombmax);
1.266 brouard 11367: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 11368: agemingood = vector(1,ncovcombmax);
1.266 brouard 11369: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 11370: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 11371: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 11372:
11373: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 11374: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 11375: sumnewp[cptcod]=0.;
1.266 brouard 11376: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
11377: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 11378: }
11379: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
11380:
1.266 brouard 11381: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
11382: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 11383: else mobilavrange=mobilav;
11384: for (age=bage; age<=fage; age++)
11385: for (i=1; i<=nlstate;i++)
11386: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
11387: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
11388: /* We keep the original values on the extreme ages bage, fage and for
11389: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
11390: we use a 5 terms etc. until the borders are no more concerned.
11391: */
11392: for (mob=3;mob <=mobilavrange;mob=mob+2){
11393: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 11394: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
11395: sumnewm[cptcod]=0.;
11396: for (i=1; i<=nlstate;i++){
1.222 brouard 11397: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
11398: for (cpt=1;cpt<=(mob-1)/2;cpt++){
11399: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
11400: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
11401: }
11402: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 11403: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
11404: } /* end i */
11405: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
11406: } /* end cptcod */
1.222 brouard 11407: }/* end age */
11408: }/* end mob */
1.266 brouard 11409: }else{
11410: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 11411: return -1;
1.266 brouard 11412: }
11413:
11414: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 11415: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
11416: if(invalidvarcomb[cptcod]){
11417: printf("\nCombination (%d) ignored because no cases \n",cptcod);
11418: continue;
11419: }
1.219 brouard 11420:
1.266 brouard 11421: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
11422: sumnewm[cptcod]=0.;
11423: sumnewmr[cptcod]=0.;
11424: for (i=1; i<=nlstate;i++){
11425: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
11426: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
11427: }
11428: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
11429: agemingoodr[cptcod]=age;
11430: }
11431: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
11432: agemingood[cptcod]=age;
11433: }
11434: } /* age */
11435: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 11436: sumnewm[cptcod]=0.;
1.266 brouard 11437: sumnewmr[cptcod]=0.;
1.222 brouard 11438: for (i=1; i<=nlstate;i++){
11439: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 11440: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
11441: }
11442: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
11443: agemaxgoodr[cptcod]=age;
1.222 brouard 11444: }
11445: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 11446: agemaxgood[cptcod]=age;
11447: }
11448: } /* age */
11449: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
11450: /* but they will change */
1.288 brouard 11451: firstA1=0;firstA2=0;
1.266 brouard 11452: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
11453: sumnewm[cptcod]=0.;
11454: sumnewmr[cptcod]=0.;
11455: for (i=1; i<=nlstate;i++){
11456: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
11457: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
11458: }
11459: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
11460: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
11461: agemaxgoodr[cptcod]=age; /* age min */
11462: for (i=1; i<=nlstate;i++)
11463: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
11464: }else{ /* bad we change the value with the values of good ages */
11465: for (i=1; i<=nlstate;i++){
11466: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
11467: } /* i */
11468: } /* end bad */
11469: }else{
11470: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
11471: agemaxgood[cptcod]=age;
11472: }else{ /* bad we change the value with the values of good ages */
11473: for (i=1; i<=nlstate;i++){
11474: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
11475: } /* i */
11476: } /* end bad */
11477: }/* end else */
11478: sum=0.;sumr=0.;
11479: for (i=1; i<=nlstate;i++){
11480: sum+=mobaverage[(int)age][i][cptcod];
11481: sumr+=probs[(int)age][i][cptcod];
11482: }
11483: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 11484: if(!firstA1){
11485: firstA1=1;
11486: printf("Moving average A1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you increase bage=%d. Others in log file...\n",cptcod,sumr, (int)age, (int)bage);
11487: }
11488: fprintf(ficlog,"Moving average A1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.266 brouard 11489: } /* end bad */
11490: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
11491: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 11492: if(!firstA2){
11493: firstA2=1;
11494: printf("Moving average A2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase bage=%d. Others in log file...\n",cptcod,sumr, (int)age, (int)bage);
11495: }
11496: fprintf(ficlog,"Moving average A2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.222 brouard 11497: } /* end bad */
11498: }/* age */
1.266 brouard 11499:
11500: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 11501: sumnewm[cptcod]=0.;
1.266 brouard 11502: sumnewmr[cptcod]=0.;
1.222 brouard 11503: for (i=1; i<=nlstate;i++){
11504: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 11505: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
11506: }
11507: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
11508: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
11509: agemingoodr[cptcod]=age;
11510: for (i=1; i<=nlstate;i++)
11511: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
11512: }else{ /* bad we change the value with the values of good ages */
11513: for (i=1; i<=nlstate;i++){
11514: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
11515: } /* i */
11516: } /* end bad */
11517: }else{
11518: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
11519: agemingood[cptcod]=age;
11520: }else{ /* bad */
11521: for (i=1; i<=nlstate;i++){
11522: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
11523: } /* i */
11524: } /* end bad */
11525: }/* end else */
11526: sum=0.;sumr=0.;
11527: for (i=1; i<=nlstate;i++){
11528: sum+=mobaverage[(int)age][i][cptcod];
11529: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 11530: }
1.266 brouard 11531: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 11532: printf("Moving average B1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you decrease fage=%d?\n",cptcod, sum, (int) age, (int)fage);
1.266 brouard 11533: } /* end bad */
11534: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
11535: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 11536: printf("Moving average B2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase fage=%d\n",cptcod,sumr, (int)age, (int)fage);
1.222 brouard 11537: } /* end bad */
11538: }/* age */
1.266 brouard 11539:
1.222 brouard 11540:
11541: for (age=bage; age<=fage; age++){
1.235 brouard 11542: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 11543: sumnewp[cptcod]=0.;
11544: sumnewm[cptcod]=0.;
11545: for (i=1; i<=nlstate;i++){
11546: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
11547: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
11548: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
11549: }
11550: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
11551: }
11552: /* printf("\n"); */
11553: /* } */
1.266 brouard 11554:
1.222 brouard 11555: /* brutal averaging */
1.266 brouard 11556: /* for (i=1; i<=nlstate;i++){ */
11557: /* for (age=1; age<=bage; age++){ */
11558: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
11559: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
11560: /* } */
11561: /* for (age=fage; age<=AGESUP; age++){ */
11562: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
11563: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
11564: /* } */
11565: /* } /\* end i status *\/ */
11566: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
11567: /* for (age=1; age<=AGESUP; age++){ */
11568: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
11569: /* mobaverage[(int)age][i][cptcod]=0.; */
11570: /* } */
11571: /* } */
1.222 brouard 11572: }/* end cptcod */
1.266 brouard 11573: free_vector(agemaxgoodr,1, ncovcombmax);
11574: free_vector(agemaxgood,1, ncovcombmax);
11575: free_vector(agemingood,1, ncovcombmax);
11576: free_vector(agemingoodr,1, ncovcombmax);
11577: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 11578: free_vector(sumnewm,1, ncovcombmax);
11579: free_vector(sumnewp,1, ncovcombmax);
11580: return 0;
11581: }/* End movingaverage */
1.218 brouard 11582:
1.126 brouard 11583:
1.296 brouard 11584:
1.126 brouard 11585: /************** Forecasting ******************/
1.296 brouard 11586: /* void prevforecast(char fileres[], double dateintmean, double anprojd, double mprojd, double jprojd, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double anprojf, double p[], int cptcoveff)*/
11587: void prevforecast(char fileres[], double dateintmean, double dateprojd, double dateprojf, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double p[], int cptcoveff){
11588: /* dateintemean, mean date of interviews
11589: dateprojd, year, month, day of starting projection
11590: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 11591: agemin, agemax range of age
11592: dateprev1 dateprev2 range of dates during which prevalence is computed
11593: */
1.296 brouard 11594: /* double anprojd, mprojd, jprojd; */
11595: /* double anprojf, mprojf, jprojf; */
1.359 brouard 11596: int yearp, stepsize, hstepm, nhstepm, j, k, i, h, nres=0;
1.126 brouard 11597: double agec; /* generic age */
1.359 brouard 11598: double agelim, ppij;
11599: /*double *popcount;*/
1.126 brouard 11600: double ***p3mat;
1.218 brouard 11601: /* double ***mobaverage; */
1.126 brouard 11602: char fileresf[FILENAMELENGTH];
11603:
11604: agelim=AGESUP;
1.211 brouard 11605: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
11606: in each health status at the date of interview (if between dateprev1 and dateprev2).
11607: We still use firstpass and lastpass as another selection.
11608: */
1.214 brouard 11609: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
11610: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 11611:
1.201 brouard 11612: strcpy(fileresf,"F_");
11613: strcat(fileresf,fileresu);
1.126 brouard 11614: if((ficresf=fopen(fileresf,"w"))==NULL) {
11615: printf("Problem with forecast resultfile: %s\n", fileresf);
11616: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
11617: }
1.235 brouard 11618: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
11619: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 11620:
1.225 brouard 11621: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 11622:
11623:
11624: stepsize=(int) (stepm+YEARM-1)/YEARM;
11625: if (stepm<=12) stepsize=1;
11626: if(estepm < stepm){
11627: printf ("Problem %d lower than %d\n",estepm, stepm);
11628: }
1.270 brouard 11629: else{
11630: hstepm=estepm;
11631: }
11632: if(estepm > stepm){ /* Yes every two year */
11633: stepsize=2;
11634: }
1.296 brouard 11635: hstepm=hstepm/stepm;
1.126 brouard 11636:
1.296 brouard 11637:
11638: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
11639: /* fractional in yp1 *\/ */
11640: /* aintmean=yp; */
11641: /* yp2=modf((yp1*12),&yp); */
11642: /* mintmean=yp; */
11643: /* yp1=modf((yp2*30.5),&yp); */
11644: /* jintmean=yp; */
11645: /* if(jintmean==0) jintmean=1; */
11646: /* if(mintmean==0) mintmean=1; */
1.126 brouard 11647:
1.296 brouard 11648:
11649: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
11650: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
11651: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.351 brouard 11652: /* i1=pow(2,cptcoveff); */
11653: /* if (cptcovn < 1){i1=1;} */
1.126 brouard 11654:
1.296 brouard 11655: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 11656:
11657: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 11658:
1.126 brouard 11659: /* if (h==(int)(YEARM*yearp)){ */
1.351 brouard 11660: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11661: k=TKresult[nres];
11662: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
11663: /* for(k=1; k<=i1;k++){ /\* We want to find the combination k corresponding to the values of the dummies given in this resut line (to be cleaned one day) *\/ */
11664: /* if(i1 != 1 && TKresult[nres]!= k) */
11665: /* continue; */
11666: /* if(invalidvarcomb[k]){ */
11667: /* printf("\nCombination (%d) projection ignored because no cases \n",k); */
11668: /* continue; */
11669: /* } */
1.227 brouard 11670: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
1.351 brouard 11671: for(j=1;j<=cptcovs;j++){
11672: /* for(j=1;j<=cptcoveff;j++) { */
11673: /* /\* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); *\/ */
11674: /* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11675: /* } */
11676: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
11677: /* fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11678: /* } */
11679: fprintf(ficresf," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.235 brouard 11680: }
1.351 brouard 11681:
1.227 brouard 11682: fprintf(ficresf," yearproj age");
11683: for(j=1; j<=nlstate+ndeath;j++){
11684: for(i=1; i<=nlstate;i++)
11685: fprintf(ficresf," p%d%d",i,j);
11686: fprintf(ficresf," wp.%d",j);
11687: }
1.296 brouard 11688: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 11689: fprintf(ficresf,"\n");
1.296 brouard 11690: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 11691: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
11692: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 11693: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
11694: nhstepm = nhstepm/hstepm;
11695: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11696: oldm=oldms;savm=savms;
1.268 brouard 11697: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 11698: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 11699: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 11700: for (h=0; h<=nhstepm; h++){
11701: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 11702: break;
11703: }
11704: }
11705: fprintf(ficresf,"\n");
1.351 brouard 11706: /* for(j=1;j<=cptcoveff;j++) */
11707: for(j=1;j<=cptcovs;j++)
11708: fprintf(ficresf,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.332 brouard 11709: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Tvaraff not correct *\/ */
1.351 brouard 11710: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* TnsdVar[Tvaraff] correct *\/ */
1.296 brouard 11711: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 11712:
11713: for(j=1; j<=nlstate+ndeath;j++) {
11714: ppij=0.;
11715: for(i=1; i<=nlstate;i++) {
1.278 brouard 11716: if (mobilav>=1)
11717: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
11718: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
11719: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
11720: }
1.268 brouard 11721: fprintf(ficresf," %.3f", p3mat[i][j][h]);
11722: } /* end i */
11723: fprintf(ficresf," %.3f", ppij);
11724: }/* end j */
1.227 brouard 11725: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11726: } /* end agec */
1.266 brouard 11727: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
11728: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 11729: } /* end yearp */
11730: } /* end k */
1.219 brouard 11731:
1.126 brouard 11732: fclose(ficresf);
1.215 brouard 11733: printf("End of Computing forecasting \n");
11734: fprintf(ficlog,"End of Computing forecasting\n");
11735:
1.126 brouard 11736: }
11737:
1.269 brouard 11738: /************** Back Forecasting ******************/
1.296 brouard 11739: /* void prevbackforecast(char fileres[], double ***prevacurrent, double anback1, double mback1, double jback1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double anback2, double p[], int cptcoveff){ */
11740: void prevbackforecast(char fileres[], double ***prevacurrent, double dateintmean, double dateprojd, double dateprojf, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double p[], int cptcoveff){
11741: /* back1, year, month, day of starting backprojection
1.267 brouard 11742: agemin, agemax range of age
11743: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 11744: anback2 year of end of backprojection (same day and month as back1).
11745: prevacurrent and prev are prevalences.
1.267 brouard 11746: */
1.359 brouard 11747: int yearp, stepsize, hstepm, nhstepm, j, k, i, h, nres=0;
1.267 brouard 11748: double agec; /* generic age */
1.359 brouard 11749: double agelim, ppij, ppi; /* ,jintmean,mintmean,aintmean;*/
11750: /*double *popcount;*/
1.267 brouard 11751: double ***p3mat;
11752: /* double ***mobaverage; */
11753: char fileresfb[FILENAMELENGTH];
11754:
1.268 brouard 11755: agelim=AGEINF;
1.267 brouard 11756: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
11757: in each health status at the date of interview (if between dateprev1 and dateprev2).
11758: We still use firstpass and lastpass as another selection.
11759: */
11760: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
11761: /* firstpass, lastpass, stepm, weightopt, model); */
11762:
11763: /*Do we need to compute prevalence again?*/
11764:
11765: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11766:
11767: strcpy(fileresfb,"FB_");
11768: strcat(fileresfb,fileresu);
11769: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
11770: printf("Problem with back forecast resultfile: %s\n", fileresfb);
11771: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
11772: }
11773: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
11774: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
11775:
11776: if (cptcoveff==0) ncodemax[cptcoveff]=1;
11777:
11778:
11779: stepsize=(int) (stepm+YEARM-1)/YEARM;
11780: if (stepm<=12) stepsize=1;
11781: if(estepm < stepm){
11782: printf ("Problem %d lower than %d\n",estepm, stepm);
11783: }
1.270 brouard 11784: else{
11785: hstepm=estepm;
11786: }
11787: if(estepm >= stepm){ /* Yes every two year */
11788: stepsize=2;
11789: }
1.267 brouard 11790:
11791: hstepm=hstepm/stepm;
1.296 brouard 11792: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
11793: /* fractional in yp1 *\/ */
11794: /* aintmean=yp; */
11795: /* yp2=modf((yp1*12),&yp); */
11796: /* mintmean=yp; */
11797: /* yp1=modf((yp2*30.5),&yp); */
11798: /* jintmean=yp; */
11799: /* if(jintmean==0) jintmean=1; */
11800: /* if(mintmean==0) jintmean=1; */
1.267 brouard 11801:
1.351 brouard 11802: /* i1=pow(2,cptcoveff); */
11803: /* if (cptcovn < 1){i1=1;} */
1.267 brouard 11804:
1.296 brouard 11805: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
11806: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 11807:
11808: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
11809:
1.351 brouard 11810: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11811: k=TKresult[nres];
11812: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
11813: /* for(k=1; k<=i1;k++){ */
11814: /* if(i1 != 1 && TKresult[nres]!= k) */
11815: /* continue; */
11816: /* if(invalidvarcomb[k]){ */
11817: /* printf("\nCombination (%d) projection ignored because no cases \n",k); */
11818: /* continue; */
11819: /* } */
1.268 brouard 11820: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.351 brouard 11821: for(j=1;j<=cptcovs;j++){
11822: /* for(j=1;j<=cptcoveff;j++) { */
11823: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11824: /* } */
11825: fprintf(ficresfb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.267 brouard 11826: }
1.351 brouard 11827: /* fprintf(ficrespij,"******\n"); */
11828: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
11829: /* fprintf(ficresfb," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11830: /* } */
1.267 brouard 11831: fprintf(ficresfb," yearbproj age");
11832: for(j=1; j<=nlstate+ndeath;j++){
11833: for(i=1; i<=nlstate;i++)
1.268 brouard 11834: fprintf(ficresfb," b%d%d",i,j);
11835: fprintf(ficresfb," b.%d",j);
1.267 brouard 11836: }
1.296 brouard 11837: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 11838: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
11839: fprintf(ficresfb,"\n");
1.296 brouard 11840: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 11841: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 11842: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
11843: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 11844: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 11845: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 11846: nhstepm = nhstepm/hstepm;
11847: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11848: oldm=oldms;savm=savms;
1.268 brouard 11849: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 11850: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 11851: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 11852: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
11853: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
11854: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 11855: for (h=0; h<=nhstepm; h++){
1.268 brouard 11856: if (h*hstepm/YEARM*stepm ==-yearp) {
11857: break;
11858: }
11859: }
11860: fprintf(ficresfb,"\n");
1.351 brouard 11861: /* for(j=1;j<=cptcoveff;j++) */
11862: for(j=1;j<=cptcovs;j++)
11863: fprintf(ficresfb,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11864: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.296 brouard 11865: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 11866: for(i=1; i<=nlstate+ndeath;i++) {
11867: ppij=0.;ppi=0.;
11868: for(j=1; j<=nlstate;j++) {
11869: /* if (mobilav==1) */
1.269 brouard 11870: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
11871: ppi=ppi+prevacurrent[(int)agec][j][k];
11872: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
11873: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 11874: /* else { */
11875: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
11876: /* } */
1.268 brouard 11877: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
11878: } /* end j */
11879: if(ppi <0.99){
11880: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
11881: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
11882: }
11883: fprintf(ficresfb," %.3f", ppij);
11884: }/* end j */
1.267 brouard 11885: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11886: } /* end agec */
11887: } /* end yearp */
11888: } /* end k */
1.217 brouard 11889:
1.267 brouard 11890: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 11891:
1.267 brouard 11892: fclose(ficresfb);
11893: printf("End of Computing Back forecasting \n");
11894: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 11895:
1.267 brouard 11896: }
1.217 brouard 11897:
1.269 brouard 11898: /* Variance of prevalence limit: varprlim */
11899: void varprlim(char fileresu[], int nresult, double ***prevacurrent, int mobilavproj, double bage, double fage, double **prlim, int *ncvyearp, double ftolpl, double p[], double **matcov, double *delti, int stepm, int cptcoveff){
1.288 brouard 11900: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 11901:
11902: char fileresvpl[FILENAMELENGTH];
11903: FILE *ficresvpl;
11904: double **oldm, **savm;
11905: double **varpl; /* Variances of prevalence limits by age */
11906: int i1, k, nres, j ;
11907:
11908: strcpy(fileresvpl,"VPL_");
11909: strcat(fileresvpl,fileresu);
11910: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 11911: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 11912: exit(0);
11913: }
1.288 brouard 11914: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11915: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 11916:
11917: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11918: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11919:
11920: i1=pow(2,cptcoveff);
11921: if (cptcovn < 1){i1=1;}
11922:
1.337 brouard 11923: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11924: k=TKresult[nres];
1.338 brouard 11925: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 11926: /* for(k=1; k<=i1;k++){ /\* We find the combination equivalent to result line values of dummies *\/ */
1.269 brouard 11927: if(i1 != 1 && TKresult[nres]!= k)
11928: continue;
11929: fprintf(ficresvpl,"\n#****** ");
11930: printf("\n#****** ");
11931: fprintf(ficlog,"\n#****** ");
1.337 brouard 11932: for(j=1;j<=cptcovs;j++) {
11933: fprintf(ficresvpl,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11934: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11935: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11936: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11937: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.269 brouard 11938: }
1.337 brouard 11939: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
11940: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
11941: /* fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
11942: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
11943: /* } */
1.269 brouard 11944: fprintf(ficresvpl,"******\n");
11945: printf("******\n");
11946: fprintf(ficlog,"******\n");
11947:
11948: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11949: oldm=oldms;savm=savms;
11950: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
11951: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
11952: /*}*/
11953: }
11954:
11955: fclose(ficresvpl);
1.288 brouard 11956: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
11957: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 11958:
11959: }
11960: /* Variance of back prevalence: varbprlim */
11961: void varbprlim(char fileresu[], int nresult, double ***prevacurrent, int mobilavproj, double bage, double fage, double **bprlim, int *ncvyearp, double ftolpl, double p[], double **matcov, double *delti, int stepm, int cptcoveff){
11962: /*------- Variance of back (stable) prevalence------*/
11963:
11964: char fileresvbl[FILENAMELENGTH];
11965: FILE *ficresvbl;
11966:
11967: double **oldm, **savm;
11968: double **varbpl; /* Variances of back prevalence limits by age */
11969: int i1, k, nres, j ;
11970:
11971: strcpy(fileresvbl,"VBL_");
11972: strcat(fileresvbl,fileresu);
11973: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
11974: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
11975: exit(0);
11976: }
11977: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
11978: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
11979:
11980:
11981: i1=pow(2,cptcoveff);
11982: if (cptcovn < 1){i1=1;}
11983:
1.337 brouard 11984: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11985: k=TKresult[nres];
1.338 brouard 11986: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 11987: /* for(k=1; k<=i1;k++){ */
11988: /* if(i1 != 1 && TKresult[nres]!= k) */
11989: /* continue; */
1.269 brouard 11990: fprintf(ficresvbl,"\n#****** ");
11991: printf("\n#****** ");
11992: fprintf(ficlog,"\n#****** ");
1.337 brouard 11993: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */
1.338 brouard 11994: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
11995: fprintf(ficresvbl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
11996: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
1.337 brouard 11997: /* for(j=1;j<=cptcoveff;j++) { */
11998: /* fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11999: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12000: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12001: /* } */
12002: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
12003: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12004: /* fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12005: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
1.269 brouard 12006: }
12007: fprintf(ficresvbl,"******\n");
12008: printf("******\n");
12009: fprintf(ficlog,"******\n");
12010:
12011: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
12012: oldm=oldms;savm=savms;
12013:
12014: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
12015: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
12016: /*}*/
12017: }
12018:
12019: fclose(ficresvbl);
12020: printf("done variance-covariance of back prevalence\n");fflush(stdout);
12021: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
12022:
12023: } /* End of varbprlim */
12024:
1.126 brouard 12025: /************** Forecasting *****not tested NB*************/
1.227 brouard 12026: /* void populforecast(char fileres[], double anpyram,double mpyram,double jpyram,double ageminpar, double agemax,double dateprev1, double dateprev2s, int mobilav, double agedeb, double fage, int popforecast, char popfile[], double anpyram1,double p[], int i2){ */
1.126 brouard 12027:
1.227 brouard 12028: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
12029: /* int *popage; */
12030: /* double calagedatem, agelim, kk1, kk2; */
12031: /* double *popeffectif,*popcount; */
12032: /* double ***p3mat,***tabpop,***tabpopprev; */
12033: /* /\* double ***mobaverage; *\/ */
12034: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 12035:
1.227 brouard 12036: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
12037: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
12038: /* agelim=AGESUP; */
12039: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 12040:
1.227 brouard 12041: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 12042:
12043:
1.227 brouard 12044: /* strcpy(filerespop,"POP_"); */
12045: /* strcat(filerespop,fileresu); */
12046: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
12047: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
12048: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
12049: /* } */
12050: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
12051: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 12052:
1.227 brouard 12053: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 12054:
1.227 brouard 12055: /* /\* if (mobilav!=0) { *\/ */
12056: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
12057: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
12058: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
12059: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
12060: /* /\* } *\/ */
12061: /* /\* } *\/ */
1.126 brouard 12062:
1.227 brouard 12063: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
12064: /* if (stepm<=12) stepsize=1; */
1.126 brouard 12065:
1.227 brouard 12066: /* agelim=AGESUP; */
1.126 brouard 12067:
1.227 brouard 12068: /* hstepm=1; */
12069: /* hstepm=hstepm/stepm; */
1.218 brouard 12070:
1.227 brouard 12071: /* if (popforecast==1) { */
12072: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
12073: /* printf("Problem with population file : %s\n",popfile);exit(0); */
12074: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
12075: /* } */
12076: /* popage=ivector(0,AGESUP); */
12077: /* popeffectif=vector(0,AGESUP); */
12078: /* popcount=vector(0,AGESUP); */
1.126 brouard 12079:
1.227 brouard 12080: /* i=1; */
12081: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 12082:
1.227 brouard 12083: /* imx=i; */
12084: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
12085: /* } */
1.218 brouard 12086:
1.227 brouard 12087: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
12088: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
12089: /* k=k+1; */
12090: /* fprintf(ficrespop,"\n#******"); */
12091: /* for(j=1;j<=cptcoveff;j++) { */
12092: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
12093: /* } */
12094: /* fprintf(ficrespop,"******\n"); */
12095: /* fprintf(ficrespop,"# Age"); */
12096: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
12097: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 12098:
1.227 brouard 12099: /* for (cpt=0; cpt<=0;cpt++) { */
12100: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 12101:
1.227 brouard 12102: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
12103: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
12104: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 12105:
1.227 brouard 12106: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
12107: /* oldm=oldms;savm=savms; */
12108: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 12109:
1.227 brouard 12110: /* for (h=0; h<=nhstepm; h++){ */
12111: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
12112: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
12113: /* } */
12114: /* for(j=1; j<=nlstate+ndeath;j++) { */
12115: /* kk1=0.;kk2=0; */
12116: /* for(i=1; i<=nlstate;i++) { */
12117: /* if (mobilav==1) */
12118: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
12119: /* else { */
12120: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
12121: /* } */
12122: /* } */
12123: /* if (h==(int)(calagedatem+12*cpt)){ */
12124: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
12125: /* /\*fprintf(ficrespop," %.3f", kk1); */
12126: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
12127: /* } */
12128: /* } */
12129: /* for(i=1; i<=nlstate;i++){ */
12130: /* kk1=0.; */
12131: /* for(j=1; j<=nlstate;j++){ */
12132: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
12133: /* } */
12134: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
12135: /* } */
1.218 brouard 12136:
1.227 brouard 12137: /* if (h==(int)(calagedatem+12*cpt)) */
12138: /* for(j=1; j<=nlstate;j++) */
12139: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
12140: /* } */
12141: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
12142: /* } */
12143: /* } */
1.218 brouard 12144:
1.227 brouard 12145: /* /\******\/ */
1.218 brouard 12146:
1.227 brouard 12147: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
12148: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
12149: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
12150: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
12151: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 12152:
1.227 brouard 12153: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
12154: /* oldm=oldms;savm=savms; */
12155: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
12156: /* for (h=0; h<=nhstepm; h++){ */
12157: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
12158: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
12159: /* } */
12160: /* for(j=1; j<=nlstate+ndeath;j++) { */
12161: /* kk1=0.;kk2=0; */
12162: /* for(i=1; i<=nlstate;i++) { */
12163: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
12164: /* } */
12165: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
12166: /* } */
12167: /* } */
12168: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
12169: /* } */
12170: /* } */
12171: /* } */
12172: /* } */
1.218 brouard 12173:
1.227 brouard 12174: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 12175:
1.227 brouard 12176: /* if (popforecast==1) { */
12177: /* free_ivector(popage,0,AGESUP); */
12178: /* free_vector(popeffectif,0,AGESUP); */
12179: /* free_vector(popcount,0,AGESUP); */
12180: /* } */
12181: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
12182: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
12183: /* fclose(ficrespop); */
12184: /* } /\* End of popforecast *\/ */
1.218 brouard 12185:
1.126 brouard 12186: int fileappend(FILE *fichier, char *optionfich)
12187: {
12188: if((fichier=fopen(optionfich,"a"))==NULL) {
12189: printf("Problem with file: %s\n", optionfich);
12190: fprintf(ficlog,"Problem with file: %s\n", optionfich);
12191: return (0);
12192: }
12193: fflush(fichier);
12194: return (1);
12195: }
12196:
12197:
12198: /**************** function prwizard **********************/
12199: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
12200: {
12201:
12202: /* Wizard to print covariance matrix template */
12203:
1.164 brouard 12204: char ca[32], cb[32];
12205: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 12206: int numlinepar;
12207:
12208: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12209: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12210: for(i=1; i <=nlstate; i++){
12211: jj=0;
12212: for(j=1; j <=nlstate+ndeath; j++){
12213: if(j==i) continue;
12214: jj++;
12215: /*ca[0]= k+'a'-1;ca[1]='\0';*/
12216: printf("%1d%1d",i,j);
12217: fprintf(ficparo,"%1d%1d",i,j);
12218: for(k=1; k<=ncovmodel;k++){
12219: /* printf(" %lf",param[i][j][k]); */
12220: /* fprintf(ficparo," %lf",param[i][j][k]); */
12221: printf(" 0.");
12222: fprintf(ficparo," 0.");
12223: }
12224: printf("\n");
12225: fprintf(ficparo,"\n");
12226: }
12227: }
12228: printf("# Scales (for hessian or gradient estimation)\n");
12229: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
12230: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
12231: for(i=1; i <=nlstate; i++){
12232: jj=0;
12233: for(j=1; j <=nlstate+ndeath; j++){
12234: if(j==i) continue;
12235: jj++;
12236: fprintf(ficparo,"%1d%1d",i,j);
12237: printf("%1d%1d",i,j);
12238: fflush(stdout);
12239: for(k=1; k<=ncovmodel;k++){
12240: /* printf(" %le",delti3[i][j][k]); */
12241: /* fprintf(ficparo," %le",delti3[i][j][k]); */
12242: printf(" 0.");
12243: fprintf(ficparo," 0.");
12244: }
12245: numlinepar++;
12246: printf("\n");
12247: fprintf(ficparo,"\n");
12248: }
12249: }
12250: printf("# Covariance matrix\n");
12251: /* # 121 Var(a12)\n\ */
12252: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12253: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12254: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12255: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12256: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12257: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12258: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12259: fflush(stdout);
12260: fprintf(ficparo,"# Covariance matrix\n");
12261: /* # 121 Var(a12)\n\ */
12262: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12263: /* # ...\n\ */
12264: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12265:
12266: for(itimes=1;itimes<=2;itimes++){
12267: jj=0;
12268: for(i=1; i <=nlstate; i++){
12269: for(j=1; j <=nlstate+ndeath; j++){
12270: if(j==i) continue;
12271: for(k=1; k<=ncovmodel;k++){
12272: jj++;
12273: ca[0]= k+'a'-1;ca[1]='\0';
12274: if(itimes==1){
12275: printf("#%1d%1d%d",i,j,k);
12276: fprintf(ficparo,"#%1d%1d%d",i,j,k);
12277: }else{
12278: printf("%1d%1d%d",i,j,k);
12279: fprintf(ficparo,"%1d%1d%d",i,j,k);
12280: /* printf(" %.5le",matcov[i][j]); */
12281: }
12282: ll=0;
12283: for(li=1;li <=nlstate; li++){
12284: for(lj=1;lj <=nlstate+ndeath; lj++){
12285: if(lj==li) continue;
12286: for(lk=1;lk<=ncovmodel;lk++){
12287: ll++;
12288: if(ll<=jj){
12289: cb[0]= lk +'a'-1;cb[1]='\0';
12290: if(ll<jj){
12291: if(itimes==1){
12292: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12293: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12294: }else{
12295: printf(" 0.");
12296: fprintf(ficparo," 0.");
12297: }
12298: }else{
12299: if(itimes==1){
12300: printf(" Var(%s%1d%1d)",ca,i,j);
12301: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
12302: }else{
12303: printf(" 0.");
12304: fprintf(ficparo," 0.");
12305: }
12306: }
12307: }
12308: } /* end lk */
12309: } /* end lj */
12310: } /* end li */
12311: printf("\n");
12312: fprintf(ficparo,"\n");
12313: numlinepar++;
12314: } /* end k*/
12315: } /*end j */
12316: } /* end i */
12317: } /* end itimes */
12318:
12319: } /* end of prwizard */
12320: /******************* Gompertz Likelihood ******************************/
12321: double gompertz(double x[])
12322: {
1.302 brouard 12323: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 12324: int i,n=0; /* n is the size of the sample */
12325:
1.220 brouard 12326: for (i=1;i<=imx ; i++) {
1.126 brouard 12327: sump=sump+weight[i];
12328: /* sump=sump+1;*/
12329: num=num+1;
12330: }
1.302 brouard 12331: L=0.0;
12332: /* agegomp=AGEGOMP; */
1.126 brouard 12333: /* for (i=0; i<=imx; i++)
12334: if (wav[i]>0) printf("i=%d ageex=%lf agecens=%lf agedc=%lf cens=%d %d\n" ,i,ageexmed[i],agecens[i],agedc[i],cens[i],wav[i]);*/
12335:
1.302 brouard 12336: for (i=1;i<=imx ; i++) {
12337: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
12338: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
12339: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
12340: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
12341: * +
12342: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
12343: */
12344: if (wav[i] > 1 || agedc[i] < AGESUP) {
12345: if (cens[i] == 1){
12346: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
12347: } else if (cens[i] == 0){
1.126 brouard 12348: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.362 brouard 12349: +log(fabs(x[1])/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
12350: /* +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM); */ /* To be seen */
1.302 brouard 12351: } else
12352: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 12353: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 12354: L=L+A*weight[i];
1.126 brouard 12355: /* printf("\ni=%d A=%f L=%lf x[1]=%lf x[2]=%lf ageex=%lf agecens=%lf cens=%d agedc=%lf weight=%lf\n",i,A,L,x[1],x[2],ageexmed[i]*12,agecens[i]*12,cens[i],agedc[i]*12,weight[i]);*/
1.302 brouard 12356: }
12357: }
1.126 brouard 12358:
1.302 brouard 12359: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 12360:
12361: return -2*L*num/sump;
12362: }
12363:
1.136 brouard 12364: #ifdef GSL
12365: /******************* Gompertz_f Likelihood ******************************/
12366: double gompertz_f(const gsl_vector *v, void *params)
12367: {
1.302 brouard 12368: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 12369: double *x= (double *) v->data;
12370: int i,n=0; /* n is the size of the sample */
12371:
12372: for (i=0;i<=imx-1 ; i++) {
12373: sump=sump+weight[i];
12374: /* sump=sump+1;*/
12375: num=num+1;
12376: }
12377:
12378:
12379: /* for (i=0; i<=imx; i++)
12380: if (wav[i]>0) printf("i=%d ageex=%lf agecens=%lf agedc=%lf cens=%d %d\n" ,i,ageexmed[i],agecens[i],agedc[i],cens[i],wav[i]);*/
12381: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
12382: for (i=1;i<=imx ; i++)
12383: {
12384: if (cens[i] == 1 && wav[i]>1)
12385: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
12386:
12387: if (cens[i] == 0 && wav[i]>1)
12388: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
12389: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
12390:
12391: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
12392: if (wav[i] > 1 ) { /* ??? */
12393: LL=LL+A*weight[i];
12394: /* printf("\ni=%d A=%f L=%lf x[1]=%lf x[2]=%lf ageex=%lf agecens=%lf cens=%d agedc=%lf weight=%lf\n",i,A,L,x[1],x[2],ageexmed[i]*12,agecens[i]*12,cens[i],agedc[i]*12,weight[i]);*/
12395: }
12396: }
12397:
12398: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
12399: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
12400:
12401: return -2*LL*num/sump;
12402: }
12403: #endif
12404:
1.126 brouard 12405: /******************* Printing html file ***********/
1.201 brouard 12406: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 12407: int lastpass, int stepm, int weightopt, char model[],\
12408: int imx, double p[],double **matcov,double agemortsup){
12409: int i,k;
12410:
12411: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
12412: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
12413: for (i=1;i<=2;i++)
12414: fprintf(fichtm," p[%d] = %lf [%f ; %f]<br>\n",i,p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.199 brouard 12415: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 12416: fprintf(fichtm,"</ul>");
12417:
12418: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
12419:
12420: fprintf(fichtm,"\nAge l<inf>x</inf> q<inf>x</inf> d(x,x+1) L<inf>x</inf> T<inf>x</inf> e<infx</inf><br>");
12421:
12422: for (k=agegomp;k<(agemortsup-2);k++)
12423: fprintf(fichtm,"%d %.0lf %lf %.0lf %.0lf %.0lf %lf<br>\n",k,lsurv[k],p[1]*exp(p[2]*(k-agegomp)),(p[1]*exp(p[2]*(k-agegomp)))*lsurv[k],lpop[k],tpop[k],tpop[k]/lsurv[k]);
12424:
12425:
12426: fflush(fichtm);
12427: }
12428:
12429: /******************* Gnuplot file **************/
1.201 brouard 12430: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 12431:
12432: char dirfileres[132],optfileres[132];
1.164 brouard 12433:
1.359 brouard 12434: /*int ng;*/
1.126 brouard 12435:
12436:
12437: /*#ifdef windows */
12438: fprintf(ficgp,"cd \"%s\" \n",pathc);
12439: /*#endif */
12440:
12441:
12442: strcpy(dirfileres,optionfilefiname);
12443: strcpy(optfileres,"vpl");
1.199 brouard 12444: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 12445: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 12446: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 12447: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 12448: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
12449:
12450: }
12451:
1.136 brouard 12452: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
12453: {
1.126 brouard 12454:
1.136 brouard 12455: /*-------- data file ----------*/
12456: FILE *fic;
12457: char dummy[]=" ";
1.359 brouard 12458: int i = 0, j = 0, n = 0, iv = 0;/* , v;*/
1.223 brouard 12459: int lstra;
1.136 brouard 12460: int linei, month, year,iout;
1.302 brouard 12461: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 12462: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 12463: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 12464: char *stratrunc;
1.223 brouard 12465:
1.349 brouard 12466: /* DummyV=ivector(-1,NCOVMAX); /\* 1 to 3 *\/ */
12467: /* FixedV=ivector(-1,NCOVMAX); /\* 1 to 3 *\/ */
1.339 brouard 12468:
12469: ncovcolt=ncovcol+nqv+ntv+nqtv; /* total of covariates in the data, not in the model equation */
12470:
1.136 brouard 12471: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 12472: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
12473: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 12474: }
1.126 brouard 12475:
1.302 brouard 12476: /* Is it a BOM UTF-8 Windows file? */
12477: /* First data line */
12478: linei=0;
12479: while(fgets(line, MAXLINE, fic)) {
12480: noffset=0;
12481: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
12482: {
12483: noffset=noffset+3;
12484: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
12485: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
12486: fflush(ficlog); return 1;
12487: }
12488: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
12489: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
12490: {
12491: noffset=noffset+2;
1.304 brouard 12492: printf("# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);fflush(stdout);
12493: fprintf(ficlog,"# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);
1.302 brouard 12494: fflush(ficlog); return 1;
12495: }
12496: else if( line[0] == 0 && line[1] == 0)
12497: {
12498: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
12499: noffset=noffset+4;
1.304 brouard 12500: printf("# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);fflush(stdout);
12501: fprintf(ficlog,"# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);
1.302 brouard 12502: fflush(ficlog); return 1;
12503: }
12504: } else{
12505: ;/*printf(" Not a BOM file\n");*/
12506: }
12507: /* If line starts with a # it is a comment */
12508: if (line[noffset] == '#') {
12509: linei=linei+1;
12510: break;
12511: }else{
12512: break;
12513: }
12514: }
12515: fclose(fic);
12516: if((fic=fopen(datafile,"r"))==NULL) {
12517: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
12518: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
12519: }
12520: /* Not a Bom file */
12521:
1.136 brouard 12522: i=1;
12523: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
12524: linei=linei+1;
12525: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
12526: if(line[j] == '\t')
12527: line[j] = ' ';
12528: }
12529: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
12530: ;
12531: };
12532: line[j+1]=0; /* Trims blanks at end of line */
12533: if(line[0]=='#'){
12534: fprintf(ficlog,"Comment line\n%s\n",line);
12535: printf("Comment line\n%s\n",line);
12536: continue;
12537: }
12538: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 12539: strcpy(line, linetmp);
1.223 brouard 12540:
12541: /* Loops on waves */
12542: for (j=maxwav;j>=1;j--){
12543: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 12544: cutv(stra, strb, line, ' ');
12545: if(strb[0]=='.') { /* Missing value */
12546: lval=-1;
12547: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
1.341 brouard 12548: cotvar[j][ncovcol+nqv+ntv+iv][i]=-1; /* For performance reasons */
1.238 brouard 12549: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
12550: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value out of %d measured at wave %d. If missing, you should remove this individual or impute a value. Exiting.\n", strb, linei,i,line,iv, nqtv, j);
12551: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value out of %d measured at wave %d. If missing, you should remove this individual or impute a value. Exiting.\n", strb, linei,i,line,iv, nqtv, j);fflush(ficlog);
12552: return 1;
12553: }
12554: }else{
12555: errno=0;
12556: /* what_kind_of_number(strb); */
12557: dval=strtod(strb,&endptr);
12558: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
12559: /* if(strb != endptr && *endptr == '\0') */
12560: /* dval=dlval; */
12561: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
12562: if( strb[0]=='\0' || (*endptr != '\0')){
12563: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value out of %d measured at wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line,iv, nqtv, j,maxwav);
12564: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value out of %d measured at wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line, iv, nqtv, j,maxwav);fflush(ficlog);
12565: return 1;
12566: }
12567: cotqvar[j][iv][i]=dval;
1.341 brouard 12568: cotvar[j][ncovcol+nqv+ntv+iv][i]=dval; /* because cotvar starts now at first ntv */
1.238 brouard 12569: }
12570: strcpy(line,stra);
1.223 brouard 12571: }/* end loop ntqv */
1.225 brouard 12572:
1.223 brouard 12573: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 12574: cutv(stra, strb, line, ' ');
12575: if(strb[0]=='.') { /* Missing value */
12576: lval=-1;
12577: }else{
12578: errno=0;
12579: lval=strtol(strb,&endptr,10);
12580: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
12581: if( strb[0]=='\0' || (*endptr != '\0')){
12582: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th dummy covariate out of %d measured at wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line,iv, ntv, j,maxwav);
12583: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d dummy covariate out of %d measured wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line,iv, ntv,j,maxwav);fflush(ficlog);
12584: return 1;
12585: }
12586: }
12587: if(lval <-1 || lval >1){
12588: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 12589: Should be a value of %d(nth) covariate of wave %d (0 should be the value for the reference and 1\n \
1.223 brouard 12590: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 12591: For example, for multinomial values like 1, 2 and 3,\n \
12592: build V1=0 V2=0 for the reference value (1),\n \
12593: V1=1 V2=0 for (2) \n \
1.223 brouard 12594: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 12595: output of IMaCh is often meaningless.\n \
1.319 brouard 12596: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 12597: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 12598: Should be a value of %d(nth) covariate of wave %d (0 should be the value for the reference and 1\n \
1.223 brouard 12599: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 12600: For example, for multinomial values like 1, 2 and 3,\n \
12601: build V1=0 V2=0 for the reference value (1),\n \
12602: V1=1 V2=0 for (2) \n \
1.223 brouard 12603: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 12604: output of IMaCh is often meaningless.\n \
1.319 brouard 12605: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 12606: return 1;
12607: }
1.341 brouard 12608: cotvar[j][ncovcol+nqv+iv][i]=(double)(lval);
1.238 brouard 12609: strcpy(line,stra);
1.223 brouard 12610: }/* end loop ntv */
1.225 brouard 12611:
1.223 brouard 12612: /* Statuses at wave */
1.137 brouard 12613: cutv(stra, strb, line, ' ');
1.223 brouard 12614: if(strb[0]=='.') { /* Missing value */
1.238 brouard 12615: lval=-1;
1.136 brouard 12616: }else{
1.238 brouard 12617: errno=0;
12618: lval=strtol(strb,&endptr,10);
12619: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
1.347 brouard 12620: if( strb[0]=='\0' || (*endptr != '\0' )){
12621: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a status of wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line,j,maxwav);
12622: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a status of wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line,j,maxwav);fflush(ficlog);
12623: return 1;
12624: }else if( lval==0 || lval > nlstate+ndeath){
1.348 brouard 12625: printf("Error in data around '%s' at line number %d for individual %d, '%s'\n Should be a state at wave %d. A state should be 1 to %d and not %ld.\n Fix your data file '%s'! Exiting.\n", strb, linei,i,line,j,nlstate+ndeath, lval, datafile);fflush(stdout);
12626: fprintf(ficlog,"Error in data around '%s' at line number %d for individual %d, '%s'\n Should be a state at wave %d. A state should be 1 to %d and not %ld.\n Fix your data file '%s'! Exiting.\n", strb, linei,i,line,j,nlstate+ndeath, lval, datafile); fflush(ficlog);
1.238 brouard 12627: return 1;
12628: }
1.136 brouard 12629: }
1.225 brouard 12630:
1.136 brouard 12631: s[j][i]=lval;
1.225 brouard 12632:
1.223 brouard 12633: /* Date of Interview */
1.136 brouard 12634: strcpy(line,stra);
12635: cutv(stra, strb,line,' ');
1.169 brouard 12636: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 12637: }
1.169 brouard 12638: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 12639: month=99;
12640: year=9999;
1.136 brouard 12641: }else{
1.225 brouard 12642: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of interview (mm/yyyy or .) at wave %d. Exiting.\n",strb, linei,i, line,j);
12643: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of interview (mm/yyyy or .) at wave %d. Exiting.\n",strb, linei,i, line,j);fflush(ficlog);
12644: return 1;
1.136 brouard 12645: }
12646: anint[j][i]= (double) year;
1.302 brouard 12647: mint[j][i]= (double)month;
12648: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
12649: /* printf("Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, mint[j][i],anint[j][i], moisnais[i],annais[i]); */
12650: /* fprintf(ficlog,"Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, mint[j][i],anint[j][i], moisnais[i],annais[i]); */
12651: /* } */
1.136 brouard 12652: strcpy(line,stra);
1.223 brouard 12653: } /* End loop on waves */
1.225 brouard 12654:
1.223 brouard 12655: /* Date of death */
1.136 brouard 12656: cutv(stra, strb,line,' ');
1.169 brouard 12657: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 12658: }
1.169 brouard 12659: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 12660: month=99;
12661: year=9999;
12662: }else{
1.141 brouard 12663: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of death (mm/yyyy or .). Exiting.\n",strb, linei,i,line);
1.225 brouard 12664: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of death (mm/yyyy or .). Exiting.\n",strb, linei,i,line);fflush(ficlog);
12665: return 1;
1.136 brouard 12666: }
12667: andc[i]=(double) year;
12668: moisdc[i]=(double) month;
12669: strcpy(line,stra);
12670:
1.223 brouard 12671: /* Date of birth */
1.136 brouard 12672: cutv(stra, strb,line,' ');
1.169 brouard 12673: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 12674: }
1.169 brouard 12675: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 12676: month=99;
12677: year=9999;
12678: }else{
1.141 brouard 12679: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of birth (mm/yyyy or .). Exiting.\n",strb, linei,i,line);
12680: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of birth (mm/yyyy or .). Exiting.\n",strb, linei,i,line);fflush(ficlog);
1.225 brouard 12681: return 1;
1.136 brouard 12682: }
12683: if (year==9999) {
1.141 brouard 12684: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of birth (mm/yyyy) but at least the year of birth should be given. Exiting.\n",strb, linei,i,line);
12685: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of birth (mm/yyyy) but at least the year of birth should be given. Exiting.\n",strb, linei,i,line);fflush(ficlog);
1.225 brouard 12686: return 1;
12687:
1.136 brouard 12688: }
12689: annais[i]=(double)(year);
1.302 brouard 12690: moisnais[i]=(double)(month);
12691: for (j=1;j<=maxwav;j++){
12692: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
12693: printf("Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, (int)mint[j][i],(int)anint[j][i], j,(int)moisnais[i],(int)annais[i]);
12694: fprintf(ficlog,"Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, (int)mint[j][i],(int)anint[j][i], j, (int)moisnais[i],(int)annais[i]);
12695: }
12696: }
12697:
1.136 brouard 12698: strcpy(line,stra);
1.225 brouard 12699:
1.223 brouard 12700: /* Sample weight */
1.136 brouard 12701: cutv(stra, strb,line,' ');
12702: errno=0;
12703: dval=strtod(strb,&endptr);
12704: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 12705: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
12706: fprintf(ficlog,"Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
1.136 brouard 12707: fflush(ficlog);
12708: return 1;
12709: }
12710: weight[i]=dval;
12711: strcpy(line,stra);
1.225 brouard 12712:
1.223 brouard 12713: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
12714: cutv(stra, strb, line, ' ');
12715: if(strb[0]=='.') { /* Missing value */
1.225 brouard 12716: lval=-1;
1.311 brouard 12717: coqvar[iv][i]=NAN;
12718: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 12719: }else{
1.225 brouard 12720: errno=0;
12721: /* what_kind_of_number(strb); */
12722: dval=strtod(strb,&endptr);
12723: /* if(strb != endptr && *endptr == '\0') */
12724: /* dval=dlval; */
12725: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
12726: if( strb[0]=='\0' || (*endptr != '\0')){
12727: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value (out of %d) constant for all waves. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line, iv, nqv, maxwav);
12728: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value (out of %d) constant for all waves. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line, iv, nqv, maxwav);fflush(ficlog);
12729: return 1;
12730: }
12731: coqvar[iv][i]=dval;
1.226 brouard 12732: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 12733: }
12734: strcpy(line,stra);
12735: }/* end loop nqv */
1.136 brouard 12736:
1.223 brouard 12737: /* Covariate values */
1.136 brouard 12738: for (j=ncovcol;j>=1;j--){
12739: cutv(stra, strb,line,' ');
1.223 brouard 12740: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 12741: lval=-1;
1.136 brouard 12742: }else{
1.225 brouard 12743: errno=0;
12744: lval=strtol(strb,&endptr,10);
12745: if( strb[0]=='\0' || (*endptr != '\0')){
12746: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\nShould be a covariate value (=0 for the reference or 1 for alternative). Exiting.\n",lval, linei,i, line);
12747: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\nShould be a covariate value (=0 for the reference or 1 for alternative). Exiting.\n",lval, linei,i, line);fflush(ficlog);
12748: return 1;
12749: }
1.136 brouard 12750: }
12751: if(lval <-1 || lval >1){
1.225 brouard 12752: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 12753: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
12754: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 12755: For example, for multinomial values like 1, 2 and 3,\n \
12756: build V1=0 V2=0 for the reference value (1),\n \
12757: V1=1 V2=0 for (2) \n \
1.136 brouard 12758: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 12759: output of IMaCh is often meaningless.\n \
1.136 brouard 12760: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 12761: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 12762: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
12763: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 12764: For example, for multinomial values like 1, 2 and 3,\n \
12765: build V1=0 V2=0 for the reference value (1),\n \
12766: V1=1 V2=0 for (2) \n \
1.136 brouard 12767: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 12768: output of IMaCh is often meaningless.\n \
1.136 brouard 12769: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 12770: return 1;
1.136 brouard 12771: }
12772: covar[j][i]=(double)(lval);
12773: strcpy(line,stra);
12774: }
12775: lstra=strlen(stra);
1.225 brouard 12776:
1.136 brouard 12777: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
12778: stratrunc = &(stra[lstra-9]);
12779: num[i]=atol(stratrunc);
12780: }
12781: else
12782: num[i]=atol(stra);
12783: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
12784: printf("%ld %.lf %.lf %.lf %.lf/%.lf %.lf/%.lf %.lf/%.lf %d %.lf/%.lf %d %.lf/%.lf %d %.lf/%.lf %d\n",num[i],(covar[1][i]), (covar[2][i]),weight[i], (moisnais[i]), (annais[i]), (moisdc[i]), (andc[i]), (mint[1][i]), (anint[1][i]), (s[1][i]), (mint[2][i]), (anint[2][i]), (s[2][i]), (mint[3][i]), (anint[3][i]), (s[3][i]), (mint[4][i]), (anint[4][i]), (s[4][i])); ij=ij+1;}*/
12785:
12786: i=i+1;
12787: } /* End loop reading data */
1.225 brouard 12788:
1.136 brouard 12789: *imax=i-1; /* Number of individuals */
12790: fclose(fic);
1.225 brouard 12791:
1.136 brouard 12792: return (0);
1.164 brouard 12793: /* endread: */
1.225 brouard 12794: printf("Exiting readdata: ");
12795: fclose(fic);
12796: return (1);
1.223 brouard 12797: }
1.126 brouard 12798:
1.234 brouard 12799: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 12800: char *p1 = *stri, *p2 = *stri;
1.235 brouard 12801: while (*p2 == ' ')
1.234 brouard 12802: p2++;
12803: /* while ((*p1++ = *p2++) !=0) */
12804: /* ; */
12805: /* do */
12806: /* while (*p2 == ' ') */
12807: /* p2++; */
12808: /* while (*p1++ == *p2++); */
12809: *stri=p2;
1.145 brouard 12810: }
12811:
1.330 brouard 12812: int decoderesult( char resultline[], int nres)
1.230 brouard 12813: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
12814: {
1.235 brouard 12815: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 12816: char resultsav[MAXLINE];
1.330 brouard 12817: /* int resultmodel[MAXLINE]; */
1.334 brouard 12818: /* int modelresult[MAXLINE]; */
1.230 brouard 12819: char stra[80], strb[80], strc[80], strd[80],stre[80];
12820:
1.234 brouard 12821: removefirstspace(&resultline);
1.332 brouard 12822: printf("decoderesult:%s\n",resultline);
1.230 brouard 12823:
1.332 brouard 12824: strcpy(resultsav,resultline);
1.342 brouard 12825: /* printf("Decoderesult resultsav=\"%s\" resultline=\"%s\"\n", resultsav, resultline); */
1.230 brouard 12826: if (strlen(resultsav) >1){
1.334 brouard 12827: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' in this resultline */
1.230 brouard 12828: }
1.353 brouard 12829: if(j == 0 && cptcovs== 0){ /* Resultline but no = and no covariate in the model */
1.253 brouard 12830: TKresult[nres]=0; /* Combination for the nresult and the model */
12831: return (0);
12832: }
1.234 brouard 12833: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.353 brouard 12834: fprintf(ficlog,"ERROR: the number of variables in the resultline which is %d, differs from the number %d of single variables used in the model line, 1+age+%s.\n",j, cptcovs, model);fflush(ficlog);
12835: printf("ERROR: the number of variables in the resultline which is %d, differs from the number %d of single variables used in the model line, 1+age+%s.\n",j, cptcovs, model);fflush(stdout);
12836: if(j==0)
12837: return 1;
1.234 brouard 12838: }
1.334 brouard 12839: for(k=1; k<=j;k++){ /* Loop on any covariate of the RESULT LINE */
1.234 brouard 12840: if(nbocc(resultsav,'=') >1){
1.318 brouard 12841: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' ' (stra is the rest of the resultline to be analyzed in the next loop *//* resultsav= "V4=1 V5=25.1 V3=0" stra= "V5=25.1 V3=0" strb= "V4=1" */
1.332 brouard 12842: /* If resultsav= "V4= 1 V5=25.1 V3=0" with a blank then strb="V4=" and stra="1 V5=25.1 V3=0" */
1.318 brouard 12843: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.332 brouard 12844: /* If a blank, then strc="V4=" and strd='\0' */
12845: if(strc[0]=='\0'){
12846: printf("Error in resultline, probably a blank after the \"%s\", \"result:%s\", stra=\"%s\" resultsav=\"%s\"\n",strb,resultline, stra, resultsav);
12847: fprintf(ficlog,"Error in resultline, probably a blank after the \"V%s=\", resultline=%s\n",strb,resultline);
12848: return 1;
12849: }
1.234 brouard 12850: }else
12851: cutl(strc,strd,resultsav,'=');
1.318 brouard 12852: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 12853:
1.230 brouard 12854: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 12855: Tvarsel[k]=atoi(strc); /* 4 */ /* Tvarsel is the id of the kth covariate in the result line Tvarsel[1] in "V4=1.." is 4.*/
1.230 brouard 12856: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
12857: /* cptcovsel++; */
12858: if (nbocc(stra,'=') >0)
12859: strcpy(resultsav,stra); /* and analyzes it */
12860: }
1.235 brouard 12861: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 12862: /* Feeds resultmodel[nres][k1]=k2 for k1th product covariate with age in the model equation fed by the index k2 of the resutline*/
1.334 brouard 12863: for(k1=1; k1<= cptcovt ;k1++){ /* Loop on MODEL LINE V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.332 brouard 12864: if(Typevar[k1]==0){ /* Single covariate in model */
12865: /* 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 12866: match=0;
1.318 brouard 12867: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
12868: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 12869: modelresult[nres][k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 12870: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 12871: break;
12872: }
12873: }
12874: if(match == 0){
1.338 brouard 12875: printf("Error in result line (Dummy single): V%d is missing in result: %s according to model=1+age+%s. Tvar[k1=%d]=%d is different from Tvarsel[k2=%d]=%d.\n",Tvar[k1], resultline, model,k1, Tvar[k1], k2, Tvarsel[k2]);
12876: fprintf(ficlog,"Error in result line (Dummy single): V%d is missing in result: %s according to model=1+age+%s\n",Tvar[k1], resultline, model);
1.310 brouard 12877: return 1;
1.234 brouard 12878: }
1.332 brouard 12879: }else if(Typevar[k1]==1){ /* Product with age We want to get the position k2 in the resultline of the product k1 in the model line*/
12880: /* We feed resultmodel[k1]=k2; */
12881: match=0;
12882: for(k2=1; k2 <=j;k2++){/* Loop on resultline. jth occurence of = signs in the result line. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
12883: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 12884: modelresult[nres][k2]=k1;/* we found a Vn=1 corrresponding to Vn*age in the model modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.332 brouard 12885: resultmodel[nres][k1]=k2; /* Added here */
1.342 brouard 12886: /* printf("Decoderesult first modelresult[k2=%d]=%d (k1) V%d*AGE\n",k2,k1,Tvar[k1]); */
1.332 brouard 12887: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
12888: break;
12889: }
12890: }
12891: if(match == 0){
1.338 brouard 12892: printf("Error in result line (Product with age): V%d is missing in result: %s according to model=1+age+%s (Tvarsel[k2=%d]=%d)\n",Tvar[k1], resultline, model, k2, Tvarsel[k2]);
12893: fprintf(ficlog,"Error in result line (Product with age): V%d is missing in result: %s according to model=1+age+%s (Tvarsel[k2=%d]=%d)\n",Tvar[k1], resultline, model, k2, Tvarsel[k2]);
1.332 brouard 12894: return 1;
12895: }
1.349 brouard 12896: }else if(Typevar[k1]==2 || Typevar[k1]==3){ /* Product with or without age. We want to get the position in the resultline of the product in the model line*/
1.332 brouard 12897: /* resultmodel[nres][of such a Vn * Vm product k1] is not unique, so can't exist, we feed Tvard[k1][1] and [2] */
12898: match=0;
1.342 brouard 12899: /* printf("Decoderesult very first Product Tvardk[k1=%d][1]=%d Tvardk[k1=%d][2]=%d V%d * V%d\n",k1,Tvardk[k1][1],k1,Tvardk[k1][2],Tvardk[k1][1],Tvardk[k1][2]); */
1.332 brouard 12900: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
12901: if(Tvardk[k1][1]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
12902: /* modelresult[k2]=k1; */
1.342 brouard 12903: /* printf("Decoderesult first Product modelresult[k2=%d]=%d (k1) V%d * \n",k2,k1,Tvarsel[k2]); */
1.332 brouard 12904: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
12905: }
12906: }
12907: if(match == 0){
1.349 brouard 12908: printf("Error in result line (Product without age first variable or double product with age): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][1], resultline, model);
12909: fprintf(ficlog,"Error in result line (Product without age first variable or double product with age): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][1], resultline, model);
1.332 brouard 12910: return 1;
12911: }
12912: match=0;
12913: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
12914: if(Tvardk[k1][2]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
12915: /* modelresult[k2]=k1;*/
1.342 brouard 12916: /* printf("Decoderesult second Product modelresult[k2=%d]=%d (k1) * V%d \n ",k2,k1,Tvarsel[k2]); */
1.332 brouard 12917: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
12918: break;
12919: }
12920: }
12921: if(match == 0){
1.349 brouard 12922: printf("Error in result line (Product without age second variable or double product with age): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][2], resultline, model);
12923: fprintf(ficlog,"Error in result line (Product without age second variable or double product with age): V%d is missing in result : %s according to model=1+age+%s\n",Tvardk[k1][2], resultline, model);
1.332 brouard 12924: return 1;
12925: }
12926: }/* End of testing */
1.333 brouard 12927: }/* End loop cptcovt */
1.235 brouard 12928: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 12929: /* Feeds resultmodel[nres][k1]=k2 for single covariate (k1) in the model equation */
1.334 brouard 12930: for(k2=1; k2 <=j;k2++){ /* j or cptcovs is the number of single covariates used either in the model line as well as in the result line (dummy or quantitative)
12931: * Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 12932: match=0;
1.318 brouard 12933: for(k1=1; k1<= cptcovt ;k1++){ /* loop on model: model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.332 brouard 12934: if(Typevar[k1]==0){ /* Single only */
1.349 brouard 12935: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 What if a product? */
1.330 brouard 12936: resultmodel[nres][k1]=k2; /* k1th position in the model equation corresponds to k2th position in the result line. resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.334 brouard 12937: modelresult[nres][k2]=k1; /* k1th position in the model equation corresponds to k2th position in the result line. modelresult[1]=2 modelresult[2]=1 modelresult[3]=3 remodelresult[4]=6 modelresult[5]=9 */
1.234 brouard 12938: ++match;
12939: }
12940: }
12941: }
12942: if(match == 0){
1.338 brouard 12943: printf("Error in result line: variable V%d is missing in model; result: %s, model=1+age+%s\n",Tvarsel[k2], resultline, model);
12944: fprintf(ficlog,"Error in result line: variable V%d is missing in model; result: %s, model=1+age+%s\n",Tvarsel[k2], resultline, model);
1.310 brouard 12945: return 1;
1.234 brouard 12946: }else if(match > 1){
1.338 brouard 12947: printf("Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
12948: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
1.310 brouard 12949: return 1;
1.234 brouard 12950: }
12951: }
1.334 brouard 12952: /* cptcovres=j /\* Number of variables in the resultline is equal to cptcovs and thus useless *\/ */
1.234 brouard 12953: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 12954: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 brouard 12955: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
12956: /* should correspond to the combination 6 of dummy: V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 1*1 + 0*2 + 1*4 = 5 + (1offset) = 6*/
12957: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 12958: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
12959: /* 1 0 0 0 */
12960: /* 2 1 0 0 */
12961: /* 3 0 1 0 */
1.330 brouard 12962: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 12963: /* 5 0 0 1 */
1.330 brouard 12964: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 12965: /* 7 0 1 1 */
12966: /* 8 1 1 1 */
1.237 brouard 12967: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
12968: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
12969: /* V5*age V5 known which value for nres? */
12970: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.334 brouard 12971: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* cptcovt number of covariates (excluding 1 and age or age*age) in the MODEL equation.
12972: * loop on position k1 in the MODEL LINE */
1.331 brouard 12973: /* k counting number of combination of single dummies in the equation model */
12974: /* k4 counting single dummies in the equation model */
12975: /* k4q counting single quantitatives in the equation model */
1.344 brouard 12976: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Dummy and Single, fixed or timevarying, k1 is sorting according to MODEL, but k3 to resultline */
1.334 brouard 12977: /* k4+1= (not always if quant in model) position in the resultline V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) */
1.332 brouard 12978: /* modelresult[k3]=k1: k3th position in the result line corresponds to the k1 position in the model line (doesn't work with products)*/
1.330 brouard 12979: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
1.332 brouard 12980: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
12981: /* k3 is the position in the nres result line of the k1th variable of the model equation */
12982: /* Tvarsel[k3]: Name of the variable at the k3th position in the result line. */
12983: /* Tvalsel[k3]: Value of the variable at the k3th position in the result line. */
12984: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.334 brouard 12985: /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 12986: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1.330 brouard 12987: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.332 brouard 12988: k3= resultmodel[nres][k1]; /* From position k1 in model get position k3 in result line */
12989: /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
12990: k2=(int)Tvarsel[k3]; /* from position k3 in resultline get name k2: nres=1 k1=2=>k3=1 Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 (V4); k1=3=>k3=2 Tvarsel[2]=3 (V3)*/
1.330 brouard 12991: k+=Tvalsel[k3]*pow(2,k4); /* nres=1 k1=2 Tvalsel[1]=1 (V4=1); k1=3 k3=2 Tvalsel[2]=0 (V3=0) */
1.334 brouard 12992: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][Name]=Value; stores the value into the name of the variable. */
1.332 brouard 12993: /* Tinvresult[nres][4]=1 */
1.334 brouard 12994: /* Tresult[nres][k4+1]=Tvalsel[k3];/\* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) *\/ */
12995: Tresult[nres][k3]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
12996: /* Tvresult[nres][k4+1]=(int)Tvarsel[k3];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
12997: Tvresult[nres][k3]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
1.237 brouard 12998: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.334 brouard 12999: precov[nres][k1]=Tvalsel[k3]; /* Value from resultline of the variable at the k1 position in the model */
1.342 brouard 13000: /* printf("Decoderesult Dummy k=%d, k1=%d precov[nres=%d][k1=%d]=%.f V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k1, nres, k1,precov[nres][k1], k2, k3, (int)Tvalsel[k3], k4); */
1.235 brouard 13001: k4++;;
1.331 brouard 13002: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Quantitative and single */
1.330 brouard 13003: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.332 brouard 13004: /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.330 brouard 13005: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
1.332 brouard 13006: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 5 =k3q */
13007: k2q=(int)Tvarsel[k3q]; /* Name of variable at k3q th position in the resultline */
13008: /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 13009: /* Tqresult[nres][k4q+1]=Tvalsel[k3q]; /\* Tqresult[nres][1]=25.1 *\/ */
13010: /* Tvresult[nres][k4q+1]=(int)Tvarsel[k3q];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
13011: /* Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /\* Tvqresult[nres][1]=5 *\/ */
13012: Tqresult[nres][k3q]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
13013: Tvresult[nres][k3q]=(int)Tvarsel[k3q];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
13014: Tvqresult[nres][k3q]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
1.237 brouard 13015: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 brouard 13016: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 13017: precov[nres][k1]=Tvalsel[k3q];
1.342 brouard 13018: /* printf("Decoderesult Quantitative nres=%d,precov[nres=%d][k1=%d]=%.f V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, nres, k1,precov[nres][k1], k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]); */
1.235 brouard 13019: k4q++;;
1.350 brouard 13020: }else if( Dummy[k1]==2 ){ /* For dummy with age product "V2+V3+V4+V6+V7+V6*V2+V7*V2+V6*V3+V7*V3+V6*V4+V7*V4+age*V2+age*V3+age*V4+age*V6+age*V7+age*V6*V2+age*V6*V3+age*V7*V3+age*V6*V4+age*V7*V4\r"*/
13021: /* Tvar[k1]; */ /* Age variable */ /* 17 age*V6*V2 ?*/
1.332 brouard 13022: /* Wrong we want the value of variable name Tvar[k1] */
1.350 brouard 13023: if(Typevar[k1]==2 || Typevar[k1]==3 ){ /* For product quant or dummy (with or without age) */
13024: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
13025: /* printf("Decoderesult Quantitative or Dummy (not with age) nres=%d k1=%d precov[nres=%d][k1=%d]=%.f V%d(=%.f) * V%d(=%.f) \n",nres, k1, nres, k1,precov[nres][k1], Tvardk[k1][1], TinvDoQresult[nres][Tvardk[k1][1]], Tvardk[k1][2], TinvDoQresult[nres][Tvardk[k1][2]]); */
13026: }else{
13027: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
13028: k2=(int)Tvarsel[k3]; /* nres=1 k1=2=>k3=1 Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 (V4); k1=3=>k3=2 Tvarsel[2]=3 (V3)*/
13029: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][4]=1 */
13030: precov[nres][k1]=Tvalsel[k3];
13031: }
1.342 brouard 13032: /* printf("Decoderesult Dummy with age k=%d, k1=%d precov[nres=%d][k1=%d]=%.f Tvar[%d]=V%d k2=Tvarsel[%d]=%d Tvalsel[%d]=%d\n",k, k1, nres, k1,precov[nres][k1], k1, Tvar[k1], k3,(int)Tvarsel[k3], k3, (int)Tvalsel[k3]); */
1.331 brouard 13033: }else if( Dummy[k1]==3 ){ /* For quant with age product */
1.350 brouard 13034: if(Typevar[k1]==2 || Typevar[k1]==3 ){ /* For product quant or dummy (with or without age) */
13035: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
13036: /* printf("Decoderesult Quantitative or Dummy (not with age) nres=%d k1=%d precov[nres=%d][k1=%d]=%.f V%d(=%.f) * V%d(=%.f) \n",nres, k1, nres, k1,precov[nres][k1], Tvardk[k1][1], TinvDoQresult[nres][Tvardk[k1][1]], Tvardk[k1][2], TinvDoQresult[nres][Tvardk[k1][2]]); */
13037: }else{
13038: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
13039: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
13040: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* TinvDoQresult[nres][5]=25.1 */
13041: precov[nres][k1]=Tvalsel[k3q];
13042: }
1.342 brouard 13043: /* printf("Decoderesult Quantitative with age nres=%d, k1=%d, precov[nres=%d][k1=%d]=%f Tvar[%d]=V%d V(k2q=%d)= Tvarsel[%d]=%d, Tvalsel[%d]=%f\n",nres, k1, nres, k1,precov[nres][k1], k1, Tvar[k1], k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]); */
1.349 brouard 13044: }else if(Typevar[k1]==2 || Typevar[k1]==3 ){ /* For product quant or dummy (with or without age) */
1.332 brouard 13045: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
1.342 brouard 13046: /* printf("Decoderesult Quantitative or Dummy (not with age) nres=%d k1=%d precov[nres=%d][k1=%d]=%.f V%d(=%.f) * V%d(=%.f) \n",nres, k1, nres, k1,precov[nres][k1], Tvardk[k1][1], TinvDoQresult[nres][Tvardk[k1][1]], Tvardk[k1][2], TinvDoQresult[nres][Tvardk[k1][2]]); */
1.330 brouard 13047: }else{
1.332 brouard 13048: printf("Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
13049: fprintf(ficlog,"Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 13050: }
13051: }
1.234 brouard 13052:
1.334 brouard 13053: TKresult[nres]=++k; /* Number of combinations of dummies for the nresult and the model =Tvalsel[k3]*pow(2,k4) + 1*/
1.230 brouard 13054: return (0);
13055: }
1.235 brouard 13056:
1.230 brouard 13057: int decodemodel( char model[], int lastobs)
13058: /**< This routine decodes the model and returns:
1.224 brouard 13059: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
13060: * - nagesqr = 1 if age*age in the model, otherwise 0.
13061: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
13062: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
13063: * - cptcovage number of covariates with age*products =2
13064: * - cptcovs number of simple covariates
1.339 brouard 13065: * ncovcolt=ncovcol+nqv+ntv+nqtv total of covariates in the data, not in the model equation
1.224 brouard 13066: * - Tvar[k] is the id of the kth covariate Tvar[1]@12 {1, 2, 3, 8, 10, 11, 8, 3, 7, 8, 5, 6}, thus Tvar[5=V7*V8]=10
1.339 brouard 13067: * which is a new column after the 9 (ncovcol+nqv+ntv+nqtv) variables.
1.319 brouard 13068: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 13069: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
13070: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
13071: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
13072: */
1.319 brouard 13073: /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 */
1.136 brouard 13074: {
1.359 brouard 13075: int i, j, k, ks;/* , v;*/
1.349 brouard 13076: int n,m;
13077: int j1, k1, k11, k12, k2, k3, k4;
13078: char modelsav[300];
13079: char stra[300], strb[300], strc[300], strd[300],stre[300],strf[300];
1.187 brouard 13080: char *strpt;
1.349 brouard 13081: int **existcomb;
13082:
13083: existcomb=imatrix(1,NCOVMAX,1,NCOVMAX);
13084: for(i=1;i<=NCOVMAX;i++)
13085: for(j=1;j<=NCOVMAX;j++)
13086: existcomb[i][j]=0;
13087:
1.145 brouard 13088: /*removespace(model);*/
1.136 brouard 13089: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.349 brouard 13090: j=0, j1=0, k1=0, k12=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 13091: if (strstr(model,"AGE") !=0){
1.192 brouard 13092: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
13093: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 13094: return 1;
13095: }
1.141 brouard 13096: if (strstr(model,"v") !=0){
1.338 brouard 13097: printf("Error. 'v' must be in upper case 'V' model=1+age+%s ",model);
13098: fprintf(ficlog,"Error. 'v' must be in upper case model=1+age+%s ",model);fflush(ficlog);
1.141 brouard 13099: return 1;
13100: }
1.187 brouard 13101: strcpy(modelsav,model);
13102: if ((strpt=strstr(model,"age*age")) !=0){
1.338 brouard 13103: printf(" strpt=%s, model=1+age+%s\n",strpt, model);
1.187 brouard 13104: if(strpt != model){
1.338 brouard 13105: printf("Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 13106: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 13107: corresponding column of parameters.\n",model);
1.338 brouard 13108: fprintf(ficlog,"Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 13109: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 13110: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 13111: return 1;
1.225 brouard 13112: }
1.187 brouard 13113: nagesqr=1;
13114: if (strstr(model,"+age*age") !=0)
1.234 brouard 13115: substrchaine(modelsav, model, "+age*age");
1.187 brouard 13116: else if (strstr(model,"age*age+") !=0)
1.234 brouard 13117: substrchaine(modelsav, model, "age*age+");
1.187 brouard 13118: else
1.234 brouard 13119: substrchaine(modelsav, model, "age*age");
1.187 brouard 13120: }else
13121: nagesqr=0;
1.349 brouard 13122: if (strlen(modelsav) >1){ /* V2 +V3 +V4 +V6 +V7 +V6*V2 +V7*V2 +V6*V3 +V7*V3 +V6*V4 +V7*V4 +age*V2 +age*V3 +age*V4 +age*V6 +age*V7 +age*V6*V2 +V7*V2 +age*V6*V3 +age*V7*V3 +age*V6*V4 +age*V7*V4 */
1.187 brouard 13123: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
13124: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.351 brouard 13125: cptcovs=0; /**< Number of simple covariates V1 +V1*age +V3 +V3*V4 +age*age => V1 + V3 =4+1-3=2 Wrong */
1.187 brouard 13126: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 13127: * cst, age and age*age
13128: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
13129: /* including age products which are counted in cptcovage.
13130: * but the covariates which are products must be treated
13131: * separately: ncovn=4- 2=2 (V1+V3). */
1.349 brouard 13132: cptcovprod=0; /**< Number of products V1*V2 +v3*age = 2 */
13133: cptcovdageprod=0; /* Number of doouble products with age age*Vn*VM or Vn*age*Vm or Vn*Vm*age */
1.187 brouard 13134: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.349 brouard 13135: cptcovprodage=0;
13136: /* cptcovprodage=nboccstr(modelsav,"age");*/
1.225 brouard 13137:
1.187 brouard 13138: /* Design
13139: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
13140: * < ncovcol=8 >
13141: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
13142: * k= 1 2 3 4 5 6 7 8
13143: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
1.345 brouard 13144: * covar[k,i], are for fixed covariates, value of kth covariate if not including age for individual i:
1.224 brouard 13145: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
13146: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 13147: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
13148: * Tage[++cptcovage]=k
1.345 brouard 13149: * if products, new covar are created after ncovcol + nqv (quanti fixed) with k1
1.187 brouard 13150: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
13151: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
13152: * Tvard[k1][1]=m Tvard[k1][2]=m; Tvard[1][1]=5 (V5) Tvard[1][2]=6 Tvard[2][1]=7 (V7) Tvard[2][2]=8
13153: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
13154: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
13155: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
1.345 brouard 13156: * < ncovcol=8 8 fixed covariate. Additional starts at 9 (V5*V6) and 10(V7*V8) >
1.187 brouard 13157: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
13158: * k= 1 2 3 4 5 6 7 8 9 10 11 12
1.345 brouard 13159: * Tvard[k]= 2 1 3 3 10 11 8 8 5 6 7 8
13160: * p Tvar[1]@12={2, 1, 3, 3, 9, 10, 8, 8}
1.187 brouard 13161: * p Tprod[1]@2={ 6, 5}
13162: *p Tvard[1][1]@4= {7, 8, 5, 6}
13163: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
13164: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 13165: *How to reorganize? Tvars(orted)
1.187 brouard 13166: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
13167: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
13168: * {2, 1, 4, 8, 5, 6, 3, 7}
13169: * Struct []
13170: */
1.225 brouard 13171:
1.187 brouard 13172: /* This loop fills the array Tvar from the string 'model'.*/
13173: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
13174: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
13175: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
13176: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
13177: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
13178: /* k=1 Tvar[1]=2 (from V2) */
13179: /* k=5 Tvar[5] */
13180: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 13181: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 13182: /* } */
1.198 brouard 13183: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 13184: /*
13185: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 13186: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
13187: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
13188: }
1.187 brouard 13189: cptcovage=0;
1.351 brouard 13190:
13191: /* First loop in order to calculate */
13192: /* for age*VN*Vm
13193: * Provides, Typevar[k], Tage[cptcovage], existcomb[n][m], FixedV[ncovcolt+k12]
13194: * Tprod[k1]=k Tposprod[k]=k1; Tvard[k1][1] =m;
13195: */
13196: /* Needs FixedV[Tvardk[k][1]] */
13197: /* For others:
13198: * Sets Typevar[k];
13199: * Tvar[k]=ncovcol+nqv+ntv+nqtv+k11;
13200: * Tposprod[k]=k11;
13201: * Tprod[k11]=k;
13202: * Tvardk[k][1] =m;
13203: * Needs FixedV[Tvardk[k][1]] == 0
13204: */
13205:
1.319 brouard 13206: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
13207: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
13208: modelsav==V2+V1+V5*age+V4+V3*age strb=V3*age stra=V2+V1V5*age+V4 */ /* <model> "V5+V4+V3+V4*V3+V5*age+V1*age+V1" strb="V5" stra="V4+V3+V4*V3+V5*age+V1*age+V1" */
13209: if (nbocc(modelsav,'+')==0)
13210: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 13211: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
13212: /*scanf("%d",i);*/
1.349 brouard 13213: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age OR double product with age strb=age*V6*V2 or V6*V2*age or V6*age*V2 */
13214: cutl(strc,strd,strb,'*'); /**< k=1 strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 OR strb=age*V6*V2 strc=V6*V2 strd=age OR c=V2*age OR c=age*V2 */
13215: if(strchr(strc,'*')) { /**< Model with age and DOUBLE product: allowed since 0.99r44, strc=V6*V2 or V2*age or age*V2, strd=age or V6 or V6 */
13216: Typevar[k]=3; /* 3 for age and double product age*Vn*Vm varying of fixed */
13217: if(strstr(strc,"age")!=0) { /* It means that strc=V2*age or age*V2 and thus that strd=Vn */
13218: cutl(stre,strf,strc,'*') ; /* strf=age or Vm, stre=Vm or age. If strc=V6*V2 then strf=V6 and stre=V2 */
13219: strcpy(strc,strb); /* save strb(=age*Vn*Vm) into strc */
13220: /* We want strb=Vn*Vm */
13221: if(strcmp(strf,"age")==0){ /* strf is "age" so that stre=Vm =V2 . */
13222: strcpy(strb,strd);
13223: strcat(strb,"*");
13224: strcat(strb,stre);
13225: }else{ /* strf=Vm If strf=V6 then stre=V2 */
13226: strcpy(strb,strf);
13227: strcat(strb,"*");
13228: strcat(strb,stre);
13229: strcpy(strd,strb); /* in order for strd to not be "age" for next test (will be Vn*Vm */
13230: }
1.351 brouard 13231: /* printf("DEBUG FIXED k=%d, Tage[k]=%d, Tvar[Tage[k]=%d,FixedV[Tvar[Tage[k]]]=%d\n",k,Tage[k],Tvar[Tage[k]],FixedV[Tvar[Tage[k]]]); */
13232: /* FixedV[Tvar[Tage[k]]]=0; /\* HERY not sure if V7*V4*age Fixed might not exist yet*\/ */
1.349 brouard 13233: }else{ /* strc=Vn*Vm (and strd=age) and should be strb=Vn*Vm but want to keep original strb double product */
13234: strcpy(stre,strb); /* save full b in stre */
13235: strcpy(strb,strc); /* save short c in new short b for next block strb=Vn*Vm*/
13236: strcpy(strf,strc); /* save short c in new short f */
13237: cutl(strc,strd,strf,'*'); /* We get strd=Vn and strc=Vm for next block (strb=Vn*Vm)*/
13238: /* strcpy(strc,stre);*/ /* save full e in c for future */
13239: }
13240: cptcovdageprod++; /* double product with age Which product is it? */
13241: /* strcpy(strb,strc); /\* strb was age*V6*V2 or V6*V2*age or V6*age*V2 IS now V6*V2 or V2*age or age*V2 *\/ */
13242: /* cutl(strc,strd,strb,'*'); /\* strd= V6 or V2 or age and strc= V2 or age or V2 *\/ */
1.234 brouard 13243: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
1.349 brouard 13244: n=atoi(stre);
1.234 brouard 13245: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
1.349 brouard 13246: m=atoi(strc);
13247: cptcovage++; /* Counts the number of covariates which include age as a product */
13248: Tage[cptcovage]=k; /* For age*V3*V2 gives the position in model of covariates associated with age Tage[1]=6 HERY too*/
13249: if(existcomb[n][m] == 0){
13250: /* r /home/brouard/Documents/Recherches/REVES/Zachary/Zach-2022/Feinuo_Sun/Feinuo-threeway/femV12V15_3wayintNBe.imach */
13251: printf("Warning in model combination V%d*V%d should exist in the model before adding V%d*V%d*age !\n",n,m,n,m);
13252: fprintf(ficlog,"Warning in model combination V%d*V%d should exist in the model before adding V%d*V%d*age !\n",n,m,n,m);
13253: fflush(ficlog);
13254: k1++; /* The combination Vn*Vm will be in the model so we create it at k1 */
13255: k12++;
13256: existcomb[n][m]=k1;
13257: existcomb[m][n]=k1;
13258: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1;
13259: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2+ age*V6*V3 Gives the k position of the k1 double product Vn*Vm or age*Vn*Vm*/
13260: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 Gives the k1 double product Vn*Vm or age*Vn*Vm at the k position */
13261: Tvard[k1][1] =m; /* m 1 for V1*/
13262: Tvardk[k][1] =m; /* m 1 for V1*/
13263: Tvard[k1][2] =n; /* n 4 for V4*/
13264: Tvardk[k][2] =n; /* n 4 for V4*/
1.351 brouard 13265: /* Tvar[Tage[cptcovage]]=k1;*/ /* Tvar[6=age*V3*V2]=9 (new fixed covariate) */ /* We don't know about Fixed yet HERE */
1.349 brouard 13266: if( FixedV[Tvardk[k][1]] == 0 && FixedV[Tvardk[k][2]] == 0){ /* If the product is a fixed covariate then we feed the new column with Vn*Vm */
13267: for (i=1; i<=lastobs;i++){/* For fixed product */
13268: /* Computes the new covariate which is a product of
13269: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
13270: covar[ncovcolt+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
13271: }
13272: cptcovprodage++; /* Counting the number of fixed covariate with age */
13273: FixedV[ncovcolt+k12]=0; /* We expand Vn*Vm */
13274: k12++;
13275: FixedV[ncovcolt+k12]=0;
13276: }else{ /*End of FixedV */
13277: cptcovprodvage++; /* Counting the number of varying covariate with age */
13278: FixedV[ncovcolt+k12]=1; /* We expand Vn*Vm */
13279: k12++;
13280: FixedV[ncovcolt+k12]=1;
13281: }
13282: }else{ /* k1 Vn*Vm already exists */
13283: k11=existcomb[n][m];
13284: Tposprod[k]=k11; /* OK */
13285: Tvar[k]=Tvar[Tprod[k11]]; /* HERY */
13286: Tvardk[k][1]=m;
13287: Tvardk[k][2]=n;
13288: if( FixedV[Tvardk[k][1]] == 0 && FixedV[Tvardk[k][2]] == 0){ /* If the product is a fixed covariate then we feed the new column with Vn*Vm */
13289: /*cptcovage++;*/ /* Counts the number of covariates which include age as a product */
13290: cptcovprodage++; /* Counting the number of fixed covariate with age */
13291: /*Tage[cptcovage]=k;*/ /* For age*V3*V2 Tage[1]=V3*V3=9 HERY too*/
13292: Tvar[Tage[cptcovage]]=k1;
13293: FixedV[ncovcolt+k12]=0; /* We expand Vn*Vm */
13294: k12++;
13295: FixedV[ncovcolt+k12]=0;
13296: }else{ /* Already exists but time varying (and age) */
13297: /*cptcovage++;*/ /* Counts the number of covariates which include age as a product */
13298: /*Tage[cptcovage]=k;*/ /* For age*V3*V2 Tage[1]=V3*V3=9 HERY too*/
13299: /* Tvar[Tage[cptcovage]]=k1; */
13300: cptcovprodvage++;
13301: FixedV[ncovcolt+k12]=1; /* We expand Vn*Vm */
13302: k12++;
13303: FixedV[ncovcolt+k12]=1;
13304: }
13305: }
13306: /* Tage[cptcovage]=k; /\* V2+V1+V4+V3*age Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
13307: /* Tvar[k]=k11; /\* HERY *\/ */
13308: } else {/* simple product strb=age*Vn so that c=Vn and d=age, or strb=Vn*age so that c=age and d=Vn, or b=Vn*Vm so that c=Vm and d=Vn */
13309: cptcovprod++;
13310: if (strcmp(strc,"age")==0) { /**< Model includes age: strb= Vn*age c=age d=Vn*/
13311: /* covar is not filled and then is empty */
13312: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
13313: Tvar[k]=atoi(stre); /* V2+V1+V5*age+V4+V3*age Tvar[5]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
13314: Typevar[k]=1; /* 1 for age product */
13315: cptcovage++; /* Counts the number of covariates which include age as a product */
13316: Tage[cptcovage]=k; /* V2+V1+V4+V3*age Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
13317: if( FixedV[Tvar[k]] == 0){
13318: cptcovprodage++; /* Counting the number of fixed covariate with age */
13319: }else{
13320: cptcovprodvage++; /* Counting the number of fixedvarying covariate with age */
13321: }
13322: /*printf("stre=%s ", stre);*/
13323: } else if (strcmp(strd,"age")==0) { /* strb= age*Vn c=Vn */
13324: cutl(stre,strb,strc,'V');
13325: Tvar[k]=atoi(stre);
13326: Typevar[k]=1; /* 1 for age product */
13327: cptcovage++;
13328: Tage[cptcovage]=k;
13329: if( FixedV[Tvar[k]] == 0){
13330: cptcovprodage++; /* Counting the number of fixed covariate with age */
13331: }else{
13332: cptcovprodvage++; /* Counting the number of fixedvarying covariate with age */
1.339 brouard 13333: }
1.349 brouard 13334: }else{ /* for product Vn*Vm */
13335: Typevar[k]=2; /* 2 for product Vn*Vm */
13336: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
13337: n=atoi(stre);
13338: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
13339: m=atoi(strc);
13340: k1++;
13341: cptcovprodnoage++;
13342: if(existcomb[n][m] != 0 || existcomb[m][n] != 0){
13343: printf("Warning in model combination V%d*V%d already exists in the model in position k1=%d!\n",n,m,existcomb[n][m]);
13344: fprintf(ficlog,"Warning in model combination V%d*V%d already exists in the model in position k1=%d!\n",n,m,existcomb[n][m]);
13345: fflush(ficlog);
13346: k11=existcomb[n][m];
13347: Tvar[k]=ncovcol+nqv+ntv+nqtv+k11;
13348: Tposprod[k]=k11;
13349: Tprod[k11]=k;
13350: Tvardk[k][1] =m; /* m 1 for V1*/
13351: /* Tvard[k11][1] =m; /\* n 4 for V4*\/ */
13352: Tvardk[k][2] =n; /* n 4 for V4*/
13353: /* Tvard[k11][2] =n; /\* n 4 for V4*\/ */
13354: }else{ /* combination Vn*Vm doesn't exist we create it (no age)*/
13355: existcomb[n][m]=k1;
13356: existcomb[m][n]=k1;
13357: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* ncovcolt+k1; For model-covariate k tells which data-covariate to use but
13358: because this model-covariate is a construction we invent a new column
13359: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
13360: If already ncovcol=4 and model= V2 + V1 + V1*V4 + age*V3 + V3*V2
13361: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
13362: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=3 etc */
13363: /* Please remark that the new variables are model dependent */
13364: /* If we have 4 variable but the model uses only 3, like in
13365: * model= V1 + age*V1 + V2 + V3 + age*V2 + age*V3 + V1*V2 + V1*V3
13366: * k= 1 2 3 4 5 6 7 8
13367: * Tvar[k]=1 1 2 3 2 3 (5 6) (and not 4 5 because of V4 missing)
13368: * Tage[kk] [1]= 2 [2]=5 [3]=6 kk=1 to cptcovage=3
13369: * Tvar[Tage[kk]][1]=2 [2]=2 [3]=3
13370: */
13371: /* We need to feed some variables like TvarVV, but later on next loop because of ncovv (k2) is not correct */
13372: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 +V6*V2*age */
13373: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
13374: Tvard[k1][1] =m; /* m 1 for V1*/
13375: Tvardk[k][1] =m; /* m 1 for V1*/
13376: Tvard[k1][2] =n; /* n 4 for V4*/
13377: Tvardk[k][2] =n; /* n 4 for V4*/
13378: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
13379: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
13380: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
13381: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
13382: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
13383: if( FixedV[Tvardk[k][1]] == 0 && FixedV[Tvardk[k][2]] == 0){ /* If the product is a fixed covariate then we feed the new column with Vn*Vm */
13384: for (i=1; i<=lastobs;i++){/* For fixed product */
13385: /* Computes the new covariate which is a product of
13386: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
13387: covar[ncovcolt+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
13388: }
13389: /* TvarVV[k2]=n; */
13390: /* FixedV[ncovcolt+k2]=0; /\* or FixedV[Tvar[k]]=0; FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13391: /* TvarVV[k2+1]=m; */
13392: /* FixedV[ncovcolt+k2]=0; /\* or FixedV[Tvar[k]]=0; FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13393: }else{ /* not FixedV */
13394: /* TvarVV[k2]=n; */
13395: /* FixedV[ncovcolt+k2]=0; /\* or FixedV[Tvar[k]]=0; FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13396: /* TvarVV[k2+1]=m; */
13397: /* FixedV[ncovcolt+k2]=0; /\* or FixedV[Tvar[k]]=0; FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13398: }
13399: } /* End of creation of Vn*Vm if not created by age*Vn*Vm earlier */
13400: } /* End of product Vn*Vm */
13401: } /* End of age*double product or simple product */
13402: }else { /* not a product */
1.234 brouard 13403: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
13404: /* scanf("%d",i);*/
13405: cutl(strd,strc,strb,'V');
13406: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
13407: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
13408: Tvar[k]=atoi(strd);
13409: Typevar[k]=0; /* 0 for simple covariates */
13410: }
13411: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 13412: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 13413: scanf("%d",i);*/
1.187 brouard 13414: } /* end of loop + on total covariates */
1.351 brouard 13415:
13416:
1.187 brouard 13417: } /* end if strlen(modelsave == 0) age*age might exist */
13418: } /* end if strlen(model == 0) */
1.349 brouard 13419: cptcovs=cptcovt - cptcovdageprod - cptcovprod;/**< Number of simple covariates V1 +V1*age +V3 +V3*V4 +age*age + age*v4*V3=> V1 + V3 =4+1-3=2 */
13420:
1.136 brouard 13421: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
13422: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 13423:
1.136 brouard 13424: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 13425: printf("cptcovprod=%d ", cptcovprod);
13426: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
13427: scanf("%d ",i);*/
13428:
13429:
1.230 brouard 13430: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
13431: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 13432: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
13433: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
13434: k = 1 2 3 4 5 6 7 8 9
13435: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 13436: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 13437: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
13438: Dummy[k] 1 0 0 0 3 1 1 2 3
13439: Tmodelind[combination of covar]=k;
1.225 brouard 13440: */
13441: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 13442: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 13443: /* Tvar[k] is the value n of Vn with n varying for 1 to nvcol, or p Vp=Vn*Vm for product */
1.226 brouard 13444: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 13445: printf("Model=1+age+%s\n\
1.349 brouard 13446: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product, 3 for double product with age \n\
1.227 brouard 13447: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
13448: Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product\n",model);
1.318 brouard 13449: fprintf(ficlog,"Model=1+age+%s\n\
1.349 brouard 13450: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product, 3 for double product with age \n\
1.227 brouard 13451: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
13452: Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product\n",model);
1.342 brouard 13453: for(k=-1;k<=NCOVMAX; k++){ Fixed[k]=0; Dummy[k]=0;}
13454: for(k=1;k<=NCOVMAX; k++){TvarFind[k]=0; TvarVind[k]=0;}
1.351 brouard 13455:
13456:
13457: /* Second loop for calculating Fixed[k], Dummy[k]*/
13458:
13459:
1.349 brouard 13460: for(k=1, ncovf=0, nsd=0, nsq=0, ncovv=0,ncovva=0,ncovvta=0, ncova=0, ncoveff=0, nqfveff=0, ntveff=0, nqtveff=0, ncovvt=0;k<=cptcovt; k++){ /* or cptocvt loop on k from model */
1.234 brouard 13461: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 13462: Fixed[k]= 0;
13463: Dummy[k]= 0;
1.225 brouard 13464: ncoveff++;
1.232 brouard 13465: ncovf++;
1.234 brouard 13466: nsd++;
13467: modell[k].maintype= FTYPE;
13468: TvarsD[nsd]=Tvar[k];
13469: TvarsDind[nsd]=k;
1.330 brouard 13470: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 13471: TvarF[ncovf]=Tvar[k];
13472: TvarFind[ncovf]=k;
13473: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
13474: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.339 brouard 13475: /* }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /\* Product of fixed dummy (<=ncovcol) covariates For a fixed product k is higher than ncovcol *\/ */
1.240 brouard 13476: }else if( Tvar[k] <=ncovcol+nqv && Typevar[k]==0){/* Remind that product Vn*Vm are added in k Only simple fixed quantitative variable */
1.227 brouard 13477: Fixed[k]= 0;
13478: Dummy[k]= 1;
1.230 brouard 13479: nqfveff++;
1.234 brouard 13480: modell[k].maintype= FTYPE;
13481: modell[k].subtype= FQ;
13482: nsq++;
1.334 brouard 13483: TvarsQ[nsq]=Tvar[k]; /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary see below */
13484: TvarsQind[nsq]=k; /* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.232 brouard 13485: ncovf++;
1.234 brouard 13486: TvarF[ncovf]=Tvar[k];
13487: TvarFind[ncovf]=k;
1.231 brouard 13488: TvarFQ[nqfveff]=Tvar[k]-ncovcol; /* TvarFQ[1]=V2-1=1st in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1.230 brouard 13489: TvarFQind[nqfveff]=k; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1.242 brouard 13490: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.339 brouard 13491: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
13492: /* model V1+V3+age*V1+age*V3+V1*V3 */
13493: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
13494: ncovvt++;
13495: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
13496: TvarVVind[ncovvt]=k; /* TvarVVind[1]=2 (second position in the model equation */
13497:
1.227 brouard 13498: Fixed[k]= 1;
13499: Dummy[k]= 0;
1.225 brouard 13500: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 13501: modell[k].maintype= VTYPE;
13502: modell[k].subtype= VD;
13503: nsd++;
13504: TvarsD[nsd]=Tvar[k];
13505: TvarsDind[nsd]=k;
1.330 brouard 13506: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 13507: ncovv++; /* Only simple time varying variables */
13508: TvarV[ncovv]=Tvar[k];
1.242 brouard 13509: TvarVind[ncovv]=k; /* TvarVind[2]=2 TvarVind[3]=3 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Any time varying singele */
1.231 brouard 13510: TvarVD[ntveff]=Tvar[k]; /* TvarVD[1]=V4 TvarVD[2]=V3 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying dummy variable */
13511: TvarVDind[ntveff]=k; /* TvarVDind[1]=2 TvarVDind[2]=3 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying dummy variable */
1.228 brouard 13512: printf("Quasi Tmodelind[%d]=%d,Tvar[Tmodelind[%d]]=V%d, ncovcol=%d, nqv=%d,Tvar[k]- ncovcol-nqv=%d\n",ntveff,k,ntveff,Tvar[k], ncovcol, nqv,Tvar[k]- ncovcol-nqv);
13513: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 13514: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.339 brouard 13515: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
13516: /* model V1+V3+age*V1+age*V3+V1*V3 */
13517: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
13518: ncovvt++;
13519: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
13520: TvarVVind[ncovvt]=k; /* TvarVV[1]=V3 (first time varying in the model equation */
13521:
1.234 brouard 13522: Fixed[k]= 1;
13523: Dummy[k]= 1;
13524: nqtveff++;
13525: modell[k].maintype= VTYPE;
13526: modell[k].subtype= VQ;
13527: ncovv++; /* Only simple time varying variables */
13528: nsq++;
1.334 brouard 13529: TvarsQ[nsq]=Tvar[k]; /* k=1 Tvar=5 nsq=1 TvarsQ[1]=5 */ /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary here) */
13530: TvarsQind[nsq]=k; /* For single quantitative covariate gives the model position of each single quantitative covariate *//* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.234 brouard 13531: TvarV[ncovv]=Tvar[k];
1.242 brouard 13532: TvarVind[ncovv]=k; /* TvarVind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Any time varying singele */
1.231 brouard 13533: TvarVQ[nqtveff]=Tvar[k]; /* TvarVQ[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying quantitative variable */
13534: TvarVQind[nqtveff]=k; /* TvarVQind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying quantitative variable */
1.234 brouard 13535: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
13536: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
1.349 brouard 13537: /* printf("Quasi TmodelQind[%d]=%d,Tvar[TmodelQind[%d]]=V%d, ncovcol=%d, nqv=%d, ntv=%Ad,Tvar[k]- ncovcol-nqv-ntv=%d\n",nqtveff,k,nqtveff,Tvar[k], ncovcol, nqv, ntv, Tvar[k]- ncovcol-nqv-ntv); */
1.342 brouard 13538: /* printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv); */
1.227 brouard 13539: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 13540: ncova++;
13541: TvarA[ncova]=Tvar[k];
13542: TvarAind[ncova]=k;
1.349 brouard 13543: /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
13544: /** Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product */
1.231 brouard 13545: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 13546: Fixed[k]= 2;
13547: Dummy[k]= 2;
13548: modell[k].maintype= ATYPE;
13549: modell[k].subtype= APFD;
1.349 brouard 13550: ncovta++;
13551: TvarAVVA[ncovta]=Tvar[k]; /* (2)age*V3 */
13552: TvarAVVAind[ncovta]=k;
1.240 brouard 13553: /* ncoveff++; */
1.227 brouard 13554: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 13555: Fixed[k]= 2;
13556: Dummy[k]= 3;
13557: modell[k].maintype= ATYPE;
13558: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
1.349 brouard 13559: ncovta++;
13560: TvarAVVA[ncovta]=Tvar[k]; /* */
13561: TvarAVVAind[ncovta]=k;
1.240 brouard 13562: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 13563: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 13564: Fixed[k]= 3;
13565: Dummy[k]= 2;
13566: modell[k].maintype= ATYPE;
13567: modell[k].subtype= APVD; /* Product age * varying dummy */
1.349 brouard 13568: ncovva++;
13569: TvarVVA[ncovva]=Tvar[k]; /* (1)+age*V6 + (2)age*V7 */
13570: TvarVVAind[ncovva]=k;
13571: ncovta++;
13572: TvarAVVA[ncovta]=Tvar[k]; /* */
13573: TvarAVVAind[ncovta]=k;
1.240 brouard 13574: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 13575: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 13576: Fixed[k]= 3;
13577: Dummy[k]= 3;
13578: modell[k].maintype= ATYPE;
13579: modell[k].subtype= APVQ; /* Product age * varying quantitative */
1.349 brouard 13580: ncovva++;
13581: TvarVVA[ncovva]=Tvar[k]; /* */
13582: TvarVVAind[ncovva]=k;
13583: ncovta++;
13584: TvarAVVA[ncovta]=Tvar[k]; /* (1)+age*V6 + (2)age*V7 */
13585: TvarAVVAind[ncovta]=k;
1.240 brouard 13586: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 13587: }
1.349 brouard 13588: }else if( Tposprod[k]>0 && Typevar[k]==2){ /* Detects if fixed product no age Vm*Vn */
13589: printf("MEMORY ERRORR k=%d Tposprod[k]=%d, Typevar[k]=%d\n ",k, Tposprod[k], Typevar[k]);
13590: if(FixedV[Tvardk[k][1]] == 0 && FixedV[Tvardk[k][2]] == 0){ /* Needs a fixed product Product of fixed dummy (<=ncovcol) covariates For a fixed product k is higher than ncovcol V3*V2 */
13591: printf("MEMORY ERRORR k=%d Tvardk[k][1]=%d, Tvardk[k][2]=%d, FixedV[Tvardk[k][1]]=%d,FixedV[Tvardk[k][2]]=%d\n ",k,Tvardk[k][1],Tvardk[k][2],FixedV[Tvardk[k][1]],FixedV[Tvardk[k][2]]);
13592: Fixed[k]= 0;
13593: Dummy[k]= 0;
13594: ncoveff++;
13595: ncovf++;
13596: /* ncovv++; */
13597: /* TvarVV[ncovv]=Tvardk[k][1]; */
13598: /* FixedV[ncovcolt+ncovv]=0; /\* or FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13599: /* ncovv++; */
13600: /* TvarVV[ncovv]=Tvardk[k][2]; */
13601: /* FixedV[ncovcolt+ncovv]=0; /\* or FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13602: modell[k].maintype= FTYPE;
13603: TvarF[ncovf]=Tvar[k];
13604: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
13605: TvarFind[ncovf]=k;
13606: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
13607: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
13608: }else{/* product varying Vn * Vm without age, V1+V3+age*V1+age*V3+V1*V3 looking at V1*V3, Typevar={0, 0, 1, 1, 2}, k=5, V1 is fixed, V3 is timevary, V5 is a product */
13609: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
13610: /* model V1+V3+age*V1+age*V3+V1*V3 + V1*V3*age*/
13611: /* Tvar={1, 3, 1, 3, 6, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
13612: k1=Tposprod[k]; /* Position in the products of product k, Tposprod={0, 0, 0, 0, 1, 1} k1=1 first product but second time varying because of V3 */
13613: ncovvt++;
13614: TvarVV[ncovvt]=Tvard[k1][1]; /* TvarVV[2]=V1 (because TvarVV[1] was V3, first time varying covariates */
13615: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
13616: ncovvt++;
13617: TvarVV[ncovvt]=Tvard[k1][2]; /* TvarVV[3]=V3 */
13618: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
13619:
13620: /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
13621: /** Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product */
13622:
13623: if(Tvard[k1][1] <=ncovcol){ /* Vn is dummy fixed, (Tvard[1][1]=V1), (Tvard[1][1]=V3 time varying) */
13624: if(Tvard[k1][2] <=ncovcol){ /* Vm is dummy fixed */
13625: Fixed[k]= 1;
13626: Dummy[k]= 0;
13627: modell[k].maintype= FTYPE;
13628: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
13629: ncovf++; /* Fixed variables without age */
13630: TvarF[ncovf]=Tvar[k];
13631: TvarFind[ncovf]=k;
13632: }else if(Tvard[k1][2] <=ncovcol+nqv){ /* Vm is quanti fixed */
13633: Fixed[k]= 0; /* Fixed product */
13634: Dummy[k]= 1;
13635: modell[k].maintype= FTYPE;
13636: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
13637: ncovf++; /* Varying variables without age */
13638: TvarF[ncovf]=Tvar[k];
13639: TvarFind[ncovf]=k;
13640: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is a time varying dummy covariate */
13641: Fixed[k]= 1;
13642: Dummy[k]= 0;
13643: modell[k].maintype= VTYPE;
13644: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
13645: ncovv++; /* Varying variables without age */
13646: TvarV[ncovv]=Tvar[k]; /* TvarV[1]=Tvar[5]=5 because there is a V4 */
13647: TvarVind[ncovv]=k;/* TvarVind[1]=5 */
13648: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is a time varying quantitative covariate */
13649: Fixed[k]= 1;
13650: Dummy[k]= 1;
13651: modell[k].maintype= VTYPE;
13652: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
13653: ncovv++; /* Varying variables without age */
13654: TvarV[ncovv]=Tvar[k];
13655: TvarVind[ncovv]=k;
13656: }
13657: }else if(Tvard[k1][1] <=ncovcol+nqv){ /* Vn is fixed quanti */
13658: if(Tvard[k1][2] <=ncovcol){ /* Vm is fixed dummy */
13659: Fixed[k]= 0; /* Fixed product */
13660: Dummy[k]= 1;
13661: modell[k].maintype= FTYPE;
13662: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
13663: ncovf++; /* Fixed variables without age */
13664: TvarF[ncovf]=Tvar[k];
13665: TvarFind[ncovf]=k;
13666: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is time varying */
13667: Fixed[k]= 1;
13668: Dummy[k]= 1;
13669: modell[k].maintype= VTYPE;
13670: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
13671: ncovv++; /* Varying variables without age */
13672: TvarV[ncovv]=Tvar[k];
13673: TvarVind[ncovv]=k;
13674: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is time varying quanti */
13675: Fixed[k]= 1;
13676: Dummy[k]= 1;
13677: modell[k].maintype= VTYPE;
13678: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
13679: ncovv++; /* Varying variables without age */
13680: TvarV[ncovv]=Tvar[k];
13681: TvarVind[ncovv]=k;
13682: ncovv++; /* Varying variables without age */
13683: TvarV[ncovv]=Tvar[k];
13684: TvarVind[ncovv]=k;
13685: }
13686: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){ /* Vn is time varying dummy */
13687: if(Tvard[k1][2] <=ncovcol){
13688: Fixed[k]= 1;
13689: Dummy[k]= 1;
13690: modell[k].maintype= VTYPE;
13691: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
13692: ncovv++; /* Varying variables without age */
13693: TvarV[ncovv]=Tvar[k];
13694: TvarVind[ncovv]=k;
13695: }else if(Tvard[k1][2] <=ncovcol+nqv){
13696: Fixed[k]= 1;
13697: Dummy[k]= 1;
13698: modell[k].maintype= VTYPE;
13699: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
13700: ncovv++; /* Varying variables without age */
13701: TvarV[ncovv]=Tvar[k];
13702: TvarVind[ncovv]=k;
13703: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
13704: Fixed[k]= 1;
13705: Dummy[k]= 0;
13706: modell[k].maintype= VTYPE;
13707: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
13708: ncovv++; /* Varying variables without age */
13709: TvarV[ncovv]=Tvar[k];
13710: TvarVind[ncovv]=k;
13711: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
13712: Fixed[k]= 1;
13713: Dummy[k]= 1;
13714: modell[k].maintype= VTYPE;
13715: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
13716: ncovv++; /* Varying variables without age */
13717: TvarV[ncovv]=Tvar[k];
13718: TvarVind[ncovv]=k;
13719: }
13720: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){ /* Vn is time varying quanti */
13721: if(Tvard[k1][2] <=ncovcol){
13722: Fixed[k]= 1;
13723: Dummy[k]= 1;
13724: modell[k].maintype= VTYPE;
13725: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
13726: ncovv++; /* Varying variables without age */
13727: TvarV[ncovv]=Tvar[k];
13728: TvarVind[ncovv]=k;
13729: }else if(Tvard[k1][2] <=ncovcol+nqv){
13730: Fixed[k]= 1;
13731: Dummy[k]= 1;
13732: modell[k].maintype= VTYPE;
13733: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
13734: ncovv++; /* Varying variables without age */
13735: TvarV[ncovv]=Tvar[k];
13736: TvarVind[ncovv]=k;
13737: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
13738: Fixed[k]= 1;
13739: Dummy[k]= 1;
13740: modell[k].maintype= VTYPE;
13741: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
13742: ncovv++; /* Varying variables without age */
13743: TvarV[ncovv]=Tvar[k];
13744: TvarVind[ncovv]=k;
13745: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
13746: Fixed[k]= 1;
13747: Dummy[k]= 1;
13748: modell[k].maintype= VTYPE;
13749: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
13750: ncovv++; /* Varying variables without age */
13751: TvarV[ncovv]=Tvar[k];
13752: TvarVind[ncovv]=k;
13753: }
13754: }else{
13755: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
13756: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
13757: } /*end k1*/
13758: }
13759: }else if(Typevar[k] == 3){ /* product Vn * Vm with age, V1+V3+age*V1+age*V3+V1*V3 looking at V1*V3, Typevar={0, 0, 1, 1, 2}, k=5, V1 is fixed, V3 is timevary, V5 is a product */
1.339 brouard 13760: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
1.349 brouard 13761: /* model V1+V3+age*V1+age*V3+V1*V3 + V1*V3*age*/
13762: /* Tvar={1, 3, 1, 3, 6, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
13763: k1=Tposprod[k]; /* Position in the products of product k, Tposprod={0, 0, 0, 0, 1, 1} k1=1 first product but second time varying because of V3 */
13764: ncova++;
13765: TvarA[ncova]=Tvard[k1][1]; /* TvarVV[2]=V1 (because TvarVV[1] was V3, first time varying covariates */
13766: TvarAind[ncova]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
13767: ncova++;
13768: TvarA[ncova]=Tvard[k1][2]; /* TvarVV[3]=V3 */
13769: TvarAind[ncova]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
1.339 brouard 13770:
1.349 brouard 13771: /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
13772: /** Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product */
13773: if( FixedV[Tvardk[k][1]] == 0 && FixedV[Tvardk[k][2]] == 0){
13774: ncovta++;
13775: TvarAVVA[ncovta]=Tvard[k1][1]; /* age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 */
13776: TvarAVVAind[ncovta]=k;
13777: ncovta++;
13778: TvarAVVA[ncovta]=Tvard[k1][2]; /* age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 */
13779: TvarAVVAind[ncovta]=k;
13780: }else{
13781: ncovva++; /* HERY reached */
13782: TvarVVA[ncovva]=Tvard[k1][1]; /* age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 */
13783: TvarVVAind[ncovva]=k;
13784: ncovva++;
13785: TvarVVA[ncovva]=Tvard[k1][2]; /* */
13786: TvarVVAind[ncovva]=k;
13787: ncovta++;
13788: TvarAVVA[ncovta]=Tvard[k1][1]; /* age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 */
13789: TvarAVVAind[ncovta]=k;
13790: ncovta++;
13791: TvarAVVA[ncovta]=Tvard[k1][2]; /* age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 */
13792: TvarAVVAind[ncovta]=k;
13793: }
1.339 brouard 13794: if(Tvard[k1][1] <=ncovcol){ /* Vn is dummy fixed, (Tvard[1][1]=V1), (Tvard[1][1]=V3 time varying) */
13795: if(Tvard[k1][2] <=ncovcol){ /* Vm is dummy fixed */
1.349 brouard 13796: Fixed[k]= 2;
13797: Dummy[k]= 2;
1.240 brouard 13798: modell[k].maintype= FTYPE;
13799: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
1.349 brouard 13800: /* TvarF[ncova]=Tvar[k]; /\* Problem to solve *\/ */
13801: /* TvarFind[ncova]=k; */
1.339 brouard 13802: }else if(Tvard[k1][2] <=ncovcol+nqv){ /* Vm is quanti fixed */
1.349 brouard 13803: Fixed[k]= 2; /* Fixed product */
13804: Dummy[k]= 3;
1.240 brouard 13805: modell[k].maintype= FTYPE;
13806: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
1.349 brouard 13807: /* TvarF[ncova]=Tvar[k]; */
13808: /* TvarFind[ncova]=k; */
1.339 brouard 13809: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is a time varying dummy covariate */
1.349 brouard 13810: Fixed[k]= 3;
13811: Dummy[k]= 2;
1.240 brouard 13812: modell[k].maintype= VTYPE;
13813: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
1.349 brouard 13814: TvarV[ncova]=Tvar[k]; /* TvarV[1]=Tvar[5]=5 because there is a V4 */
13815: TvarVind[ncova]=k;/* TvarVind[1]=5 */
1.339 brouard 13816: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is a time varying quantitative covariate */
1.349 brouard 13817: Fixed[k]= 3;
13818: Dummy[k]= 3;
1.240 brouard 13819: modell[k].maintype= VTYPE;
13820: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
1.349 brouard 13821: /* ncovv++; /\* Varying variables without age *\/ */
13822: /* TvarV[ncovv]=Tvar[k]; */
13823: /* TvarVind[ncovv]=k; */
1.240 brouard 13824: }
1.339 brouard 13825: }else if(Tvard[k1][1] <=ncovcol+nqv){ /* Vn is fixed quanti */
13826: if(Tvard[k1][2] <=ncovcol){ /* Vm is fixed dummy */
1.349 brouard 13827: Fixed[k]= 2; /* Fixed product */
13828: Dummy[k]= 2;
1.240 brouard 13829: modell[k].maintype= FTYPE;
13830: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
1.349 brouard 13831: /* ncova++; /\* Fixed variables with age *\/ */
13832: /* TvarF[ncovf]=Tvar[k]; */
13833: /* TvarFind[ncovf]=k; */
1.339 brouard 13834: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is time varying */
1.349 brouard 13835: Fixed[k]= 2;
13836: Dummy[k]= 3;
1.240 brouard 13837: modell[k].maintype= VTYPE;
13838: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
1.349 brouard 13839: /* ncova++; /\* Varying variables with age *\/ */
13840: /* TvarV[ncova]=Tvar[k]; */
13841: /* TvarVind[ncova]=k; */
1.339 brouard 13842: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is time varying quanti */
1.349 brouard 13843: Fixed[k]= 3;
13844: Dummy[k]= 2;
1.240 brouard 13845: modell[k].maintype= VTYPE;
13846: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
1.349 brouard 13847: ncova++; /* Varying variables without age */
13848: TvarV[ncova]=Tvar[k];
13849: TvarVind[ncova]=k;
13850: /* ncova++; /\* Varying variables without age *\/ */
13851: /* TvarV[ncova]=Tvar[k]; */
13852: /* TvarVind[ncova]=k; */
1.240 brouard 13853: }
1.339 brouard 13854: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){ /* Vn is time varying dummy */
1.240 brouard 13855: if(Tvard[k1][2] <=ncovcol){
1.349 brouard 13856: Fixed[k]= 2;
13857: Dummy[k]= 2;
1.240 brouard 13858: modell[k].maintype= VTYPE;
13859: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
1.349 brouard 13860: /* ncova++; /\* Varying variables with age *\/ */
13861: /* TvarV[ncova]=Tvar[k]; */
13862: /* TvarVind[ncova]=k; */
1.240 brouard 13863: }else if(Tvard[k1][2] <=ncovcol+nqv){
1.349 brouard 13864: Fixed[k]= 2;
13865: Dummy[k]= 3;
1.240 brouard 13866: modell[k].maintype= VTYPE;
13867: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
1.349 brouard 13868: /* ncova++; /\* Varying variables with age *\/ */
13869: /* TvarV[ncova]=Tvar[k]; */
13870: /* TvarVind[ncova]=k; */
1.240 brouard 13871: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
1.349 brouard 13872: Fixed[k]= 3;
13873: Dummy[k]= 2;
1.240 brouard 13874: modell[k].maintype= VTYPE;
13875: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
1.349 brouard 13876: /* ncova++; /\* Varying variables with age *\/ */
13877: /* TvarV[ncova]=Tvar[k]; */
13878: /* TvarVind[ncova]=k; */
1.240 brouard 13879: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
1.349 brouard 13880: Fixed[k]= 3;
13881: Dummy[k]= 3;
1.240 brouard 13882: modell[k].maintype= VTYPE;
13883: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
1.349 brouard 13884: /* ncova++; /\* Varying variables with age *\/ */
13885: /* TvarV[ncova]=Tvar[k]; */
13886: /* TvarVind[ncova]=k; */
1.240 brouard 13887: }
1.339 brouard 13888: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){ /* Vn is time varying quanti */
1.240 brouard 13889: if(Tvard[k1][2] <=ncovcol){
1.349 brouard 13890: Fixed[k]= 2;
13891: Dummy[k]= 2;
1.240 brouard 13892: modell[k].maintype= VTYPE;
13893: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
1.349 brouard 13894: /* ncova++; /\* Varying variables with age *\/ */
13895: /* TvarV[ncova]=Tvar[k]; */
13896: /* TvarVind[ncova]=k; */
1.240 brouard 13897: }else if(Tvard[k1][2] <=ncovcol+nqv){
1.349 brouard 13898: Fixed[k]= 2;
13899: Dummy[k]= 3;
1.240 brouard 13900: modell[k].maintype= VTYPE;
13901: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
1.349 brouard 13902: /* ncova++; /\* Varying variables with age *\/ */
13903: /* TvarV[ncova]=Tvar[k]; */
13904: /* TvarVind[ncova]=k; */
1.240 brouard 13905: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
1.349 brouard 13906: Fixed[k]= 3;
13907: Dummy[k]= 2;
1.240 brouard 13908: modell[k].maintype= VTYPE;
13909: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
1.349 brouard 13910: /* ncova++; /\* Varying variables with age *\/ */
13911: /* TvarV[ncova]=Tvar[k]; */
13912: /* TvarVind[ncova]=k; */
1.240 brouard 13913: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
1.349 brouard 13914: Fixed[k]= 3;
13915: Dummy[k]= 3;
1.240 brouard 13916: modell[k].maintype= VTYPE;
13917: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
1.349 brouard 13918: /* ncova++; /\* Varying variables with age *\/ */
13919: /* TvarV[ncova]=Tvar[k]; */
13920: /* TvarVind[ncova]=k; */
1.240 brouard 13921: }
1.227 brouard 13922: }else{
1.240 brouard 13923: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
13924: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
13925: } /*end k1*/
1.349 brouard 13926: } else{
1.226 brouard 13927: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
13928: fprintf(ficlog,"Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
1.225 brouard 13929: }
1.342 brouard 13930: /* printf("Decodemodel, k=%d, Tvar[%d]=V%d,Typevar=%d, Fixed=%d, Dummy=%d\n",k, k,Tvar[k],Typevar[k],Fixed[k],Dummy[k]); */
13931: /* printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype); */
1.227 brouard 13932: fprintf(ficlog,"Decodemodel, k=%d, Tvar[%d]=V%d,Typevar=%d, Fixed=%d, Dummy=%d\n",k, k,Tvar[k],Typevar[k],Fixed[k],Dummy[k]);
13933: }
1.349 brouard 13934: ncovvta=ncovva;
1.227 brouard 13935: /* Searching for doublons in the model */
13936: for(k1=1; k1<= cptcovt;k1++){
13937: for(k2=1; k2 <k1;k2++){
1.285 brouard 13938: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
13939: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 13940: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
13941: if(Tvar[k1]==Tvar[k2]){
1.338 brouard 13942: printf("Error duplication in the model=1+age+%s at positions (+) %d and %d, Tvar[%d]=V%d, Tvar[%d]=V%d, Typevar=%d, Fixed=%d, Dummy=%d\n", model, k1,k2, k1, Tvar[k1], k2, Tvar[k2],Typevar[k1],Fixed[k1],Dummy[k1]);
13943: fprintf(ficlog,"Error duplication in the model=1+age+%s at positions (+) %d and %d, Tvar[%d]=V%d, Tvar[%d]=V%d, Typevar=%d, Fixed=%d, Dummy=%d\n", model, k1,k2, k1, Tvar[k1], k2, Tvar[k2],Typevar[k1],Fixed[k1],Dummy[k1]); fflush(ficlog);
1.234 brouard 13944: return(1);
13945: }
13946: }else if (Typevar[k1] ==2){
13947: k3=Tposprod[k1];
13948: k4=Tposprod[k2];
13949: if( ((Tvard[k3][1]== Tvard[k4][1])&&(Tvard[k3][2]== Tvard[k4][2])) || ((Tvard[k3][1]== Tvard[k4][2])&&(Tvard[k3][2]== Tvard[k4][1])) ){
1.338 brouard 13950: printf("Error duplication in the model=1+age+%s at positions (+) %d and %d, V%d*V%d, Typevar=%d, Fixed=%d, Dummy=%d\n",model, k1,k2, Tvard[k3][1], Tvard[k3][2],Typevar[k1],Fixed[Tvar[k1]],Dummy[Tvar[k1]]);
13951: fprintf(ficlog,"Error duplication in the model=1+age+%s at positions (+) %d and %d, V%d*V%d, Typevar=%d, Fixed=%d, Dummy=%d\n",model, k1,k2, Tvard[k3][1], Tvard[k3][2],Typevar[k1],Fixed[Tvar[k1]],Dummy[Tvar[k1]]); fflush(ficlog);
1.234 brouard 13952: return(1);
13953: }
13954: }
1.227 brouard 13955: }
13956: }
1.225 brouard 13957: }
13958: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
13959: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 13960: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
13961: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.349 brouard 13962:
13963: free_imatrix(existcomb,1,NCOVMAX,1,NCOVMAX);
1.137 brouard 13964: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 13965: /*endread:*/
1.225 brouard 13966: printf("Exiting decodemodel: ");
13967: return (1);
1.136 brouard 13968: }
13969:
1.169 brouard 13970: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 13971: {/* Check ages at death */
1.136 brouard 13972: int i, m;
1.218 brouard 13973: int firstone=0;
13974:
1.136 brouard 13975: for (i=1; i<=imx; i++) {
13976: for(m=2; (m<= maxwav); m++) {
13977: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
13978: anint[m][i]=9999;
1.216 brouard 13979: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
13980: s[m][i]=-1;
1.136 brouard 13981: }
13982: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 13983: *nberr = *nberr + 1;
1.218 brouard 13984: if(firstone == 0){
13985: firstone=1;
1.260 brouard 13986: printf("Warning (#%d)! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown but status is a death state %d at wave %d. If you don't know the vital status, please enter -2. If he/she is still alive but don't know the state, please code with '-1 or '.'. Here, we do not believe in a death, skipped.\nOther similar cases in log file\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
1.218 brouard 13987: }
1.262 brouard 13988: fprintf(ficlog,"Warning (#%d)! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown but status is a death state %d at wave %d. If you don't know the vital status, please enter -2. If he/she is still alive but don't know the state, please code with '-1 or '.'. Here, we do not believe in a death, skipped.\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
1.260 brouard 13989: s[m][i]=-1; /* Droping the death status */
1.136 brouard 13990: }
13991: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 13992: (*nberr)++;
1.259 brouard 13993: printf("Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\nOther similar cases in log file\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
1.262 brouard 13994: fprintf(ficlog,"Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
1.259 brouard 13995: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 13996: }
13997: }
13998: }
13999:
14000: for (i=1; i<=imx; i++) {
14001: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
14002: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 14003: if(s[m][i] >0 || s[m][i]==-1 || s[m][i]==-2 || s[m][i]==-4 || s[m][i]==-5){ /* What if s[m][i]=-1 */
1.136 brouard 14004: if (s[m][i] >= nlstate+1) {
1.169 brouard 14005: if(agedc[i]>0){
14006: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 14007: agev[m][i]=agedc[i];
1.214 brouard 14008: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 14009: }else {
1.136 brouard 14010: if ((int)andc[i]!=9999){
14011: nbwarn++;
14012: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
14013: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
14014: agev[m][i]=-1;
14015: }
14016: }
1.169 brouard 14017: } /* agedc > 0 */
1.214 brouard 14018: } /* end if */
1.136 brouard 14019: else if(s[m][i] !=9){ /* Standard case, age in fractional
14020: years but with the precision of a month */
14021: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
14022: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
14023: agev[m][i]=1;
14024: else if(agev[m][i] < *agemin){
14025: *agemin=agev[m][i];
14026: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
14027: }
14028: else if(agev[m][i] >*agemax){
14029: *agemax=agev[m][i];
1.156 brouard 14030: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 14031: }
14032: /*agev[m][i]=anint[m][i]-annais[i];*/
14033: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 14034: } /* en if 9*/
1.136 brouard 14035: else { /* =9 */
1.214 brouard 14036: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 14037: agev[m][i]=1;
14038: s[m][i]=-1;
14039: }
14040: }
1.214 brouard 14041: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 14042: agev[m][i]=1;
1.214 brouard 14043: else{
14044: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
14045: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
14046: agev[m][i]=0;
14047: }
14048: } /* End for lastpass */
14049: }
1.136 brouard 14050:
14051: for (i=1; i<=imx; i++) {
14052: for(m=firstpass; (m<=lastpass); m++){
14053: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 14054: (*nberr)++;
1.136 brouard 14055: printf("Error: on wave %d of individual %d status %d > (nlstate+ndeath)=(%d+%d)=%d\n",m,i,s[m][i],nlstate, ndeath, nlstate+ndeath);
14056: fprintf(ficlog,"Error: on wave %d of individual %d status %d > (nlstate+ndeath)=(%d+%d)=%d\n",m,i,s[m][i],nlstate, ndeath, nlstate+ndeath);
14057: return 1;
14058: }
14059: }
14060: }
14061:
14062: /*for (i=1; i<=imx; i++){
14063: for (m=firstpass; (m<lastpass); m++){
14064: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
14065: }
14066:
14067: }*/
14068:
14069:
1.139 brouard 14070: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
14071: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 14072:
14073: return (0);
1.164 brouard 14074: /* endread:*/
1.136 brouard 14075: printf("Exiting calandcheckages: ");
14076: return (1);
14077: }
14078:
1.172 brouard 14079: #if defined(_MSC_VER)
14080: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
14081: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
14082: //#include "stdafx.h"
14083: //#include <stdio.h>
14084: //#include <tchar.h>
14085: //#include <windows.h>
14086: //#include <iostream>
14087: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
14088:
14089: LPFN_ISWOW64PROCESS fnIsWow64Process;
14090:
14091: BOOL IsWow64()
14092: {
14093: BOOL bIsWow64 = FALSE;
14094:
14095: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
14096: // (HANDLE, PBOOL);
14097:
14098: //LPFN_ISWOW64PROCESS fnIsWow64Process;
14099:
14100: HMODULE module = GetModuleHandle(_T("kernel32"));
14101: const char funcName[] = "IsWow64Process";
14102: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
14103: GetProcAddress(module, funcName);
14104:
14105: if (NULL != fnIsWow64Process)
14106: {
14107: if (!fnIsWow64Process(GetCurrentProcess(),
14108: &bIsWow64))
14109: //throw std::exception("Unknown error");
14110: printf("Unknown error\n");
14111: }
14112: return bIsWow64 != FALSE;
14113: }
14114: #endif
1.177 brouard 14115:
1.191 brouard 14116: void syscompilerinfo(int logged)
1.292 brouard 14117: {
14118: #include <stdint.h>
14119:
14120: /* #include "syscompilerinfo.h"*/
1.185 brouard 14121: /* command line Intel compiler 32bit windows, XP compatible:*/
14122: /* /GS /W3 /Gy
14123: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
14124: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
14125: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 14126: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
14127: */
14128: /* 64 bits */
1.185 brouard 14129: /*
14130: /GS /W3 /Gy
14131: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
14132: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
14133: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
14134: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
14135: /* Optimization are useless and O3 is slower than O2 */
14136: /*
14137: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
14138: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
14139: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
14140: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
14141: */
1.186 brouard 14142: /* Link is */ /* /OUT:"visual studio
1.185 brouard 14143: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
14144: /PDB:"visual studio
14145: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
14146: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
14147: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
14148: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
14149: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
14150: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
14151: uiAccess='false'"
14152: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
14153: /NOLOGO /TLBID:1
14154: */
1.292 brouard 14155:
14156:
1.177 brouard 14157: #if defined __INTEL_COMPILER
1.178 brouard 14158: #if defined(__GNUC__)
14159: struct utsname sysInfo; /* For Intel on Linux and OS/X */
14160: #endif
1.177 brouard 14161: #elif defined(__GNUC__)
1.179 brouard 14162: #ifndef __APPLE__
1.174 brouard 14163: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 14164: #endif
1.177 brouard 14165: struct utsname sysInfo;
1.178 brouard 14166: int cross = CROSS;
14167: if (cross){
14168: printf("Cross-");
1.191 brouard 14169: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 14170: }
1.174 brouard 14171: #endif
14172:
1.191 brouard 14173: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 14174: #if defined(__clang__)
1.191 brouard 14175: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 14176: #endif
14177: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 14178: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 14179: #endif
14180: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 14181: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 14182: #endif
14183: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 14184: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 14185: #endif
14186: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 14187: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 14188: #endif
14189: #if defined(_MSC_VER)
1.191 brouard 14190: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 14191: #endif
14192: #if defined(__PGI)
1.191 brouard 14193: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 14194: #endif
14195: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 14196: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 14197: #endif
1.191 brouard 14198: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 14199:
1.167 brouard 14200: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
14201: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
14202: // Windows (x64 and x86)
1.191 brouard 14203: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 14204: #elif __unix__ // all unices, not all compilers
14205: // Unix
1.191 brouard 14206: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 14207: #elif __linux__
14208: // linux
1.191 brouard 14209: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 14210: #elif __APPLE__
1.174 brouard 14211: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 14212: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 14213: #endif
14214:
14215: /* __MINGW32__ */
14216: /* __CYGWIN__ */
14217: /* __MINGW64__ */
14218: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
14219: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
14220: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
14221: /* _WIN64 // Defined for applications for Win64. */
14222: /* _M_X64 // Defined for compilations that target x64 processors. */
14223: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 14224:
1.167 brouard 14225: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 14226: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 14227: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 14228: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 14229: #else
1.191 brouard 14230: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 14231: #endif
14232:
1.169 brouard 14233: #if defined(__GNUC__)
14234: # if defined(__GNUC_PATCHLEVEL__)
14235: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
14236: + __GNUC_MINOR__ * 100 \
14237: + __GNUC_PATCHLEVEL__)
14238: # else
14239: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
14240: + __GNUC_MINOR__ * 100)
14241: # endif
1.174 brouard 14242: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 14243: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 14244:
14245: if (uname(&sysInfo) != -1) {
14246: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 14247: if(logged) fprintf(ficlog,"Running on: %s %s %s %s %s\n ",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.176 brouard 14248: }
14249: else
14250: perror("uname() error");
1.179 brouard 14251: //#ifndef __INTEL_COMPILER
14252: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 14253: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 14254: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 14255: #endif
1.169 brouard 14256: #endif
1.172 brouard 14257:
1.286 brouard 14258: // void main ()
1.172 brouard 14259: // {
1.169 brouard 14260: #if defined(_MSC_VER)
1.174 brouard 14261: if (IsWow64()){
1.191 brouard 14262: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
14263: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 14264: }
14265: else{
1.191 brouard 14266: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
14267: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 14268: }
1.172 brouard 14269: // printf("\nPress Enter to continue...");
14270: // getchar();
14271: // }
14272:
1.169 brouard 14273: #endif
14274:
1.167 brouard 14275:
1.219 brouard 14276: }
1.136 brouard 14277:
1.219 brouard 14278: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 14279: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.332 brouard 14280: /* Computes the prevalence limit for each combination of the dummy covariates */
1.235 brouard 14281: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 14282: /* double ftolpl = 1.e-10; */
1.180 brouard 14283: double age, agebase, agelim;
1.203 brouard 14284: double tot;
1.180 brouard 14285:
1.202 brouard 14286: strcpy(filerespl,"PL_");
14287: strcat(filerespl,fileresu);
14288: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 14289: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
14290: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 14291: }
1.288 brouard 14292: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
14293: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 14294: pstamp(ficrespl);
1.288 brouard 14295: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 14296: fprintf(ficrespl,"#Age ");
14297: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
14298: fprintf(ficrespl,"\n");
1.180 brouard 14299:
1.219 brouard 14300: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 14301:
1.219 brouard 14302: agebase=ageminpar;
14303: agelim=agemaxpar;
1.180 brouard 14304:
1.227 brouard 14305: /* i1=pow(2,ncoveff); */
1.234 brouard 14306: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 14307: if (cptcovn < 1){i1=1;}
1.180 brouard 14308:
1.337 brouard 14309: /* for(k=1; k<=i1;k++){ /\* For each combination k of dummy covariates in the model *\/ */
1.238 brouard 14310: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 14311: k=TKresult[nres];
1.338 brouard 14312: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 14313: /* if(i1 != 1 && TKresult[nres]!= k) /\* We found the combination k corresponding to the resultline value of dummies *\/ */
14314: /* continue; */
1.235 brouard 14315:
1.238 brouard 14316: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
14317: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
14318: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
14319: /* k=k+1; */
14320: /* to clean */
1.332 brouard 14321: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 14322: fprintf(ficrespl,"#******");
14323: printf("#******");
14324: fprintf(ficlog,"#******");
1.337 brouard 14325: for(j=1;j<=cptcovs ;j++) {/**< cptcovs number of SIMPLE covariates in the model or resultline V2+V1 =2 (dummy or quantit or time varying) */
1.332 brouard 14326: /* fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Here problem for varying dummy*\/ */
1.337 brouard 14327: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14328: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14329: fprintf(ficrespl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14330: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14331: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14332: }
14333: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
14334: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
14335: /* fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
14336: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
14337: /* } */
1.238 brouard 14338: fprintf(ficrespl,"******\n");
14339: printf("******\n");
14340: fprintf(ficlog,"******\n");
14341: if(invalidvarcomb[k]){
14342: printf("\nCombination (%d) ignored because no case \n",k);
14343: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
14344: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
14345: continue;
14346: }
1.219 brouard 14347:
1.238 brouard 14348: fprintf(ficrespl,"#Age ");
1.337 brouard 14349: /* for(j=1;j<=cptcoveff;j++) { */
14350: /* fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14351: /* } */
14352: for(j=1;j<=cptcovs;j++) { /* New the quanti variable is added */
14353: fprintf(ficrespl,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 14354: }
14355: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
14356: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 14357:
1.238 brouard 14358: for (age=agebase; age<=agelim; age++){
14359: /* for (age=agebase; age<=agebase; age++){ */
1.337 brouard 14360: /**< Computes the prevalence limit in each live state at age x and for covariate combination (k and) nres */
14361: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres); /* Nicely done */
1.238 brouard 14362: fprintf(ficrespl,"%.0f ",age );
1.337 brouard 14363: /* for(j=1;j<=cptcoveff;j++) */
14364: /* fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14365: for(j=1;j<=cptcovs;j++)
14366: fprintf(ficrespl,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 14367: tot=0.;
14368: for(i=1; i<=nlstate;i++){
14369: tot += prlim[i][i];
14370: fprintf(ficrespl," %.5f", prlim[i][i]);
14371: }
14372: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
14373: } /* Age */
14374: /* was end of cptcod */
1.337 brouard 14375: } /* nres */
14376: /* } /\* for each combination *\/ */
1.219 brouard 14377: return 0;
1.180 brouard 14378: }
14379:
1.218 brouard 14380: int back_prevalence_limit(double *p, double **bprlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp, double dateprev1,double dateprev2, int firstpass, int lastpass, int mobilavproj){
1.288 brouard 14381: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 14382:
14383: /* Computes the back prevalence limit for any combination of covariate values
14384: * at any age between ageminpar and agemaxpar
14385: */
1.235 brouard 14386: int i, j, k, i1, nres=0 ;
1.217 brouard 14387: /* double ftolpl = 1.e-10; */
14388: double age, agebase, agelim;
14389: double tot;
1.218 brouard 14390: /* double ***mobaverage; */
14391: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 14392:
14393: strcpy(fileresplb,"PLB_");
14394: strcat(fileresplb,fileresu);
14395: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 14396: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
14397: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 14398: }
1.288 brouard 14399: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
14400: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 14401: pstamp(ficresplb);
1.288 brouard 14402: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 14403: fprintf(ficresplb,"#Age ");
14404: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
14405: fprintf(ficresplb,"\n");
14406:
1.218 brouard 14407:
14408: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
14409:
14410: agebase=ageminpar;
14411: agelim=agemaxpar;
14412:
14413:
1.227 brouard 14414: i1=pow(2,cptcoveff);
1.218 brouard 14415: if (cptcovn < 1){i1=1;}
1.227 brouard 14416:
1.238 brouard 14417: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.338 brouard 14418: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
14419: k=TKresult[nres];
14420: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
14421: /* if(i1 != 1 && TKresult[nres]!= k) */
14422: /* continue; */
14423: /* /\*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*\/ */
1.238 brouard 14424: fprintf(ficresplb,"#******");
14425: printf("#******");
14426: fprintf(ficlog,"#******");
1.338 brouard 14427: for(j=1;j<=cptcovs ;j++) {/**< cptcovs number of SIMPLE covariates in the model or resultline V2+V1 =2 (dummy or quantit or time varying) */
14428: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14429: fprintf(ficresplb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14430: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 14431: }
1.338 brouard 14432: /* for(j=1;j<=cptcoveff ;j++) {/\* all covariates *\/ */
14433: /* fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14434: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14435: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14436: /* } */
14437: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
14438: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14439: /* fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14440: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14441: /* } */
1.238 brouard 14442: fprintf(ficresplb,"******\n");
14443: printf("******\n");
14444: fprintf(ficlog,"******\n");
14445: if(invalidvarcomb[k]){
14446: printf("\nCombination (%d) ignored because no cases \n",k);
14447: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
14448: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
14449: continue;
14450: }
1.218 brouard 14451:
1.238 brouard 14452: fprintf(ficresplb,"#Age ");
1.338 brouard 14453: for(j=1;j<=cptcovs;j++) {
14454: fprintf(ficresplb,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 14455: }
14456: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
14457: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 14458:
14459:
1.238 brouard 14460: for (age=agebase; age<=agelim; age++){
14461: /* for (age=agebase; age<=agebase; age++){ */
14462: if(mobilavproj > 0){
14463: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
14464: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 14465: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 14466: }else if (mobilavproj == 0){
14467: printf("There is no chance to get back prevalence limit if data aren't non zero and summing to 1, please try a non null mobil_average(=%d) parameter or mobil_average=-1 if you want to try at your own risk.\n",mobilavproj);
14468: fprintf(ficlog,"There is no chance to get back prevalence limit if data aren't non zero and summing to 1, please try a non null mobil_average(=%d) parameter or mobil_average=-1 if you want to try at your own risk.\n",mobilavproj);
14469: exit(1);
14470: }else{
14471: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 14472: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 14473: /* printf("TOTOT\n"); */
14474: /* exit(1); */
1.238 brouard 14475: }
14476: fprintf(ficresplb,"%.0f ",age );
1.338 brouard 14477: for(j=1;j<=cptcovs;j++)
14478: fprintf(ficresplb,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 14479: tot=0.;
14480: for(i=1; i<=nlstate;i++){
14481: tot += bprlim[i][i];
14482: fprintf(ficresplb," %.5f", bprlim[i][i]);
14483: }
14484: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
14485: } /* Age */
14486: /* was end of cptcod */
1.255 brouard 14487: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.338 brouard 14488: /* } /\* end of any combination *\/ */
1.238 brouard 14489: } /* end of nres */
1.218 brouard 14490: /* hBijx(p, bage, fage); */
14491: /* fclose(ficrespijb); */
14492:
14493: return 0;
1.217 brouard 14494: }
1.218 brouard 14495:
1.180 brouard 14496: int hPijx(double *p, int bage, int fage){
14497: /*------------- h Pij x at various ages ------------*/
1.336 brouard 14498: /* to be optimized with precov */
1.180 brouard 14499: int stepsize;
14500: int agelim;
14501: int hstepm;
14502: int nhstepm;
1.359 brouard 14503: int h, i, i1, j, k, nres=0;
1.180 brouard 14504:
14505: double agedeb;
14506: double ***p3mat;
14507:
1.337 brouard 14508: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
14509: if((ficrespij=fopen(filerespij,"w"))==NULL) {
14510: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
14511: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
14512: }
14513: printf("Computing pij: result on file '%s' \n", filerespij);
14514: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
14515:
14516: stepsize=(int) (stepm+YEARM-1)/YEARM;
14517: /*if (stepm<=24) stepsize=2;*/
14518:
14519: agelim=AGESUP;
14520: hstepm=stepsize*YEARM; /* Every year of age */
14521: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
14522:
14523: /* hstepm=1; aff par mois*/
14524: pstamp(ficrespij);
14525: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
14526: i1= pow(2,cptcoveff);
14527: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
14528: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
14529: /* k=k+1; */
14530: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
14531: k=TKresult[nres];
1.338 brouard 14532: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 14533: /* for(k=1; k<=i1;k++){ */
14534: /* if(i1 != 1 && TKresult[nres]!= k) */
14535: /* continue; */
14536: fprintf(ficrespij,"\n#****** ");
14537: for(j=1;j<=cptcovs;j++){
14538: fprintf(ficrespij," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14539: /* fprintf(ficrespij,"@wV%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14540: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
14541: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
14542: /* fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
14543: }
14544: fprintf(ficrespij,"******\n");
14545:
14546: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
14547: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
14548: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
14549:
14550: /* nhstepm=nhstepm*YEARM; aff par mois*/
14551:
14552: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
14553: oldm=oldms;savm=savms;
14554: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
14555: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
14556: for(i=1; i<=nlstate;i++)
14557: for(j=1; j<=nlstate+ndeath;j++)
14558: fprintf(ficrespij," %1d-%1d",i,j);
14559: fprintf(ficrespij,"\n");
14560: for (h=0; h<=nhstepm; h++){
14561: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
14562: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.183 brouard 14563: for(i=1; i<=nlstate;i++)
14564: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 14565: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.183 brouard 14566: fprintf(ficrespij,"\n");
14567: }
1.337 brouard 14568: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
14569: fprintf(ficrespij,"\n");
1.180 brouard 14570: }
1.337 brouard 14571: }
14572: /*}*/
14573: return 0;
1.180 brouard 14574: }
1.218 brouard 14575:
14576: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 14577: /*------------- h Bij x at various ages ------------*/
1.336 brouard 14578: /* To be optimized with precov */
1.217 brouard 14579: int stepsize;
1.218 brouard 14580: /* int agelim; */
14581: int ageminl;
1.217 brouard 14582: int hstepm;
14583: int nhstepm;
1.238 brouard 14584: int h, i, i1, j, k, nres;
1.218 brouard 14585:
1.217 brouard 14586: double agedeb;
14587: double ***p3mat;
1.218 brouard 14588:
14589: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
14590: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
14591: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
14592: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
14593: }
14594: printf("Computing pij back: result on file '%s' \n", filerespijb);
14595: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
14596:
14597: stepsize=(int) (stepm+YEARM-1)/YEARM;
14598: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 14599:
1.218 brouard 14600: /* agelim=AGESUP; */
1.289 brouard 14601: ageminl=AGEINF; /* was 30 */
1.218 brouard 14602: hstepm=stepsize*YEARM; /* Every year of age */
14603: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
14604:
14605: /* hstepm=1; aff par mois*/
14606: pstamp(ficrespijb);
1.255 brouard 14607: fprintf(ficrespijb,"#****** h Bij x Back probability to be in state i at age x-h being in j at x: B1j+B2j+...=1 ");
1.227 brouard 14608: i1= pow(2,cptcoveff);
1.218 brouard 14609: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
14610: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
14611: /* k=k+1; */
1.238 brouard 14612: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 14613: k=TKresult[nres];
1.338 brouard 14614: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 14615: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
14616: /* if(i1 != 1 && TKresult[nres]!= k) */
14617: /* continue; */
14618: fprintf(ficrespijb,"\n#****** ");
14619: for(j=1;j<=cptcovs;j++){
1.338 brouard 14620: fprintf(ficrespijb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.337 brouard 14621: /* for(j=1;j<=cptcoveff;j++) */
14622: /* fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14623: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
14624: /* fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14625: }
14626: fprintf(ficrespijb,"******\n");
14627: if(invalidvarcomb[k]){ /* Is it necessary here? */
14628: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
14629: continue;
14630: }
14631:
14632: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
14633: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
14634: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
14635: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm+0.1)-1; /* Typically 20 years = 20*12/6=40 or 55*12/24=27.5-1.1=>27 */
14636: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
14637:
14638: /* nhstepm=nhstepm*YEARM; aff par mois*/
14639:
14640: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
14641: /* and memory limitations if stepm is small */
14642:
14643: /* oldm=oldms;savm=savms; */
14644: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
14645: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
14646: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
14647: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
14648: for(i=1; i<=nlstate;i++)
14649: for(j=1; j<=nlstate+ndeath;j++)
14650: fprintf(ficrespijb," %1d-%1d",i,j);
14651: fprintf(ficrespijb,"\n");
14652: for (h=0; h<=nhstepm; h++){
14653: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
14654: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
14655: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
1.217 brouard 14656: for(i=1; i<=nlstate;i++)
14657: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 14658: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.217 brouard 14659: fprintf(ficrespijb,"\n");
1.337 brouard 14660: }
14661: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
14662: fprintf(ficrespijb,"\n");
14663: } /* end age deb */
14664: /* } /\* end combination *\/ */
1.238 brouard 14665: } /* end nres */
1.218 brouard 14666: return 0;
14667: } /* hBijx */
1.217 brouard 14668:
1.180 brouard 14669:
1.136 brouard 14670: /***********************************************/
14671: /**************** Main Program *****************/
14672: /***********************************************/
14673:
14674: int main(int argc, char *argv[])
14675: {
14676: #ifdef GSL
14677: const gsl_multimin_fminimizer_type *T;
14678: size_t iteri = 0, it;
14679: int rval = GSL_CONTINUE;
14680: int status = GSL_SUCCESS;
14681: double ssval;
14682: #endif
14683: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 14684: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
14685: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 14686: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 14687: int jj, ll, li, lj, lk;
1.136 brouard 14688: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 14689: int num_filled;
1.136 brouard 14690: int itimes;
14691: int NDIM=2;
14692: int vpopbased=0;
1.235 brouard 14693: int nres=0;
1.258 brouard 14694: int endishere=0;
1.277 brouard 14695: int noffset=0;
1.274 brouard 14696: int ncurrv=0; /* Temporary variable */
14697:
1.164 brouard 14698: char ca[32], cb[32];
1.136 brouard 14699: /* FILE *fichtm; *//* Html File */
14700: /* FILE *ficgp;*/ /*Gnuplot File */
14701: struct stat info;
1.191 brouard 14702: double agedeb=0.;
1.194 brouard 14703:
14704: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 14705: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 14706:
1.361 brouard 14707: double stdpercent; /* for computing the std error of percent e.i: e.i/e.. */
1.165 brouard 14708: double fret;
1.191 brouard 14709: double dum=0.; /* Dummy variable */
1.359 brouard 14710: /* double*** p3mat;*/
1.218 brouard 14711: /* double ***mobaverage; */
1.319 brouard 14712: double wald;
1.164 brouard 14713:
1.351 brouard 14714: char line[MAXLINE], linetmp[MAXLINE];
1.197 brouard 14715: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
14716:
1.234 brouard 14717: char modeltemp[MAXLINE];
1.332 brouard 14718: char resultline[MAXLINE], resultlineori[MAXLINE];
1.230 brouard 14719:
1.136 brouard 14720: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 14721: char *tok, *val; /* pathtot */
1.334 brouard 14722: /* int firstobs=1, lastobs=10; /\* nobs = lastobs-firstobs declared globally ;*\/ */
1.359 brouard 14723: int c, h; /* c2; */
1.191 brouard 14724: int jl=0;
14725: int i1, j1, jk, stepsize=0;
1.194 brouard 14726: int count=0;
14727:
1.164 brouard 14728: int *tab;
1.136 brouard 14729: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 14730: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
14731: /* double anprojf, mprojf, jprojf; */
14732: /* double jintmean,mintmean,aintmean; */
14733: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
14734: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
14735: double yrfproj= 10.0; /* Number of years of forward projections */
14736: double yrbproj= 10.0; /* Number of years of backward projections */
14737: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 14738: int mobilav=0,popforecast=0;
1.191 brouard 14739: int hstepm=0, nhstepm=0;
1.136 brouard 14740: int agemortsup;
14741: float sumlpop=0.;
14742: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
14743: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
14744:
1.191 brouard 14745: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 14746: double ftolpl=FTOL;
14747: double **prlim;
1.217 brouard 14748: double **bprlim;
1.317 brouard 14749: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
14750: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 14751: double ***paramstart; /* Matrix of starting parameter values */
14752: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 14753: double **matcov; /* Matrix of covariance */
1.203 brouard 14754: double **hess; /* Hessian matrix */
1.136 brouard 14755: double ***delti3; /* Scale */
14756: double *delti; /* Scale */
14757: double ***eij, ***vareij;
1.359 brouard 14758: //double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 14759:
1.136 brouard 14760: double *epj, vepp;
1.164 brouard 14761:
1.273 brouard 14762: double dateprev1, dateprev2;
1.296 brouard 14763: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
14764: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
14765:
1.217 brouard 14766:
1.136 brouard 14767: double **ximort;
1.145 brouard 14768: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 14769: int *dcwave;
14770:
1.164 brouard 14771: char z[1]="c";
1.136 brouard 14772:
14773: /*char *strt;*/
14774: char strtend[80];
1.126 brouard 14775:
1.164 brouard 14776:
1.126 brouard 14777: /* setlocale (LC_ALL, ""); */
14778: /* bindtextdomain (PACKAGE, LOCALEDIR); */
14779: /* textdomain (PACKAGE); */
14780: /* setlocale (LC_CTYPE, ""); */
14781: /* setlocale (LC_MESSAGES, ""); */
14782:
14783: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 14784: rstart_time = time(NULL);
14785: /* (void) gettimeofday(&start_time,&tzp);*/
14786: start_time = *localtime(&rstart_time);
1.126 brouard 14787: curr_time=start_time;
1.157 brouard 14788: /*tml = *localtime(&start_time.tm_sec);*/
14789: /* strcpy(strstart,asctime(&tml)); */
14790: strcpy(strstart,asctime(&start_time));
1.126 brouard 14791:
14792: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 14793: /* tp.tm_sec = tp.tm_sec +86400; */
14794: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 14795: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
14796: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
14797: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 14798: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 14799: /* strt=asctime(&tmg); */
14800: /* printf("Time(after) =%s",strstart); */
14801: /* (void) time (&time_value);
14802: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
14803: * tm = *localtime(&time_value);
14804: * strstart=asctime(&tm);
14805: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
14806: */
14807:
14808: nberr=0; /* Number of errors and warnings */
14809: nbwarn=0;
1.184 brouard 14810: #ifdef WIN32
14811: _getcwd(pathcd, size);
14812: #else
1.126 brouard 14813: getcwd(pathcd, size);
1.184 brouard 14814: #endif
1.191 brouard 14815: syscompilerinfo(0);
1.359 brouard 14816: printf("\nIMaCh prax version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 14817: if(argc <=1){
14818: printf("\nEnter the parameter file name: ");
1.205 brouard 14819: if(!fgets(pathr,FILENAMELENGTH,stdin)){
14820: printf("ERROR Empty parameter file name\n");
14821: goto end;
14822: }
1.126 brouard 14823: i=strlen(pathr);
14824: if(pathr[i-1]=='\n')
14825: pathr[i-1]='\0';
1.156 brouard 14826: i=strlen(pathr);
1.205 brouard 14827: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 14828: pathr[i-1]='\0';
1.205 brouard 14829: }
14830: i=strlen(pathr);
14831: if( i==0 ){
14832: printf("ERROR Empty parameter file name\n");
14833: goto end;
14834: }
14835: for (tok = pathr; tok != NULL; ){
1.126 brouard 14836: printf("Pathr |%s|\n",pathr);
14837: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
14838: printf("val= |%s| pathr=%s\n",val,pathr);
14839: strcpy (pathtot, val);
14840: if(pathr[0] == '\0') break; /* Dirty */
14841: }
14842: }
1.281 brouard 14843: else if (argc<=2){
14844: strcpy(pathtot,argv[1]);
14845: }
1.126 brouard 14846: else{
14847: strcpy(pathtot,argv[1]);
1.281 brouard 14848: strcpy(z,argv[2]);
14849: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 14850: }
14851: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
14852: /*cygwin_split_path(pathtot,path,optionfile);
14853: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
14854: /* cutv(path,optionfile,pathtot,'\\');*/
14855:
14856: /* Split argv[0], imach program to get pathimach */
14857: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
14858: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
14859: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
14860: /* strcpy(pathimach,argv[0]); */
14861: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
14862: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
14863: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 14864: #ifdef WIN32
14865: _chdir(path); /* Can be a relative path */
14866: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
14867: #else
1.126 brouard 14868: chdir(path); /* Can be a relative path */
1.184 brouard 14869: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
14870: #endif
14871: printf("Current directory %s!\n",pathcd);
1.126 brouard 14872: strcpy(command,"mkdir ");
14873: strcat(command,optionfilefiname);
14874: if((outcmd=system(command)) != 0){
1.169 brouard 14875: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 14876: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
14877: /* fclose(ficlog); */
14878: /* exit(1); */
14879: }
14880: /* if((imk=mkdir(optionfilefiname))<0){ */
14881: /* perror("mkdir"); */
14882: /* } */
14883:
14884: /*-------- arguments in the command line --------*/
14885:
1.186 brouard 14886: /* Main Log file */
1.126 brouard 14887: strcat(filelog, optionfilefiname);
14888: strcat(filelog,".log"); /* */
14889: if((ficlog=fopen(filelog,"w"))==NULL) {
14890: printf("Problem with logfile %s\n",filelog);
14891: goto end;
14892: }
14893: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 14894: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 14895: fprintf(ficlog,"\nEnter the parameter file name: \n");
14896: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
14897: path=%s \n\
14898: optionfile=%s\n\
14899: optionfilext=%s\n\
1.156 brouard 14900: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 14901:
1.197 brouard 14902: syscompilerinfo(1);
1.167 brouard 14903:
1.126 brouard 14904: printf("Local time (at start):%s",strstart);
14905: fprintf(ficlog,"Local time (at start): %s",strstart);
14906: fflush(ficlog);
14907: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 14908: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 14909:
14910: /* */
14911: strcpy(fileres,"r");
14912: strcat(fileres, optionfilefiname);
1.201 brouard 14913: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 14914: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 14915: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 14916:
1.186 brouard 14917: /* Main ---------arguments file --------*/
1.126 brouard 14918:
14919: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 14920: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
14921: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 14922: fflush(ficlog);
1.149 brouard 14923: /* goto end; */
14924: exit(70);
1.126 brouard 14925: }
14926:
14927: strcpy(filereso,"o");
1.201 brouard 14928: strcat(filereso,fileresu);
1.126 brouard 14929: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
14930: printf("Problem with Output resultfile: %s\n", filereso);
14931: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
14932: fflush(ficlog);
14933: goto end;
14934: }
1.278 brouard 14935: /*-------- Rewriting parameter file ----------*/
14936: strcpy(rfileres,"r"); /* "Rparameterfile */
14937: strcat(rfileres,optionfilefiname); /* Parameter file first name */
14938: strcat(rfileres,"."); /* */
14939: strcat(rfileres,optionfilext); /* Other files have txt extension */
14940: if((ficres =fopen(rfileres,"w"))==NULL) {
14941: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
14942: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
14943: fflush(ficlog);
14944: goto end;
14945: }
14946: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 14947:
1.278 brouard 14948:
1.126 brouard 14949: /* Reads comments: lines beginning with '#' */
14950: numlinepar=0;
1.277 brouard 14951: /* Is it a BOM UTF-8 Windows file? */
14952: /* First parameter line */
1.197 brouard 14953: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 14954: noffset=0;
14955: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
14956: {
14957: noffset=noffset+3;
14958: printf("# File is an UTF8 Bom.\n"); // 0xBF
14959: }
1.302 brouard 14960: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
14961: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 14962: {
14963: noffset=noffset+2;
14964: printf("# File is an UTF16BE BOM file\n");
14965: }
14966: else if( line[0] == 0 && line[1] == 0)
14967: {
14968: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
14969: noffset=noffset+4;
14970: printf("# File is an UTF16BE BOM file\n");
14971: }
14972: } else{
14973: ;/*printf(" Not a BOM file\n");*/
14974: }
14975:
1.197 brouard 14976: /* If line starts with a # it is a comment */
1.277 brouard 14977: if (line[noffset] == '#') {
1.197 brouard 14978: numlinepar++;
14979: fputs(line,stdout);
14980: fputs(line,ficparo);
1.278 brouard 14981: fputs(line,ficres);
1.197 brouard 14982: fputs(line,ficlog);
14983: continue;
14984: }else
14985: break;
14986: }
14987: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
14988: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
14989: if (num_filled != 5) {
14990: printf("Should be 5 parameters\n");
1.283 brouard 14991: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 14992: }
1.126 brouard 14993: numlinepar++;
1.197 brouard 14994: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 14995: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
14996: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
14997: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 14998: }
14999: /* Second parameter line */
15000: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 15001: /* while(fscanf(ficpar,"%[^\n]", line)) { */
15002: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 15003: if (line[0] == '#') {
15004: numlinepar++;
1.283 brouard 15005: printf("%s",line);
15006: fprintf(ficres,"%s",line);
15007: fprintf(ficparo,"%s",line);
15008: fprintf(ficlog,"%s",line);
1.197 brouard 15009: continue;
15010: }else
15011: break;
15012: }
1.223 brouard 15013: if((num_filled=sscanf(line,"ftol=%lf stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\n", \
15014: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
15015: if (num_filled != 11) {
15016: printf("Not 11 parameters, for example:ftol=1.e-8 stepm=12 ncovcol=2 nqv=1 ntv=2 nqtv=1 nlstate=2 ndeath=1 maxwav=3 mle=1 weight=1\n");
1.209 brouard 15017: printf("but line=%s\n",line);
1.283 brouard 15018: fprintf(ficlog,"Not 11 parameters, for example:ftol=1.e-8 stepm=12 ncovcol=2 nqv=1 ntv=2 nqtv=1 nlstate=2 ndeath=1 maxwav=3 mle=1 weight=1\n");
15019: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 15020: }
1.286 brouard 15021: if( lastpass > maxwav){
15022: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
15023: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
15024: fflush(ficlog);
15025: goto end;
15026: }
15027: printf("ftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\n",ftol, stepm, ncovcol, nqv, ntv, nqtv, nlstate, ndeath, maxwav, mle, weightopt);
1.283 brouard 15028: fprintf(ficparo,"ftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\n",ftol, stepm, ncovcol, nqv, ntv, nqtv, nlstate, ndeath, maxwav, mle, weightopt);
1.286 brouard 15029: fprintf(ficres,"ftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\n",ftol, stepm, ncovcol, nqv, ntv, nqtv, nlstate, ndeath, maxwav, 0, weightopt);
1.283 brouard 15030: fprintf(ficlog,"ftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\n",ftol, stepm, ncovcol, nqv, ntv, nqtv, nlstate, ndeath, maxwav, mle, weightopt);
1.126 brouard 15031: }
1.203 brouard 15032: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 15033: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 15034: /* Third parameter line */
15035: while(fgets(line, MAXLINE, ficpar)) {
15036: /* If line starts with a # it is a comment */
15037: if (line[0] == '#') {
15038: numlinepar++;
1.283 brouard 15039: printf("%s",line);
15040: fprintf(ficres,"%s",line);
15041: fprintf(ficparo,"%s",line);
15042: fprintf(ficlog,"%s",line);
1.197 brouard 15043: continue;
15044: }else
15045: break;
15046: }
1.351 brouard 15047: if((num_filled=sscanf(line,"model=%[^.\n]", model)) !=EOF){ /* Every character after model but dot and return */
15048: if (num_filled != 1){
15049: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
15050: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
15051: model[0]='\0';
15052: goto end;
15053: }else{
15054: trimbtab(linetmp,line); /* Trims multiple blanks in line */
15055: strcpy(line, linetmp);
15056: }
15057: }
15058: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){ /* Every character after 1+age but dot and return */
1.279 brouard 15059: if (num_filled != 1){
1.302 brouard 15060: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
15061: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 15062: model[0]='\0';
15063: goto end;
15064: }
15065: else{
15066: if (model[0]=='+'){
15067: for(i=1; i<=strlen(model);i++)
15068: modeltemp[i-1]=model[i];
1.201 brouard 15069: strcpy(model,modeltemp);
1.197 brouard 15070: }
15071: }
1.338 brouard 15072: /* printf(" model=1+age%s modeltemp= %s, model=1+age+%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 15073: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 15074: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
15075: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
15076: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 15077: }
15078: /* fscanf(ficpar,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\nftol=%lf stepm=%d ncovcol=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d model=1+age+%s\n",title, datafile, &lastobs, &firstpass,&lastpass,&ftol, &stepm, &ncovcol, &nlstate,&ndeath, &maxwav, &mle, &weightopt,model); */
15079: /* numlinepar=numlinepar+3; /\* In general *\/ */
15080: /* printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\nftol=%e stepm=%d ncovcol=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\nmodel=1+age+%s\n", title, datafile, lastobs, firstpass,lastpass,ftol, stepm, ncovcol, nlstate,ndeath, maxwav, mle, weightopt,model); */
1.283 brouard 15081: /* fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\nftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\nmodel=1+age+%s.\n", title, datafile, lastobs, firstpass,lastpass,ftol,stepm,ncovcol, nqv, ntv, nqtv, nlstate,ndeath,maxwav, mle, weightopt,model); */
15082: /* fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\nftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\nmodel=1+age+%s.\n", title, datafile, lastobs, firstpass,lastpass,ftol,stepm,ncovcol, nqv, ntv, nqtv, nlstate,ndeath,maxwav, mle, weightopt,model); */
1.126 brouard 15083: fflush(ficlog);
1.190 brouard 15084: /* if(model[0]=='#'|| model[0]== '\0'){ */
15085: if(model[0]=='#'){
1.279 brouard 15086: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
15087: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
15088: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 15089: if(mle != -1){
1.279 brouard 15090: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter vectors and subdiagonal covariance matrix.\n");
1.187 brouard 15091: exit(1);
15092: }
15093: }
1.126 brouard 15094: while((c=getc(ficpar))=='#' && c!= EOF){
15095: ungetc(c,ficpar);
15096: fgets(line, MAXLINE, ficpar);
15097: numlinepar++;
1.195 brouard 15098: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
15099: z[0]=line[1];
1.342 brouard 15100: }else if(line[1]=='d'){ /* For debugging individual values of covariates in ficresilk */
1.343 brouard 15101: debugILK=1;printf("DebugILK\n");
1.195 brouard 15102: }
15103: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 15104: fputs(line, stdout);
15105: //puts(line);
1.126 brouard 15106: fputs(line,ficparo);
15107: fputs(line,ficlog);
15108: }
15109: ungetc(c,ficpar);
15110:
15111:
1.290 brouard 15112: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
15113: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
15114: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
1.341 brouard 15115: /* if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /\**< Time varying covariate (dummy and quantitative)*\/ */
15116: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs); /**< Might be better */
1.136 brouard 15117: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
15118: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
15119: v1+v2*age+v2*v3 makes cptcovn = 3
15120: */
15121: if (strlen(model)>1)
1.187 brouard 15122: ncovmodel=2+nbocc(model,'+')+1; /*Number of variables including intercept and age = cptcovn + intercept + age : v1+v2+v3+v2*v4+v5*age makes 5+2=7,age*age makes 3*/
1.145 brouard 15123: else
1.187 brouard 15124: ncovmodel=2; /* Constant and age */
1.133 brouard 15125: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
15126: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 15127: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
15128: printf("Too complex model for current IMaCh: npar=(nlstate+ndeath-1)*nlstate*ncovmodel=%d >= %d(MAXPARM) or nlstate=%d >= %d(NLSTATEMAX) or ndeath=%d >= %d(NDEATHMAX) or ncovmodel=(k+age+#of+signs)=%d(NCOVMAX) >= %d\n",npar, MAXPARM, nlstate, NLSTATEMAX, ndeath, NDEATHMAX, ncovmodel, NCOVMAX);
15129: fprintf(ficlog,"Too complex model for current IMaCh: %d >=%d(MAXPARM) or %d >=%d(NLSTATEMAX) or %d >=%d(NDEATHMAX) or %d(NCOVMAX) >=%d\n",npar, MAXPARM, nlstate, NLSTATEMAX, ndeath, NDEATHMAX, ncovmodel, NCOVMAX);
15130: fflush(stdout);
15131: fclose (ficlog);
15132: goto end;
15133: }
1.126 brouard 15134: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
15135: delti=delti3[1][1];
15136: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
15137: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 15138: /* We could also provide initial parameters values giving by simple logistic regression
15139: * only one way, that is without matrix product. We will have nlstate maximizations */
15140: /* for(i=1;i<nlstate;i++){ */
15141: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
15142: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
15143: /* } */
1.126 brouard 15144: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 15145: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
15146: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 15147: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
15148: fclose (ficparo);
15149: fclose (ficlog);
15150: goto end;
15151: exit(0);
1.220 brouard 15152: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 15153: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 15154: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
15155: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 15156: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
15157: matcov=matrix(1,npar,1,npar);
1.203 brouard 15158: hess=matrix(1,npar,1,npar);
1.220 brouard 15159: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 15160: /* Read guessed parameters */
1.126 brouard 15161: /* Reads comments: lines beginning with '#' */
15162: while((c=getc(ficpar))=='#' && c!= EOF){
15163: ungetc(c,ficpar);
15164: fgets(line, MAXLINE, ficpar);
15165: numlinepar++;
1.141 brouard 15166: fputs(line,stdout);
1.126 brouard 15167: fputs(line,ficparo);
15168: fputs(line,ficlog);
15169: }
15170: ungetc(c,ficpar);
15171:
15172: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 15173: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 15174: for(i=1; i <=nlstate; i++){
1.234 brouard 15175: j=0;
1.126 brouard 15176: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 15177: if(jj==i) continue;
15178: j++;
1.292 brouard 15179: while((c=getc(ficpar))=='#' && c!= EOF){
15180: ungetc(c,ficpar);
15181: fgets(line, MAXLINE, ficpar);
15182: numlinepar++;
15183: fputs(line,stdout);
15184: fputs(line,ficparo);
15185: fputs(line,ficlog);
15186: }
15187: ungetc(c,ficpar);
1.234 brouard 15188: fscanf(ficpar,"%1d%1d",&i1,&j1);
15189: if ((i1 != i) || (j1 != jj)){
15190: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 15191: It might be a problem of design; if ncovcol and the model are correct\n \
15192: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 15193: exit(1);
15194: }
15195: fprintf(ficparo,"%1d%1d",i1,j1);
15196: if(mle==1)
15197: printf("%1d%1d",i,jj);
15198: fprintf(ficlog,"%1d%1d",i,jj);
15199: for(k=1; k<=ncovmodel;k++){
15200: fscanf(ficpar," %lf",¶m[i][j][k]);
15201: if(mle==1){
15202: printf(" %lf",param[i][j][k]);
15203: fprintf(ficlog," %lf",param[i][j][k]);
15204: }
15205: else
15206: fprintf(ficlog," %lf",param[i][j][k]);
15207: fprintf(ficparo," %lf",param[i][j][k]);
15208: }
15209: fscanf(ficpar,"\n");
15210: numlinepar++;
15211: if(mle==1)
15212: printf("\n");
15213: fprintf(ficlog,"\n");
15214: fprintf(ficparo,"\n");
1.126 brouard 15215: }
15216: }
15217: fflush(ficlog);
1.234 brouard 15218:
1.251 brouard 15219: /* Reads parameters values */
1.126 brouard 15220: p=param[1][1];
1.251 brouard 15221: pstart=paramstart[1][1];
1.126 brouard 15222:
15223: /* Reads comments: lines beginning with '#' */
15224: while((c=getc(ficpar))=='#' && c!= EOF){
15225: ungetc(c,ficpar);
15226: fgets(line, MAXLINE, ficpar);
15227: numlinepar++;
1.141 brouard 15228: fputs(line,stdout);
1.126 brouard 15229: fputs(line,ficparo);
15230: fputs(line,ficlog);
15231: }
15232: ungetc(c,ficpar);
15233:
15234: for(i=1; i <=nlstate; i++){
15235: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 15236: fscanf(ficpar,"%1d%1d",&i1,&j1);
15237: if ( (i1-i) * (j1-j) != 0){
15238: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
15239: exit(1);
15240: }
15241: printf("%1d%1d",i,j);
15242: fprintf(ficparo,"%1d%1d",i1,j1);
15243: fprintf(ficlog,"%1d%1d",i1,j1);
15244: for(k=1; k<=ncovmodel;k++){
15245: fscanf(ficpar,"%le",&delti3[i][j][k]);
15246: printf(" %le",delti3[i][j][k]);
15247: fprintf(ficparo," %le",delti3[i][j][k]);
15248: fprintf(ficlog," %le",delti3[i][j][k]);
15249: }
15250: fscanf(ficpar,"\n");
15251: numlinepar++;
15252: printf("\n");
15253: fprintf(ficparo,"\n");
15254: fprintf(ficlog,"\n");
1.126 brouard 15255: }
15256: }
15257: fflush(ficlog);
1.234 brouard 15258:
1.145 brouard 15259: /* Reads covariance matrix */
1.126 brouard 15260: delti=delti3[1][1];
1.220 brouard 15261:
15262:
1.126 brouard 15263: /* free_ma3x(delti3,1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */ /* Hasn't to to freed here otherwise delti is no more allocated */
1.220 brouard 15264:
1.126 brouard 15265: /* Reads comments: lines beginning with '#' */
15266: while((c=getc(ficpar))=='#' && c!= EOF){
15267: ungetc(c,ficpar);
15268: fgets(line, MAXLINE, ficpar);
15269: numlinepar++;
1.141 brouard 15270: fputs(line,stdout);
1.126 brouard 15271: fputs(line,ficparo);
15272: fputs(line,ficlog);
15273: }
15274: ungetc(c,ficpar);
1.220 brouard 15275:
1.126 brouard 15276: matcov=matrix(1,npar,1,npar);
1.203 brouard 15277: hess=matrix(1,npar,1,npar);
1.131 brouard 15278: for(i=1; i <=npar; i++)
15279: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 15280:
1.194 brouard 15281: /* Scans npar lines */
1.126 brouard 15282: for(i=1; i <=npar; i++){
1.226 brouard 15283: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 15284: if(count != 3){
1.226 brouard 15285: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 15286: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
15287: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 15288: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 15289: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
15290: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 15291: exit(1);
1.220 brouard 15292: }else{
1.226 brouard 15293: if(mle==1)
15294: printf("%1d%1d%d",i1,j1,jk);
15295: }
15296: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
15297: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 15298: for(j=1; j <=i; j++){
1.226 brouard 15299: fscanf(ficpar," %le",&matcov[i][j]);
15300: if(mle==1){
15301: printf(" %.5le",matcov[i][j]);
15302: }
15303: fprintf(ficlog," %.5le",matcov[i][j]);
15304: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 15305: }
15306: fscanf(ficpar,"\n");
15307: numlinepar++;
15308: if(mle==1)
1.220 brouard 15309: printf("\n");
1.126 brouard 15310: fprintf(ficlog,"\n");
15311: fprintf(ficparo,"\n");
15312: }
1.194 brouard 15313: /* End of read covariance matrix npar lines */
1.126 brouard 15314: for(i=1; i <=npar; i++)
15315: for(j=i+1;j<=npar;j++)
1.226 brouard 15316: matcov[i][j]=matcov[j][i];
1.126 brouard 15317:
15318: if(mle==1)
15319: printf("\n");
15320: fprintf(ficlog,"\n");
15321:
15322: fflush(ficlog);
15323:
15324: } /* End of mle != -3 */
1.218 brouard 15325:
1.186 brouard 15326: /* Main data
15327: */
1.290 brouard 15328: nobs=lastobs-firstobs+1; /* was = lastobs;*/
15329: /* num=lvector(1,n); */
15330: /* moisnais=vector(1,n); */
15331: /* annais=vector(1,n); */
15332: /* moisdc=vector(1,n); */
15333: /* andc=vector(1,n); */
15334: /* weight=vector(1,n); */
15335: /* agedc=vector(1,n); */
15336: /* cod=ivector(1,n); */
15337: /* for(i=1;i<=n;i++){ */
15338: num=lvector(firstobs,lastobs);
15339: moisnais=vector(firstobs,lastobs);
15340: annais=vector(firstobs,lastobs);
15341: moisdc=vector(firstobs,lastobs);
15342: andc=vector(firstobs,lastobs);
15343: weight=vector(firstobs,lastobs);
15344: agedc=vector(firstobs,lastobs);
15345: cod=ivector(firstobs,lastobs);
15346: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 15347: num[i]=0;
15348: moisnais[i]=0;
15349: annais[i]=0;
15350: moisdc[i]=0;
15351: andc[i]=0;
15352: agedc[i]=0;
15353: cod[i]=0;
15354: weight[i]=1.0; /* Equal weights, 1 by default */
15355: }
1.290 brouard 15356: mint=matrix(1,maxwav,firstobs,lastobs);
15357: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 15358: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.336 brouard 15359: /* printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel)); */
1.126 brouard 15360: tab=ivector(1,NCOVMAX);
1.144 brouard 15361: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 15362: ncodemaxwundef=ivector(1,NCOVMAX); /* Number of code per covariate; if - 1 O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.126 brouard 15363:
1.136 brouard 15364: /* Reads data from file datafile */
15365: if (readdata(datafile, firstobs, lastobs, &imx)==1)
15366: goto end;
15367:
15368: /* Calculation of the number of parameters from char model */
1.234 brouard 15369: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 15370: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
15371: k=3 V4 Tvar[k=3]= 4 (from V4)
15372: k=2 V1 Tvar[k=2]= 1 (from V1)
15373: k=1 Tvar[1]=2 (from V2)
1.234 brouard 15374: */
15375:
15376: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
15377: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 brouard 15378: TnsdVar=ivector(1,NCOVMAX); /* */
1.335 brouard 15379: /* for(i=1; i<=NCOVMAX;i++) TnsdVar[i]=3; */
1.234 brouard 15380: TvarsD=ivector(1,NCOVMAX); /* */
15381: TvarsQind=ivector(1,NCOVMAX); /* */
15382: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 15383: TvarF=ivector(1,NCOVMAX); /* */
15384: TvarFind=ivector(1,NCOVMAX); /* */
15385: TvarV=ivector(1,NCOVMAX); /* */
15386: TvarVind=ivector(1,NCOVMAX); /* */
15387: TvarA=ivector(1,NCOVMAX); /* */
15388: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 15389: TvarFD=ivector(1,NCOVMAX); /* */
15390: TvarFDind=ivector(1,NCOVMAX); /* */
15391: TvarFQ=ivector(1,NCOVMAX); /* */
15392: TvarFQind=ivector(1,NCOVMAX); /* */
15393: TvarVD=ivector(1,NCOVMAX); /* */
15394: TvarVDind=ivector(1,NCOVMAX); /* */
15395: TvarVQ=ivector(1,NCOVMAX); /* */
15396: TvarVQind=ivector(1,NCOVMAX); /* */
1.339 brouard 15397: TvarVV=ivector(1,NCOVMAX); /* */
15398: TvarVVind=ivector(1,NCOVMAX); /* */
1.349 brouard 15399: TvarVVA=ivector(1,NCOVMAX); /* */
15400: TvarVVAind=ivector(1,NCOVMAX); /* */
15401: TvarAVVA=ivector(1,NCOVMAX); /* */
15402: TvarAVVAind=ivector(1,NCOVMAX); /* */
1.231 brouard 15403:
1.230 brouard 15404: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 15405: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 15406: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
15407: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
15408: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.349 brouard 15409: DummyV=ivector(-1,NCOVMAX); /* 1 to 3 */
15410: FixedV=ivector(-1,NCOVMAX); /* 1 to 3 */
15411:
1.137 brouard 15412: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
15413: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
15414: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
15415: */
15416: /* For model-covariate k tells which data-covariate to use but
15417: because this model-covariate is a construction we invent a new column
15418: ncovcol + k1
15419: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
15420: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 15421: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
15422: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 15423: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
15424: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 15425: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 15426: */
1.145 brouard 15427: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
15428: Tvard=imatrix(1,NCOVMAX,1,2); /* n=Tvard[k1][1] and m=Tvard[k1][2] gives the couple n,m of the k1 th product Vn*Vm
1.141 brouard 15429: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
15430: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.351 brouard 15431: Tvardk=imatrix(0,NCOVMAX,1,2);
1.145 brouard 15432: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 15433: 4 covariates (3 plus signs)
15434: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 15435: */
15436: for(i=1;i<NCOVMAX;i++)
15437: Tage[i]=0;
1.230 brouard 15438: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 15439: * individual dummy, fixed or varying:
15440: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
15441: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 15442: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
15443: * V1 df, V2 qf, V3 & V4 dv, V5 qv
15444: * Tmodelind[1]@9={9,0,3,2,}*/
15445: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
15446: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 15447: * individual quantitative, fixed or varying:
15448: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
15449: * 3, 1, 0, 0, 0, 0, 0, 0},
15450: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.349 brouard 15451:
15452: /* Probably useless zeroes */
15453: for(i=1;i<NCOVMAX;i++){
15454: DummyV[i]=0;
15455: FixedV[i]=0;
15456: }
15457:
15458: for(i=1; i <=ncovcol;i++){
15459: DummyV[i]=0;
15460: FixedV[i]=0;
15461: }
15462: for(i=ncovcol+1; i <=ncovcol+nqv;i++){
15463: DummyV[i]=1;
15464: FixedV[i]=0;
15465: }
15466: for(i=ncovcol+nqv+1; i <=ncovcol+nqv+ntv;i++){
15467: DummyV[i]=0;
15468: FixedV[i]=1;
15469: }
15470: for(i=ncovcol+nqv+ntv+1; i <=ncovcol+nqv+ntv+nqtv;i++){
15471: DummyV[i]=1;
15472: FixedV[i]=1;
15473: }
15474: for(i=1; i <=ncovcol+nqv+ntv+nqtv;i++){
15475: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",i,i,DummyV[i],i,FixedV[i]);
15476: fprintf(ficlog,"Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",i,i,DummyV[i],i,FixedV[i]);
15477: }
15478:
15479:
15480:
1.186 brouard 15481: /* Main decodemodel */
15482:
1.187 brouard 15483:
1.223 brouard 15484: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 15485: goto end;
15486:
1.137 brouard 15487: if((double)(lastobs-imx)/(double)imx > 1.10){
15488: nbwarn++;
15489: printf("Warning: The value of parameter lastobs=%d is big compared to the \n effective number of cases imx=%d, please adjust, \n otherwise you are allocating more memory than necessary.\n",lastobs, imx);
15490: fprintf(ficlog,"Warning: The value of parameter lastobs=%d is big compared to the \n effective number of cases imx=%d, please adjust, \n otherwise you are allocating more memory than necessary.\n",lastobs, imx);
15491: }
1.136 brouard 15492: /* if(mle==1){*/
1.137 brouard 15493: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
15494: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 15495: }
15496:
15497: /*-calculation of age at interview from date of interview and age at death -*/
15498: agev=matrix(1,maxwav,1,imx);
15499:
15500: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
15501: goto end;
15502:
1.126 brouard 15503:
1.136 brouard 15504: agegomp=(int)agemin;
1.290 brouard 15505: free_vector(moisnais,firstobs,lastobs);
15506: free_vector(annais,firstobs,lastobs);
1.126 brouard 15507: /* free_matrix(mint,1,maxwav,1,n);
15508: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 15509: /* free_vector(moisdc,1,n); */
15510: /* free_vector(andc,1,n); */
1.145 brouard 15511: /* */
15512:
1.126 brouard 15513: wav=ivector(1,imx);
1.214 brouard 15514: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
15515: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
15516: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
15517: dh=imatrix(1,lastpass-firstpass+2,1,imx); /* We are adding a wave if status is unknown at last wave but death occurs after last wave.*/
15518: bh=imatrix(1,lastpass-firstpass+2,1,imx);
15519: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 15520:
15521: /* Concatenates waves */
1.214 brouard 15522: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
15523: Death is a valid wave (if date is known).
15524: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
15525: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
15526: and mw[mi+1][i]. dh depends on stepm.
15527: */
15528:
1.126 brouard 15529: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 15530: /* Concatenates waves */
1.145 brouard 15531:
1.290 brouard 15532: free_vector(moisdc,firstobs,lastobs);
15533: free_vector(andc,firstobs,lastobs);
1.215 brouard 15534:
1.126 brouard 15535: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
15536: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
15537: ncodemax[1]=1;
1.145 brouard 15538: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 15539: cptcoveff=0;
1.220 brouard 15540: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
1.335 brouard 15541: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; as well as calculate cptcoveff or number of total effective dummy covariates*/
1.227 brouard 15542: }
15543:
15544: ncovcombmax=pow(2,cptcoveff);
1.338 brouard 15545: invalidvarcomb=ivector(0, ncovcombmax);
15546: for(i=0;i<ncovcombmax;i++)
1.227 brouard 15547: invalidvarcomb[i]=0;
15548:
1.211 brouard 15549: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 15550: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 15551: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 15552:
1.200 brouard 15553: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 15554: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 15555: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 15556: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
15557: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
15558: * (currently 0 or 1) in the data.
15559: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
15560: * corresponding modality (h,j).
15561: */
15562:
1.145 brouard 15563: h=0;
15564: /*if (cptcovn > 0) */
1.126 brouard 15565: m=pow(2,cptcoveff);
15566:
1.144 brouard 15567: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 15568: * For k=4 covariates, h goes from 1 to m=2**k
15569: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
15570: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 15571: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
15572: *______________________________ *______________________
15573: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
15574: * 2 2 1 1 1 * 1 0 0 0 1
15575: * 3 i=2 1 2 1 1 * 2 0 0 1 0
15576: * 4 2 2 1 1 * 3 0 0 1 1
15577: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
15578: * 6 2 1 2 1 * 5 0 1 0 1
15579: * 7 i=4 1 2 2 1 * 6 0 1 1 0
15580: * 8 2 2 2 1 * 7 0 1 1 1
15581: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
15582: * 10 2 1 1 2 * 9 1 0 0 1
15583: * 11 i=6 1 2 1 2 * 10 1 0 1 0
15584: * 12 2 2 1 2 * 11 1 0 1 1
15585: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
15586: * 14 2 1 2 2 * 13 1 1 0 1
15587: * 15 i=8 1 2 2 2 * 14 1 1 1 0
15588: * 16 2 2 2 2 * 15 1 1 1 1
15589: */
1.212 brouard 15590: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 15591: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
15592: * and the value of each covariate?
15593: * V1=1, V2=1, V3=2, V4=1 ?
15594: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
15595: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
15596: * In order to get the real value in the data, we use nbcode
15597: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
15598: * We are keeping this crazy system in order to be able (in the future?)
15599: * to have more than 2 values (0 or 1) for a covariate.
15600: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
15601: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
15602: * bbbbbbbb
15603: * 76543210
15604: * h-1 00000101 (6-1=5)
1.219 brouard 15605: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 15606: * &
15607: * 1 00000001 (1)
1.219 brouard 15608: * 00000000 = 1 & ((h-1) >> (k-1))
15609: * +1= 00000001 =1
1.211 brouard 15610: *
15611: * h=14, k=3 => h'=h-1=13, k'=k-1=2
15612: * h' 1101 =2^3+2^2+0x2^1+2^0
15613: * >>k' 11
15614: * & 00000001
15615: * = 00000001
15616: * +1 = 00000010=2 = codtabm(14,3)
15617: * Reverse h=6 and m=16?
15618: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
15619: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
15620: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
15621: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
15622: * V3=decodtabm(14,3,2**4)=2
15623: * h'=13 1101 =2^3+2^2+0x2^1+2^0
15624: *(h-1) >> (j-1) 0011 =13 >> 2
15625: * &1 000000001
15626: * = 000000001
15627: * +1= 000000010 =2
15628: * 2211
15629: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
15630: * V3=2
1.220 brouard 15631: * codtabm and decodtabm are identical
1.211 brouard 15632: */
15633:
1.145 brouard 15634:
15635: free_ivector(Ndum,-1,NCOVMAX);
15636:
15637:
1.126 brouard 15638:
1.186 brouard 15639: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 15640: strcpy(optionfilegnuplot,optionfilefiname);
15641: if(mle==-3)
1.201 brouard 15642: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 15643: strcat(optionfilegnuplot,".gp");
15644:
15645: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
15646: printf("Problem with file %s",optionfilegnuplot);
15647: }
15648: else{
1.204 brouard 15649: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 15650: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 15651: //fprintf(ficgp,"set missing 'NaNq'\n");
15652: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 15653: }
15654: /* fclose(ficgp);*/
1.186 brouard 15655:
15656:
15657: /* Initialisation of --------- index.htm --------*/
1.126 brouard 15658:
15659: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
15660: if(mle==-3)
1.201 brouard 15661: strcat(optionfilehtm,"-MORT_");
1.126 brouard 15662: strcat(optionfilehtm,".htm");
15663: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 15664: printf("Problem with %s \n",optionfilehtm);
15665: exit(0);
1.126 brouard 15666: }
15667:
15668: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
15669: strcat(optionfilehtmcov,"-cov.htm");
15670: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
15671: printf("Problem with %s \n",optionfilehtmcov), exit(0);
15672: }
15673: else{
15674: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
15675: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 15676: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 15677: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
15678: }
15679:
1.335 brouard 15680: fprintf(fichtm,"<html><head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n\
15681: <title>IMaCh %s</title></head>\n\
15682: <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n\
15683: <font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>\
15684: -EUROREVES-Institut de longévité-2013-2022-Japan Society for the Promotion of Sciences 日本学術振興会 \
15685: (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - \
15686: <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \n", optionfilehtm);
15687:
15688: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 15689: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 15690: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.337 brouard 15691: This file: <a href=\"%s\">%s</a></br>Title=%s <br>Datafile=<a href=\"%s\">%s</a> Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 15692: \n\
15693: <hr size=\"2\" color=\"#EC5E5E\">\
15694: <ul><li><h4>Parameter files</h4>\n\
15695: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
15696: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
15697: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
15698: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
15699: - Date and time at start: %s</ul>\n",\
1.335 brouard 15700: version,fullversion,optionfilehtm,optionfilehtm,title,datafile,datafile,firstpass,lastpass,stepm, weightopt, model, \
1.126 brouard 15701: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
15702: fileres,fileres,\
15703: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
15704: fflush(fichtm);
15705:
15706: strcpy(pathr,path);
15707: strcat(pathr,optionfilefiname);
1.184 brouard 15708: #ifdef WIN32
15709: _chdir(optionfilefiname); /* Move to directory named optionfile */
15710: #else
1.126 brouard 15711: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 15712: #endif
15713:
1.126 brouard 15714:
1.220 brouard 15715: /* Calculates basic frequencies. Computes observed prevalence at single age
15716: and for any valid combination of covariates
1.126 brouard 15717: and prints on file fileres'p'. */
1.359 brouard 15718: freqsummary(fileres, p, pstart, (double)agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 15719: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 15720:
15721: fprintf(fichtm,"\n");
1.286 brouard 15722: fprintf(fichtm,"<h4>Parameter line 2</h4><ul><li>Tolerance for the convergence of the likelihood: ftol=%g \n<li>Interval for the elementary matrix (in month): stepm=%d",\
1.274 brouard 15723: ftol, stepm);
15724: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
15725: ncurrv=1;
15726: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
15727: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
15728: ncurrv=i;
15729: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 15730: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 15731: ncurrv=i;
15732: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 15733: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 15734: ncurrv=i;
15735: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
15736: fprintf(fichtm,"\n<li>Weights column \n<br>Number of alive states: nlstate=%d <br>Number of death states (not really implemented): ndeath=%d \n<li>Number of waves: maxwav=%d \n<li>Parameter for maximization (1), using parameter values (0), for design of parameters and variance-covariance matrix: mle=%d \n<li>Does the weight column be taken into account (1), or not (0): weight=%d</ul>\n", \
15737: nlstate, ndeath, maxwav, mle, weightopt);
15738:
15739: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
15740: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
15741:
15742:
1.317 brouard 15743: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 15744: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
15745: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 15746: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 15747: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 15748: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
15749: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
15750: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
15751: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 15752:
1.126 brouard 15753: /* For Powell, parameters are in a vector p[] starting at p[1]
15754: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
15755: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
15756:
15757: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 15758: /* For mortality only */
1.126 brouard 15759: if (mle==-3){
1.136 brouard 15760: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 15761: for(i=1;i<=NDIM;i++)
15762: for(j=1;j<=NDIM;j++)
15763: ximort[i][j]=0.;
1.186 brouard 15764: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 15765: cens=ivector(firstobs,lastobs);
15766: ageexmed=vector(firstobs,lastobs);
15767: agecens=vector(firstobs,lastobs);
15768: dcwave=ivector(firstobs,lastobs);
1.223 brouard 15769:
1.126 brouard 15770: for (i=1; i<=imx; i++){
15771: dcwave[i]=-1;
15772: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 15773: if (s[m][i]>nlstate) {
15774: dcwave[i]=m;
15775: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
15776: break;
15777: }
1.126 brouard 15778: }
1.226 brouard 15779:
1.126 brouard 15780: for (i=1; i<=imx; i++) {
15781: if (wav[i]>0){
1.226 brouard 15782: ageexmed[i]=agev[mw[1][i]][i];
15783: j=wav[i];
15784: agecens[i]=1.;
15785:
15786: if (ageexmed[i]> 1 && wav[i] > 0){
15787: agecens[i]=agev[mw[j][i]][i];
15788: cens[i]= 1;
15789: }else if (ageexmed[i]< 1)
15790: cens[i]= -1;
15791: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
15792: cens[i]=0 ;
1.126 brouard 15793: }
15794: else cens[i]=-1;
15795: }
15796:
15797: for (i=1;i<=NDIM;i++) {
15798: for (j=1;j<=NDIM;j++)
1.226 brouard 15799: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 15800: }
15801:
1.302 brouard 15802: p[1]=0.0268; p[NDIM]=0.083;
15803: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 15804:
15805:
1.136 brouard 15806: #ifdef GSL
15807: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 15808: #else
1.359 brouard 15809: printf("Powell-mort\n"); fprintf(ficlog,"Powell-mort\n");
1.136 brouard 15810: #endif
1.201 brouard 15811: strcpy(filerespow,"POW-MORT_");
15812: strcat(filerespow,fileresu);
1.126 brouard 15813: if((ficrespow=fopen(filerespow,"w"))==NULL) {
15814: printf("Problem with resultfile: %s\n", filerespow);
15815: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
15816: }
1.136 brouard 15817: #ifdef GSL
15818: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 15819: #else
1.126 brouard 15820: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 15821: #endif
1.126 brouard 15822: /* for (i=1;i<=nlstate;i++)
15823: for(j=1;j<=nlstate+ndeath;j++)
15824: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
15825: */
15826: fprintf(ficrespow,"\n");
1.136 brouard 15827: #ifdef GSL
15828: /* gsl starts here */
15829: T = gsl_multimin_fminimizer_nmsimplex;
15830: gsl_multimin_fminimizer *sfm = NULL;
15831: gsl_vector *ss, *x;
15832: gsl_multimin_function minex_func;
15833:
15834: /* Initial vertex size vector */
15835: ss = gsl_vector_alloc (NDIM);
15836:
15837: if (ss == NULL){
15838: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
15839: }
15840: /* Set all step sizes to 1 */
15841: gsl_vector_set_all (ss, 0.001);
15842:
15843: /* Starting point */
1.126 brouard 15844:
1.136 brouard 15845: x = gsl_vector_alloc (NDIM);
15846:
15847: if (x == NULL){
15848: gsl_vector_free(ss);
15849: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
15850: }
15851:
15852: /* Initialize method and iterate */
15853: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 15854: /* gsl_vector_set(x, 0, 0.0268); */
15855: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 15856: gsl_vector_set(x, 0, p[1]);
15857: gsl_vector_set(x, 1, p[2]);
15858:
15859: minex_func.f = &gompertz_f;
15860: minex_func.n = NDIM;
15861: minex_func.params = (void *)&p; /* ??? */
15862:
15863: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
15864: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
15865:
15866: printf("Iterations beginning .....\n\n");
15867: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
15868:
15869: iteri=0;
15870: while (rval == GSL_CONTINUE){
15871: iteri++;
15872: status = gsl_multimin_fminimizer_iterate(sfm);
15873:
15874: if (status) printf("error: %s\n", gsl_strerror (status));
15875: fflush(0);
15876:
15877: if (status)
15878: break;
15879:
15880: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
15881: ssval = gsl_multimin_fminimizer_size (sfm);
15882:
15883: if (rval == GSL_SUCCESS)
15884: printf ("converged to a local maximum at\n");
15885:
15886: printf("%5d ", iteri);
15887: for (it = 0; it < NDIM; it++){
15888: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
15889: }
15890: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
15891: }
15892:
15893: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
15894:
15895: gsl_vector_free(x); /* initial values */
15896: gsl_vector_free(ss); /* inital step size */
15897: for (it=0; it<NDIM; it++){
15898: p[it+1]=gsl_vector_get(sfm->x,it);
15899: fprintf(ficrespow," %.12lf", p[it]);
15900: }
15901: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
15902: #endif
15903: #ifdef POWELL
1.361 brouard 15904: #ifdef LINMINORIGINAL
15905: #else /* LINMINORIGINAL */
15906:
15907: flatdir=ivector(1,npar);
15908: for (j=1;j<=npar;j++) flatdir[j]=0;
15909: #endif /*LINMINORIGINAL */
1.362 brouard 15910: /* powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz); */
15911: /* double h0=0.25; */
15912: macheps=pow(16.0,-13.0);
15913: printf("Praxis Gegenfurtner mle=%d\n",mle);
15914: fprintf(ficlog, "Praxis Gegenfurtner mle=%d\n", mle);fflush(ficlog);
15915: /* ffmin = praxis(ftol,macheps, h0, npar, prin, p, gompertz); */
15916: /* For the Gompertz we use only two parameters */
15917: int _npar=2;
15918: ffmin = praxis(ftol,macheps, h0, _npar, 4, p, gompertz);
15919: printf("End Praxis\n");
1.126 brouard 15920: fclose(ficrespow);
1.361 brouard 15921: #ifdef LINMINORIGINAL
15922: #else
15923: free_ivector(flatdir,1,npar);
15924: #endif /* LINMINORIGINAL*/
1.364 brouard 15925: #endif /* POWELL */
1.203 brouard 15926: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 15927:
15928: for(i=1; i <=NDIM; i++)
15929: for(j=i+1;j<=NDIM;j++)
1.359 brouard 15930: matcov[i][j]=matcov[j][i];
1.126 brouard 15931:
15932: printf("\nCovariance matrix\n ");
1.203 brouard 15933: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 15934: for(i=1; i <=NDIM; i++) {
15935: for(j=1;j<=NDIM;j++){
1.220 brouard 15936: printf("%f ",matcov[i][j]);
15937: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 15938: }
1.203 brouard 15939: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 15940: }
15941:
15942: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 15943: for (i=1;i<=NDIM;i++) {
1.126 brouard 15944: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 15945: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
15946: }
1.302 brouard 15947: lsurv=vector(agegomp,AGESUP);
15948: lpop=vector(agegomp,AGESUP);
15949: tpop=vector(agegomp,AGESUP);
1.126 brouard 15950: lsurv[agegomp]=100000;
15951:
15952: for (k=agegomp;k<=AGESUP;k++) {
15953: agemortsup=k;
15954: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
15955: }
15956:
15957: for (k=agegomp;k<agemortsup;k++)
15958: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
15959:
15960: for (k=agegomp;k<agemortsup;k++){
15961: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
15962: sumlpop=sumlpop+lpop[k];
15963: }
15964:
15965: tpop[agegomp]=sumlpop;
15966: for (k=agegomp;k<(agemortsup-3);k++){
15967: /* tpop[k+1]=2;*/
15968: tpop[k+1]=tpop[k]-lpop[k];
15969: }
15970:
15971:
15972: printf("\nAge lx qx dx Lx Tx e(x)\n");
15973: for (k=agegomp;k<(agemortsup-2);k++)
15974: printf("%d %.0lf %lf %.0lf %.0lf %.0lf %lf\n",k,lsurv[k],p[1]*exp(p[2]*(k-agegomp)),(p[1]*exp(p[2]*(k-agegomp)))*lsurv[k],lpop[k],tpop[k],tpop[k]/lsurv[k]);
15975:
15976:
15977: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 15978: ageminpar=50;
15979: agemaxpar=100;
1.194 brouard 15980: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
15981: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
15982: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
15983: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
15984: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
15985: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
15986: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 15987: }else{
15988: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
15989: fprintf(ficlog,"Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
1.201 brouard 15990: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 15991: }
1.201 brouard 15992: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 15993: stepm, weightopt,\
15994: model,imx,p,matcov,agemortsup);
15995:
1.302 brouard 15996: free_vector(lsurv,agegomp,AGESUP);
15997: free_vector(lpop,agegomp,AGESUP);
15998: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 15999: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 16000: free_ivector(dcwave,firstobs,lastobs);
16001: free_vector(agecens,firstobs,lastobs);
16002: free_vector(ageexmed,firstobs,lastobs);
16003: free_ivector(cens,firstobs,lastobs);
1.220 brouard 16004: #ifdef GSL
1.136 brouard 16005: #endif
1.186 brouard 16006: } /* Endof if mle==-3 mortality only */
1.205 brouard 16007: /* Standard */
16008: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
16009: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
16010: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 16011: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 16012: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
16013: for (k=1; k<=npar;k++)
16014: printf(" %d %8.5f",k,p[k]);
16015: printf("\n");
1.205 brouard 16016: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
16017: /* mlikeli uses func not funcone */
1.247 brouard 16018: /* for(i=1;i<nlstate;i++){ */
16019: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
16020: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
16021: /* } */
1.205 brouard 16022: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
16023: }
16024: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
16025: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
16026: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
16027: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
16028: }
16029: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 16030: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
16031: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
1.335 brouard 16032: /* exit(0); */
1.126 brouard 16033: for (k=1; k<=npar;k++)
16034: printf(" %d %8.5f",k,p[k]);
16035: printf("\n");
16036:
16037: /*--------- results files --------------*/
1.283 brouard 16038: /* fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\nftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle= 0 weight=%d\nmodel=1+age+%s.\n", title, datafile, lastobs, firstpass,lastpass,ftol, stepm, ncovcol, nqv, ntv, nqtv, nlstate, ndeath, maxwav, weightopt,model); */
1.126 brouard 16039:
16040:
16041: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 16042: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 16043: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 16044:
16045: printf("#model= 1 + age ");
16046: fprintf(ficres,"#model= 1 + age ");
16047: fprintf(ficlog,"#model= 1 + age ");
16048: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
16049: </ul>", model);
16050:
16051: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
16052: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
16053: if(nagesqr==1){
16054: printf(" + age*age ");
16055: fprintf(ficres," + age*age ");
16056: fprintf(ficlog," + age*age ");
16057: fprintf(fichtm, "<th>+ age*age</th>");
16058: }
1.362 brouard 16059: for(j=1;j <=ncovmodel-2-nagesqr;j++){
1.319 brouard 16060: if(Typevar[j]==0) {
16061: printf(" + V%d ",Tvar[j]);
16062: fprintf(ficres," + V%d ",Tvar[j]);
16063: fprintf(ficlog," + V%d ",Tvar[j]);
16064: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
16065: }else if(Typevar[j]==1) {
16066: printf(" + V%d*age ",Tvar[j]);
16067: fprintf(ficres," + V%d*age ",Tvar[j]);
16068: fprintf(ficlog," + V%d*age ",Tvar[j]);
16069: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
16070: }else if(Typevar[j]==2) {
16071: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
16072: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
16073: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
16074: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.349 brouard 16075: }else if(Typevar[j]==3) { /* TO VERIFY */
16076: printf(" + V%d*V%d*age ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
16077: fprintf(ficres," + V%d*V%d*age ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
16078: fprintf(ficlog," + V%d*V%d*age ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
16079: fprintf(fichtm, "<th>+ V%d*V%d*age</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.319 brouard 16080: }
16081: }
16082: printf("\n");
16083: fprintf(ficres,"\n");
16084: fprintf(ficlog,"\n");
16085: fprintf(fichtm, "</tr>");
16086: fprintf(fichtm, "\n");
16087:
16088:
1.126 brouard 16089: for(i=1,jk=1; i <=nlstate; i++){
16090: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 16091: if (k != i) {
1.319 brouard 16092: fprintf(fichtm, "<tr>");
1.225 brouard 16093: printf("%d%d ",i,k);
16094: fprintf(ficlog,"%d%d ",i,k);
16095: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 16096: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 16097: for(j=1; j <=ncovmodel; j++){
16098: printf("%12.7f ",p[jk]);
16099: fprintf(ficlog,"%12.7f ",p[jk]);
16100: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 16101: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 16102: jk++;
16103: }
16104: printf("\n");
16105: fprintf(ficlog,"\n");
16106: fprintf(ficres,"\n");
1.319 brouard 16107: fprintf(fichtm, "</tr>\n");
1.225 brouard 16108: }
1.126 brouard 16109: }
16110: }
1.319 brouard 16111: /* fprintf(fichtm,"</tr>\n"); */
16112: fprintf(fichtm,"</table>\n");
16113: fprintf(fichtm, "\n");
16114:
1.203 brouard 16115: if(mle != 0){
16116: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 16117: ftolhess=ftol; /* Usually correct */
1.203 brouard 16118: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
16119: printf("Parameters and 95%% confidence intervals\n W is simply the result of the division of the parameter by the square root of covariance of the parameter.\n And Wald-based confidence intervals plus and minus 1.96 * W .\n But be careful that parameters are highly correlated because incidence of disability is highly correlated to incidence of recovery.\n It might be better to visualize the covariance matrix. See the page 'Matrix of variance-covariance of one-step probabilities' and its graphs.\n");
16120: fprintf(ficlog, "Parameters, Wald tests and Wald-based confidence intervals\n W is simply the result of the division of the parameter by the square root of covariance of the parameter.\n And Wald-based confidence intervals plus and minus 1.96 * W \n It might be better to visualize the covariance matrix. See the page 'Matrix of variance-covariance of one-step probabilities' and its graphs.\n");
1.322 brouard 16121: fprintf(fichtm, "\n<p>The Wald test results are output only if the maximimzation of the Likelihood is performed (mle=1)\n</br>Parameters, Wald tests and Wald-based confidence intervals\n</br> W is simply the result of the division of the parameter by the square root of covariance of the parameter.\n</br> And Wald-based confidence intervals plus and minus 1.96 * W \n </br> It might be better to visualize the covariance matrix. See the page '<a href=\"%s\">Matrix of variance-covariance of one-step probabilities and its graphs</a>'.\n</br>",optionfilehtmcov);
1.319 brouard 16122: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
16123: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
16124: if(nagesqr==1){
16125: printf(" + age*age ");
16126: fprintf(ficres," + age*age ");
16127: fprintf(ficlog," + age*age ");
16128: fprintf(fichtm, "<th>+ age*age</th>");
16129: }
1.362 brouard 16130: for(j=1;j <=ncovmodel-2-nagesqr;j++){
1.319 brouard 16131: if(Typevar[j]==0) {
16132: printf(" + V%d ",Tvar[j]);
16133: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
16134: }else if(Typevar[j]==1) {
16135: printf(" + V%d*age ",Tvar[j]);
16136: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
16137: }else if(Typevar[j]==2) {
16138: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.349 brouard 16139: }else if(Typevar[j]==3) { /* TO VERIFY */
16140: fprintf(fichtm, "<th>+ V%d*V%d*age</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.319 brouard 16141: }
16142: }
16143: fprintf(fichtm, "</tr>\n");
16144:
1.203 brouard 16145: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 16146: for(k=1; k <=(nlstate+ndeath); k++){
16147: if (k != i) {
1.319 brouard 16148: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 16149: printf("%d%d ",i,k);
16150: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 16151: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 16152: for(j=1; j <=ncovmodel; j++){
1.319 brouard 16153: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 16154: printf("%12.7f(%12.7f) W=%8.3f CI=[%12.7f ; %12.7f] ",p[jk],sqrt(matcov[jk][jk]), p[jk]/sqrt(matcov[jk][jk]), p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
16155: fprintf(ficlog,"%12.7f(%12.7f) W=%8.3f CI=[%12.7f ; %12.7f] ",p[jk],sqrt(matcov[jk][jk]), p[jk]/sqrt(matcov[jk][jk]), p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
1.319 brouard 16156: if(fabs(wald) > 1.96){
1.321 brouard 16157: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 16158: }else{
16159: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
16160: }
1.324 brouard 16161: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 16162: fprintf(fichtm,"[%12.7f;%12.7f]</br></td>", p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
1.225 brouard 16163: jk++;
16164: }
16165: printf("\n");
16166: fprintf(ficlog,"\n");
1.319 brouard 16167: fprintf(fichtm, "</tr>\n");
1.225 brouard 16168: }
16169: }
1.193 brouard 16170: }
1.203 brouard 16171: } /* end of hesscov and Wald tests */
1.319 brouard 16172: fprintf(fichtm,"</table>\n");
1.225 brouard 16173:
1.203 brouard 16174: /* */
1.126 brouard 16175: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
16176: printf("# Scales (for hessian or gradient estimation)\n");
16177: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
16178: for(i=1,jk=1; i <=nlstate; i++){
16179: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 16180: if (j!=i) {
16181: fprintf(ficres,"%1d%1d",i,j);
16182: printf("%1d%1d",i,j);
16183: fprintf(ficlog,"%1d%1d",i,j);
16184: for(k=1; k<=ncovmodel;k++){
16185: printf(" %.5e",delti[jk]);
16186: fprintf(ficlog," %.5e",delti[jk]);
16187: fprintf(ficres," %.5e",delti[jk]);
16188: jk++;
16189: }
16190: printf("\n");
16191: fprintf(ficlog,"\n");
16192: fprintf(ficres,"\n");
16193: }
1.126 brouard 16194: }
16195: }
16196:
16197: fprintf(ficres,"# Covariance matrix \n# 121 Var(a12)\n# 122 Cov(b12,a12) Var(b12)\n# ...\n# 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n");
1.349 brouard 16198: if(mle >= 1) /* Too big for the screen */
1.126 brouard 16199: printf("# Covariance matrix \n# 121 Var(a12)\n# 122 Cov(b12,a12) Var(b12)\n# ...\n# 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n");
16200: fprintf(ficlog,"# Covariance matrix \n# 121 Var(a12)\n# 122 Cov(b12,a12) Var(b12)\n# ...\n# 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n");
16201: /* # 121 Var(a12)\n\ */
16202: /* # 122 Cov(b12,a12) Var(b12)\n\ */
16203: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
16204: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
16205: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
16206: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
16207: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
16208: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
16209:
16210:
16211: /* Just to have a covariance matrix which will be more understandable
16212: even is we still don't want to manage dictionary of variables
16213: */
16214: for(itimes=1;itimes<=2;itimes++){
16215: jj=0;
16216: for(i=1; i <=nlstate; i++){
1.225 brouard 16217: for(j=1; j <=nlstate+ndeath; j++){
16218: if(j==i) continue;
16219: for(k=1; k<=ncovmodel;k++){
16220: jj++;
16221: ca[0]= k+'a'-1;ca[1]='\0';
16222: if(itimes==1){
16223: if(mle>=1)
16224: printf("#%1d%1d%d",i,j,k);
16225: fprintf(ficlog,"#%1d%1d%d",i,j,k);
16226: fprintf(ficres,"#%1d%1d%d",i,j,k);
16227: }else{
16228: if(mle>=1)
16229: printf("%1d%1d%d",i,j,k);
16230: fprintf(ficlog,"%1d%1d%d",i,j,k);
16231: fprintf(ficres,"%1d%1d%d",i,j,k);
16232: }
16233: ll=0;
16234: for(li=1;li <=nlstate; li++){
16235: for(lj=1;lj <=nlstate+ndeath; lj++){
16236: if(lj==li) continue;
16237: for(lk=1;lk<=ncovmodel;lk++){
16238: ll++;
16239: if(ll<=jj){
16240: cb[0]= lk +'a'-1;cb[1]='\0';
16241: if(ll<jj){
16242: if(itimes==1){
16243: if(mle>=1)
16244: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
16245: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
16246: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
16247: }else{
16248: if(mle>=1)
16249: printf(" %.5e",matcov[jj][ll]);
16250: fprintf(ficlog," %.5e",matcov[jj][ll]);
16251: fprintf(ficres," %.5e",matcov[jj][ll]);
16252: }
16253: }else{
16254: if(itimes==1){
16255: if(mle>=1)
16256: printf(" Var(%s%1d%1d)",ca,i,j);
16257: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
16258: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
16259: }else{
16260: if(mle>=1)
16261: printf(" %.7e",matcov[jj][ll]);
16262: fprintf(ficlog," %.7e",matcov[jj][ll]);
16263: fprintf(ficres," %.7e",matcov[jj][ll]);
16264: }
16265: }
16266: }
16267: } /* end lk */
16268: } /* end lj */
16269: } /* end li */
16270: if(mle>=1)
16271: printf("\n");
16272: fprintf(ficlog,"\n");
16273: fprintf(ficres,"\n");
16274: numlinepar++;
16275: } /* end k*/
16276: } /*end j */
1.126 brouard 16277: } /* end i */
16278: } /* end itimes */
16279:
16280: fflush(ficlog);
16281: fflush(ficres);
1.225 brouard 16282: while(fgets(line, MAXLINE, ficpar)) {
16283: /* If line starts with a # it is a comment */
16284: if (line[0] == '#') {
16285: numlinepar++;
16286: fputs(line,stdout);
16287: fputs(line,ficparo);
16288: fputs(line,ficlog);
1.299 brouard 16289: fputs(line,ficres);
1.225 brouard 16290: continue;
16291: }else
16292: break;
16293: }
16294:
1.209 brouard 16295: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
16296: /* ungetc(c,ficpar); */
16297: /* fgets(line, MAXLINE, ficpar); */
16298: /* fputs(line,stdout); */
16299: /* fputs(line,ficparo); */
16300: /* } */
16301: /* ungetc(c,ficpar); */
1.126 brouard 16302:
16303: estepm=0;
1.209 brouard 16304: if((num_filled=sscanf(line,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm, &ftolpl)) !=EOF){
1.225 brouard 16305:
16306: if (num_filled != 6) {
16307: printf("Error: Not 6 parameters in line, for example:agemin=60 agemax=95 bage=55 fage=95 estepm=24 ftolpl=6e-4\n, your line=%s . Probably you are running an older format.\n",line);
16308: fprintf(ficlog,"Error: Not 6 parameters in line, for example:agemin=60 agemax=95 bage=55 fage=95 estepm=24 ftolpl=6e-4\n, your line=%s . Probably you are running an older format.\n",line);
16309: goto end;
16310: }
16311: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
16312: }
16313: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
16314: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
16315:
1.209 brouard 16316: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 16317: if (estepm==0 || estepm < stepm) estepm=stepm;
16318: if (fage <= 2) {
16319: bage = ageminpar;
16320: fage = agemaxpar;
16321: }
16322:
16323: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 16324: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
16325: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 16326:
1.186 brouard 16327: /* Other stuffs, more or less useful */
1.254 brouard 16328: while(fgets(line, MAXLINE, ficpar)) {
16329: /* If line starts with a # it is a comment */
16330: if (line[0] == '#') {
16331: numlinepar++;
16332: fputs(line,stdout);
16333: fputs(line,ficparo);
16334: fputs(line,ficlog);
1.299 brouard 16335: fputs(line,ficres);
1.254 brouard 16336: continue;
16337: }else
16338: break;
16339: }
16340:
16341: if((num_filled=sscanf(line,"begin-prev-date=%lf/%lf/%lf end-prev-date=%lf/%lf/%lf mov_average=%d\n",&jprev1, &mprev1,&anprev1,&jprev2, &mprev2,&anprev2,&mobilav)) !=EOF){
16342:
16343: if (num_filled != 7) {
16344: printf("Error: Not 7 (data)parameters in line but %d, for example:begin-prev-date=1/1/1990 end-prev-date=1/6/2004 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
16345: fprintf(ficlog,"Error: Not 7 (data)parameters in line but %d, for example:begin-prev-date=1/1/1990 end-prev-date=1/6/2004 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
16346: goto end;
16347: }
16348: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
16349: fprintf(ficparo,"begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
16350: fprintf(ficres,"begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
16351: fprintf(ficlog,"begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
1.126 brouard 16352: }
1.254 brouard 16353:
16354: while(fgets(line, MAXLINE, ficpar)) {
16355: /* If line starts with a # it is a comment */
16356: if (line[0] == '#') {
16357: numlinepar++;
16358: fputs(line,stdout);
16359: fputs(line,ficparo);
16360: fputs(line,ficlog);
1.299 brouard 16361: fputs(line,ficres);
1.254 brouard 16362: continue;
16363: }else
16364: break;
1.126 brouard 16365: }
16366:
16367:
16368: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
16369: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
16370:
1.254 brouard 16371: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
16372: if (num_filled != 1) {
16373: printf("Error: Not 1 (data)parameters in line but %d, for example:pop_based=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
16374: fprintf(ficlog,"Error: Not 1 (data)parameters in line but %d, for example: pop_based=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
16375: goto end;
16376: }
16377: printf("pop_based=%d\n",popbased);
16378: fprintf(ficlog,"pop_based=%d\n",popbased);
16379: fprintf(ficparo,"pop_based=%d\n",popbased);
16380: fprintf(ficres,"pop_based=%d\n",popbased);
16381: }
16382:
1.258 brouard 16383: /* Results */
1.359 brouard 16384: /* Value of covariate in each resultine will be computed (if product) and sorted according to model rank */
1.332 brouard 16385: /* It is precov[] because we need the varying age in order to compute the real cov[] of the model equation */
16386: precov=matrix(1,MAXRESULTLINESPONE,1,NCOVMAX+1);
1.307 brouard 16387: endishere=0;
1.258 brouard 16388: nresult=0;
1.308 brouard 16389: parameterline=0;
1.258 brouard 16390: do{
16391: if(!fgets(line, MAXLINE, ficpar)){
16392: endishere=1;
1.308 brouard 16393: parameterline=15;
1.258 brouard 16394: }else if (line[0] == '#') {
16395: /* If line starts with a # it is a comment */
1.254 brouard 16396: numlinepar++;
16397: fputs(line,stdout);
16398: fputs(line,ficparo);
16399: fputs(line,ficlog);
1.299 brouard 16400: fputs(line,ficres);
1.254 brouard 16401: continue;
1.258 brouard 16402: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
16403: parameterline=11;
1.296 brouard 16404: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 16405: parameterline=12;
1.307 brouard 16406: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 16407: parameterline=13;
1.307 brouard 16408: }
1.258 brouard 16409: else{
16410: parameterline=14;
1.254 brouard 16411: }
1.308 brouard 16412: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 16413: case 11:
1.296 brouard 16414: if((num_filled=sscanf(line,"prevforecast=%d starting-proj-date=%lf/%lf/%lf final-proj-date=%lf/%lf/%lf mobil_average=%d\n",&prevfcast,&jproj1,&mproj1,&anproj1,&jproj2,&mproj2,&anproj2,&mobilavproj)) !=EOF && (num_filled == 8)){
16415: fprintf(ficparo,"prevforecast=%d starting-proj-date=%.lf/%.lf/%.lf final-proj-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevfcast,jproj1,mproj1,anproj1,jproj2,mproj2,anproj2,mobilavproj);
1.258 brouard 16416: printf("prevforecast=%d starting-proj-date=%.lf/%.lf/%.lf final-proj-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevfcast,jproj1,mproj1,anproj1,jproj2,mproj2,anproj2,mobilavproj);
16417: fprintf(ficlog,"prevforecast=%d starting-proj-date=%.lf/%.lf/%.lf final-proj-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevfcast,jproj1,mproj1,anproj1,jproj2,mproj2,anproj2,mobilavproj);
16418: fprintf(ficres,"prevforecast=%d starting-proj-date=%.lf/%.lf/%.lf final-proj-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevfcast,jproj1,mproj1,anproj1,jproj2,mproj2,anproj2,mobilavproj);
16419: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 16420: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
16421: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 16422: prvforecast = 1;
16423: }
16424: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 16425: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
16426: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
16427: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 16428: prvforecast = 2;
16429: }
16430: else {
16431: printf("Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevforecast=1 yearsfproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
16432: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevforecast=1 yearproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
16433: goto end;
1.258 brouard 16434: }
1.254 brouard 16435: break;
1.258 brouard 16436: case 12:
1.296 brouard 16437: if((num_filled=sscanf(line,"prevbackcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&prevbcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj)) !=EOF && (num_filled == 8)){
16438: fprintf(ficparo,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
16439: printf("prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
16440: fprintf(ficlog,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
16441: fprintf(ficres,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
16442: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 16443: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
16444: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 16445: prvbackcast = 1;
16446: }
16447: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 16448: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
16449: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
16450: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 16451: prvbackcast = 2;
16452: }
16453: else {
16454: printf("Error: Not 8 (data)parameters in line but %d, for example:prevbackcast=1 starting-back-date=1/1/1990 final-back-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevbackcast=1 yearsbproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
16455: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:prevbackcast=1 starting-back-date=1/1/1990 final-back-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevbackcast=1 yearbproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
16456: goto end;
1.258 brouard 16457: }
1.230 brouard 16458: break;
1.258 brouard 16459: case 13:
1.332 brouard 16460: num_filled=sscanf(line,"result:%[^\n]\n",resultlineori);
1.307 brouard 16461: nresult++; /* Sum of resultlines */
1.342 brouard 16462: /* printf("Result %d: result:%s\n",nresult, resultlineori); */
1.332 brouard 16463: /* removefirstspace(&resultlineori); */
16464:
16465: if(strstr(resultlineori,"v") !=0){
16466: printf("Error. 'v' must be in upper case 'V' result: %s ",resultlineori);
16467: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultlineori);fflush(ficlog);
16468: return 1;
16469: }
16470: trimbb(resultline, resultlineori); /* Suppressing double blank in the resultline */
1.342 brouard 16471: /* printf("Decoderesult resultline=\"%s\" resultlineori=\"%s\"\n", resultline, resultlineori); */
1.318 brouard 16472: if(nresult > MAXRESULTLINESPONE-1){
16473: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\nYou can use the 'r' parameter file '%s' which uses option mle=0 to get other results. ",MAXRESULTLINESPONE-1,nresult,rfileres);
16474: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\nYou can use the 'r' parameter file '%s' which uses option mle=0 to get other results. ",MAXRESULTLINESPONE-1,nresult,rfileres);
1.307 brouard 16475: goto end;
16476: }
1.332 brouard 16477:
1.310 brouard 16478: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 16479: fprintf(ficparo,"result: %s\n",resultline);
16480: fprintf(ficres,"result: %s\n",resultline);
16481: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 16482: } else
16483: goto end;
1.307 brouard 16484: break;
16485: case 14:
16486: printf("Error: Unknown command '%s'\n",line);
16487: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 16488: if(line[0] == ' ' || line[0] == '\n'){
16489: printf("It should not be an empty line '%s'\n",line);
16490: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
16491: }
1.307 brouard 16492: if(ncovmodel >=2 && nresult==0 ){
16493: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
16494: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 16495: }
1.307 brouard 16496: /* goto end; */
16497: break;
1.308 brouard 16498: case 15:
16499: printf("End of resultlines.\n");
16500: fprintf(ficlog,"End of resultlines.\n");
16501: break;
16502: default: /* parameterline =0 */
1.307 brouard 16503: nresult=1;
16504: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 16505: } /* End switch parameterline */
16506: }while(endishere==0); /* End do */
1.126 brouard 16507:
1.230 brouard 16508: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 16509: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 16510:
16511: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 16512: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 16513: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 16514: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
16515: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 16516: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 16517: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
16518: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 16519: }else{
1.270 brouard 16520: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 16521: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
16522: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
16523: if(prvforecast==1){
16524: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
16525: jprojd=jproj1;
16526: mprojd=mproj1;
16527: anprojd=anproj1;
16528: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
16529: jprojf=jproj2;
16530: mprojf=mproj2;
16531: anprojf=anproj2;
16532: } else if(prvforecast == 2){
16533: dateprojd=dateintmean;
16534: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
16535: dateprojf=dateintmean+yrfproj;
16536: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
16537: }
16538: if(prvbackcast==1){
16539: datebackd=(jback1+12*mback1+365*anback1)/365;
16540: jbackd=jback1;
16541: mbackd=mback1;
16542: anbackd=anback1;
16543: datebackf=(jback2+12*mback2+365*anback2)/365;
16544: jbackf=jback2;
16545: mbackf=mback2;
16546: anbackf=anback2;
16547: } else if(prvbackcast == 2){
16548: datebackd=dateintmean;
16549: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
16550: datebackf=dateintmean-yrbproj;
16551: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
16552: }
16553:
1.350 brouard 16554: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);/* HERE valgrind Tvard*/
1.220 brouard 16555: }
16556: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 16557: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
16558: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 16559:
1.225 brouard 16560: /*------------ free_vector -------------*/
16561: /* chdir(path); */
1.220 brouard 16562:
1.215 brouard 16563: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
16564: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
16565: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
16566: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 16567: free_lvector(num,firstobs,lastobs);
16568: free_vector(agedc,firstobs,lastobs);
1.126 brouard 16569: /*free_matrix(covar,0,NCOVMAX,1,n);*/
16570: /*free_matrix(covar,1,NCOVMAX,1,n);*/
16571: fclose(ficparo);
16572: fclose(ficres);
1.220 brouard 16573:
16574:
1.186 brouard 16575: /* Other results (useful)*/
1.220 brouard 16576:
16577:
1.126 brouard 16578: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 16579: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
16580: prlim=matrix(1,nlstate,1,nlstate);
1.332 brouard 16581: /* Computes the prevalence limit for each combination k of the dummy covariates by calling prevalim(k) */
1.209 brouard 16582: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 16583: fclose(ficrespl);
16584:
16585: /*------------- h Pij x at various ages ------------*/
1.180 brouard 16586: /*#include "hpijx.h"*/
1.332 brouard 16587: /** h Pij x Probability to be in state j at age x+h being in i at x, for each combination k of dummies in the model line or to nres?*/
16588: /* calls hpxij with combination k */
1.180 brouard 16589: hPijx(p, bage, fage);
1.145 brouard 16590: fclose(ficrespij);
1.227 brouard 16591:
1.220 brouard 16592: /* ncovcombmax= pow(2,cptcoveff); */
1.332 brouard 16593: /*-------------- Variance of one-step probabilities for a combination ij or for nres ?---*/
1.145 brouard 16594: k=1;
1.126 brouard 16595: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 16596:
1.269 brouard 16597: /* Prevalence for each covariate combination in probs[age][status][cov] */
16598: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
16599: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 16600: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 16601: for(k=1;k<=ncovcombmax;k++)
16602: probs[i][j][k]=0.;
1.269 brouard 16603: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
16604: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 16605: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 16606: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
16607: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 16608: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 16609: for(k=1;k<=ncovcombmax;k++)
16610: mobaverages[i][j][k]=0.;
1.219 brouard 16611: mobaverage=mobaverages;
16612: if (mobilav!=0) {
1.235 brouard 16613: printf("Movingaveraging observed prevalence\n");
1.258 brouard 16614: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 16615: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
16616: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
16617: printf(" Error in movingaverage mobilav=%d\n",mobilav);
16618: }
1.269 brouard 16619: } else if (mobilavproj !=0) {
1.235 brouard 16620: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 16621: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 16622: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
16623: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
16624: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
16625: }
1.269 brouard 16626: }else{
16627: printf("Internal error moving average\n");
16628: fflush(stdout);
16629: exit(1);
1.219 brouard 16630: }
16631: }/* end if moving average */
1.227 brouard 16632:
1.126 brouard 16633: /*---------- Forecasting ------------------*/
1.296 brouard 16634: if(prevfcast==1){
16635: /* /\* if(stepm ==1){*\/ */
16636: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
16637: /*This done previously after freqsummary.*/
16638: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
16639: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
16640:
16641: /* } else if (prvforecast==2){ */
16642: /* /\* if(stepm ==1){*\/ */
16643: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
16644: /* } */
16645: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
16646: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 16647: }
1.269 brouard 16648:
1.296 brouard 16649: /* Prevbcasting */
16650: if(prevbcast==1){
1.219 brouard 16651: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
16652: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
16653: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
16654:
16655: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
16656:
16657: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 16658:
1.219 brouard 16659: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
16660: fclose(ficresplb);
16661:
1.222 brouard 16662: hBijx(p, bage, fage, mobaverage);
16663: fclose(ficrespijb);
1.219 brouard 16664:
1.296 brouard 16665: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
16666: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
16667: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
16668: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
16669: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
16670: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
16671:
16672:
1.269 brouard 16673: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 16674:
16675:
1.269 brouard 16676: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 16677: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
16678: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
16679: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 16680: } /* end Prevbcasting */
1.268 brouard 16681:
1.186 brouard 16682:
16683: /* ------ Other prevalence ratios------------ */
1.126 brouard 16684:
1.215 brouard 16685: free_ivector(wav,1,imx);
16686: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
16687: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
16688: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 16689:
16690:
1.127 brouard 16691: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 16692:
1.201 brouard 16693: strcpy(filerese,"E_");
16694: strcat(filerese,fileresu);
1.126 brouard 16695: if((ficreseij=fopen(filerese,"w"))==NULL) {
16696: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
16697: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
16698: }
1.208 brouard 16699: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
16700: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 16701:
16702: pstamp(ficreseij);
1.219 brouard 16703:
1.351 brouard 16704: /* i1=pow(2,cptcoveff); /\* Number of combination of dummy covariates *\/ */
16705: /* if (cptcovn < 1){i1=1;} */
1.235 brouard 16706:
1.351 brouard 16707: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
16708: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
16709: /* if(i1 != 1 && TKresult[nres]!= k) */
16710: /* continue; */
1.219 brouard 16711: fprintf(ficreseij,"\n#****** ");
1.235 brouard 16712: printf("\n#****** ");
1.351 brouard 16713: for(j=1;j<=cptcovs;j++){
16714: /* for(j=1;j<=cptcoveff;j++) { */
16715: /* fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
16716: fprintf(ficreseij," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
16717: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
16718: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.235 brouard 16719: }
16720: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.337 brouard 16721: printf(" V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]); /* TvarsQ[j] gives the name of the jth quantitative (fixed or time v) */
16722: fprintf(ficreseij,"V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]);
1.219 brouard 16723: }
16724: fprintf(ficreseij,"******\n");
1.235 brouard 16725: printf("******\n");
1.219 brouard 16726:
16727: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
16728: oldm=oldms;savm=savms;
1.330 brouard 16729: /* printf("HELLO Entering evsij bage=%d fage=%d k=%d estepm=%d nres=%d\n",(int) bage, (int)fage, k, estepm, nres); */
1.235 brouard 16730: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 16731:
1.219 brouard 16732: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 16733: }
16734: fclose(ficreseij);
1.208 brouard 16735: printf("done evsij\n");fflush(stdout);
16736: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 16737:
1.218 brouard 16738:
1.227 brouard 16739: /*---------- State-specific expectancies and variances ------------*/
1.336 brouard 16740: /* Should be moved in a function */
1.201 brouard 16741: strcpy(filerest,"T_");
16742: strcat(filerest,fileresu);
1.127 brouard 16743: if((ficrest=fopen(filerest,"w"))==NULL) {
16744: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
16745: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
16746: }
1.208 brouard 16747: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
16748: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 16749: strcpy(fileresstde,"STDE_");
16750: strcat(fileresstde,fileresu);
1.126 brouard 16751: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 16752: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
16753: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 16754: }
1.227 brouard 16755: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
16756: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 16757:
1.201 brouard 16758: strcpy(filerescve,"CVE_");
16759: strcat(filerescve,fileresu);
1.126 brouard 16760: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 16761: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
16762: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 16763: }
1.227 brouard 16764: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
16765: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 16766:
1.201 brouard 16767: strcpy(fileresv,"V_");
16768: strcat(fileresv,fileresu);
1.126 brouard 16769: if((ficresvij=fopen(fileresv,"w"))==NULL) {
16770: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
16771: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
16772: }
1.227 brouard 16773: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
16774: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 16775:
1.235 brouard 16776: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
16777: if (cptcovn < 1){i1=1;}
16778:
1.334 brouard 16779: for(nres=1; nres <= nresult; nres++) /* For each resultline, find the combination and output results according to the values of dummies and then quanti. */
16780: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying. For each nres and each value at position k
16781: * we know Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline
16782: * Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline
16783: * and Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
16784: /* */
16785: if(i1 != 1 && TKresult[nres]!= k) /* TKresult[nres] is the combination of this nres resultline. All the i1 combinations are not output */
1.235 brouard 16786: continue;
1.359 brouard 16787: printf("\n# model=1+age+%s \n#****** Result for:", model); /* HERE model is empty */
16788: fprintf(ficrest,"\n# model=1+age+%s \n#****** Result for:", model);
16789: fprintf(ficlog,"\n# model=1+age+%s \n#****** Result for:", model);
1.334 brouard 16790: /* It might not be a good idea to mix dummies and quantitative */
16791: /* for(j=1;j<=cptcoveff;j++){ /\* j=resultpos. Could be a loop on cptcovs: number of single dummy covariate in the result line as well as in the model *\/ */
16792: for(j=1;j<=cptcovs;j++){ /* j=resultpos. Could be a loop on cptcovs: number of single covariate (dummy or quantitative) in the result line as well as in the model */
16793: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* Output by variables in the resultline *\/ */
16794: /* Tvaraff[j] is the name of the dummy variable in position j in the equation model:
16795: * Tvaraff[1]@9={4, 3, 0, 0, 0, 0, 0, 0, 0}, in model=V5+V4+V3+V4*V3+V5*age
16796: * (V5 is quanti) V4 and V3 are dummies
16797: * TnsdVar[4] is the position 1 and TnsdVar[3]=2 in codtabm(k,l)(V4 V3)=V4 V3
16798: * l=1 l=2
16799: * k=1 1 1 0 0
16800: * k=2 2 1 1 0
16801: * k=3 [1] [2] 0 1
16802: * k=4 2 2 1 1
16803: * If nres=1 result: V3=1 V4=0 then k=3 and outputs
16804: * If nres=2 result: V4=1 V3=0 then k=2 and outputs
16805: * nres=1 =>k=3 j=1 V4= nbcode[4][codtabm(3,1)=1)=0; j=2 V3= nbcode[3][codtabm(3,2)=2]=1
16806: * nres=2 =>k=2 j=1 V4= nbcode[4][codtabm(2,1)=2)=1; j=2 V3= nbcode[3][codtabm(2,2)=1]=0
16807: */
16808: /* Tvresult[nres][j] Name of the variable at position j in this resultline */
16809: /* Tresult[nres][j] Value of this variable at position j could be a float if quantitative */
16810: /* We give up with the combinations!! */
1.342 brouard 16811: /* if(debugILK) */
16812: /* printf("\n j=%d In computing T_ Dummy[modelresult[%d][%d]]=%d, modelresult[%d][%d]=%d cptcovs=%d, cptcoveff=%d Fixed[modelresult[nres][j]]=%d\n", j, nres, j, Dummy[modelresult[nres][j]],nres,j,modelresult[nres][j],cptcovs, cptcoveff,Fixed[modelresult[nres][j]]); /\* end if dummy or quanti *\/ */
1.334 brouard 16813:
16814: if(Dummy[modelresult[nres][j]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to j in resultline */
1.344 brouard 16815: /* printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][j]); /\* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline *\/ */ /* TinvDoQresult[nres][Name of the variable] */
16816: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); /* Output of each value for the combination TKresult[nres], ordered by the covariate values in the resultline */
16817: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
16818: fprintf(ficrest,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
1.334 brouard 16819: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
16820: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
16821: }else{
16822: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
16823: }
16824: /* fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
16825: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
16826: }else if(Dummy[modelresult[nres][j]]==1){ /* Quanti variable */
16827: /* For each selected (single) quantitative value */
1.337 brouard 16828: printf(" V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
16829: fprintf(ficlog," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
16830: fprintf(ficrest," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
1.334 brouard 16831: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
16832: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
16833: }else{
16834: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
16835: }
16836: }else{
16837: printf("Error in computing T_ Dummy[modelresult[%d][%d]]=%d, modelresult[%d][%d]=%d cptcovs=%d, cptcoveff=%d \n", nres, j, Dummy[modelresult[nres][j]],nres,j,modelresult[nres][j],cptcovs, cptcoveff); /* end if dummy or quanti */
16838: fprintf(ficlog,"Error in computing T_ Dummy[modelresult[%d][%d]]=%d, modelresult[%d][%d]=%d cptcovs=%d, cptcoveff=%d \n", nres, j, Dummy[modelresult[nres][j]],nres,j,modelresult[nres][j],cptcovs, cptcoveff); /* end if dummy or quanti */
16839: exit(1);
16840: }
1.335 brouard 16841: } /* End loop for each variable in the resultline */
1.334 brouard 16842: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
16843: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /\* Wrong j is not in the equation model *\/ */
16844: /* fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
16845: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
16846: /* } */
1.208 brouard 16847: fprintf(ficrest,"******\n");
1.227 brouard 16848: fprintf(ficlog,"******\n");
16849: printf("******\n");
1.208 brouard 16850:
16851: fprintf(ficresstdeij,"\n#****** ");
16852: fprintf(ficrescveij,"\n#****** ");
1.337 brouard 16853: /* It could have been: for(j=1;j<=cptcoveff;j++) {printf("V=%d=%lg",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);} */
16854: /* But it won't be sorted and depends on how the resultline is ordered */
1.225 brouard 16855: for(j=1;j<=cptcoveff;j++) {
1.334 brouard 16856: fprintf(ficresstdeij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
16857: /* fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
16858: /* fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
16859: }
16860: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value, TvarsQind gives the position of a quantitative in model equation */
1.337 brouard 16861: fprintf(ficresstdeij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
16862: fprintf(ficrescveij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
1.235 brouard 16863: }
1.208 brouard 16864: fprintf(ficresstdeij,"******\n");
16865: fprintf(ficrescveij,"******\n");
16866:
16867: fprintf(ficresvij,"\n#****** ");
1.238 brouard 16868: /* pstamp(ficresvij); */
1.225 brouard 16869: for(j=1;j<=cptcoveff;j++)
1.335 brouard 16870: fprintf(ficresvij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
16871: /* fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[TnsdVar[Tvaraff[j]]])]); */
1.235 brouard 16872: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 16873: /* fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]); /\* To solve *\/ */
1.337 brouard 16874: fprintf(ficresvij," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /* Solved */
1.235 brouard 16875: }
1.208 brouard 16876: fprintf(ficresvij,"******\n");
16877:
16878: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
16879: oldm=oldms;savm=savms;
1.235 brouard 16880: printf(" cvevsij ");
16881: fprintf(ficlog, " cvevsij ");
16882: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 16883: printf(" end cvevsij \n ");
16884: fprintf(ficlog, " end cvevsij \n ");
16885:
16886: /*
16887: */
16888: /* goto endfree; */
16889:
16890: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
16891: pstamp(ficrest);
16892:
1.269 brouard 16893: epj=vector(1,nlstate+1);
1.208 brouard 16894: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 16895: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
16896: cptcod= 0; /* To be deleted */
1.360 brouard 16897: printf("varevsij vpopbased=%d popbased=%d \n",vpopbased,popbased);
16898: fprintf(ficlog, "varevsij vpopbased=%d popbased=%d \n",vpopbased,popbased);
1.361 brouard 16899: /* Call to varevsij to get cov(e.i, e.j)= vareij[i][j][(int)age]=sum_h sum_k trgrad(h_p.i) V(theta) grad(k_p.k) Equation 20 */
16900: /* Depending of popbased which changes the prevalences, either cross-sectional or period */
1.235 brouard 16901: varevsij(optionfilefiname, vareij, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, estepm, cptcov,cptcod,vpopbased,mobilav, strstart, nres); /* cptcod not initialized Intel */
1.360 brouard 16902: fprintf(ficrest,"# Total life expectancy with std error and decomposition into time to be expected in each state\n\
16903: # (these are weighted average of eij where weights are ");
1.227 brouard 16904: if(vpopbased==1)
1.360 brouard 16905: fprintf(ficrest,"the age specific prevalence observed (cross-sectionally) in the population i.e cross-sectionally)\n in each health state (popbased=1) (mobilav=%d)\n",mobilav);
1.227 brouard 16906: else
1.360 brouard 16907: fprintf(ficrest,"the age specific forward period (stable) prevalences in each state) \n");
16908: fprintf(ficrest,"# with proportions of time spent in each state with standard error (on the right of the table.\n ");
1.335 brouard 16909: fprintf(ficrest,"# Age popbased mobilav e.. (std) "); /* Adding covariate values? */
1.227 brouard 16910: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
1.360 brouard 16911: for (i=1;i<=nlstate;i++) fprintf(ficrest," %% e.%d/e.. (std) ",i);
1.227 brouard 16912: fprintf(ficrest,"\n");
16913: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 16914: printf("Computing age specific forward period (stable) prevalences in each health state \n");
16915: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 16916: for(age=bage; age <=fage ;age++){
1.235 brouard 16917: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 16918: if (vpopbased==1) {
16919: if(mobilav ==0){
16920: for(i=1; i<=nlstate;i++)
16921: prlim[i][i]=probs[(int)age][i][k];
16922: }else{ /* mobilav */
16923: for(i=1; i<=nlstate;i++)
16924: prlim[i][i]=mobaverage[(int)age][i][k];
16925: }
16926: }
1.219 brouard 16927:
1.227 brouard 16928: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
16929: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
16930: /* printf(" age %4.0f ",age); */
16931: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
16932: for(i=1, epj[j]=0.;i <=nlstate;i++) {
16933: epj[j] += prlim[i][i]*eij[i][j][(int)age];
16934: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
16935: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
16936: }
1.361 brouard 16937: epj[nlstate+1] +=epj[j]; /* epp=sum_j epj = sum_j sum_i w_i e_ij */
1.227 brouard 16938: }
16939: /* printf(" age %4.0f \n",age); */
1.219 brouard 16940:
1.361 brouard 16941: for(i=1, vepp=0.;i <=nlstate;i++) /* Variance of total life expectancy e.. */
1.227 brouard 16942: for(j=1;j <=nlstate;j++)
1.361 brouard 16943: vepp += vareij[i][j][(int)age]; /* sum_i sum_j cov(e.i, e.j) = var(e..) */
1.227 brouard 16944: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
1.361 brouard 16945: /* vareij[i][j] is the covariance cov(e.i, e.j) and vareij[j][j] is the variance of e.j */
1.227 brouard 16946: for(j=1;j <=nlstate;j++){
16947: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
16948: }
1.360 brouard 16949: /* And proportion of time spent in state j */
16950: /* $$ E[r(X,Y)-E(r(X,Y))]^2=[\frac{1}{\mu_y} -\frac{\mu_x}{{\mu_y}^2}]' Var(X,Y)[\frac{1}{\mu_y} -\frac{\mu_x}{{\mu_y}^2}]$$ */
1.361 brouard 16951: /* \frac{\mu_x^2}{\mu_y^2} ( \frac{\sigma^2_x}{\mu_x^2}-2\frac{\sigma_{xy}}{\mu_x\mu_y} +\frac{\sigma^2_y}{\mu_y^2}) */
16952: /* \frac{e_{.i}^2}{e_{..}^2} ( \frac{\Var e_{.i}}{e_{.i}^2}-2\frac{\Var e_{.i} + \sum_{j\ne i} \Cov e_{.j},e_{.i}}{e_{.i}e_{..}} +\frac{\Var e_{..}}{e_{..}^2})*/
16953: /*\mu_x = epj[j], \sigma^2_x = vareij[j][j][(int)age] and \mu_y=epj[nlstate+1], \sigma^2_y=vepp \sigmaxy= */
16954: /* vareij[j][j][(int)age]/epj[nlstate+1]^2 + vepp/epj[nlstate+1]^4 */
1.360 brouard 16955: for(j=1;j <=nlstate;j++){
16956: /* fprintf(ficrest," %7.3f (%7.3f)", epj[j]/epj[nlstate+1], sqrt( vareij[j][j][(int)age]/epj[j]/epj[j] + vepp/epj[j]/epj[j]/epj[j]/epj[j] )); */
1.361 brouard 16957: /* fprintf(ficrest," %7.3f (%7.3f)", epj[j]/epj[nlstate+1], sqrt( vareij[j][j][(int)age]/epj[j]/epj[j] + vepp/epj[j]/epj[j]/epj[j]/epj[j] )); */
16958:
16959: for(i=1,stdpercent=0.;i<=nlstate;i++){ /* Computing cov(e..,e.j)=cov(sum_i e.i,e.j)=sum_i cov(e.i, e.j) */
16960: stdpercent += vareij[i][j][(int)age];
16961: }
16962: stdpercent= epj[j]*epj[j]/epj[nlstate+1]/epj[nlstate+1]* (vareij[j][j][(int)age]/epj[j]/epj[j]-2.*stdpercent/epj[j]/epj[nlstate+1]+ vepp/epj[nlstate+1]/epj[nlstate+1]);
16963: /* stdpercent= epj[j]*epj[j]/epj[nlstate+1]/epj[nlstate+1]*(vareij[j][j][(int)age]/epj[j]/epj[j] + vepp/epj[nlstate+1]/epj[nlstate+1]); */ /* Without covariance */
16964: /* fprintf(ficrest," %7.3f (%7.3f)", epj[j]/epj[nlstate+1], sqrt( vareij[j][j][(int)age]/epj[nlstate+1]/epj[nlstate+1] + epj[j]*epj[j]*vepp/epj[nlstate+1]/epj[nlstate+1]/epj[nlstate+1]/epj[nlstate+1] )); */
16965: fprintf(ficrest," %7.3f (%7.3f)", epj[j]/epj[nlstate+1], sqrt(stdpercent));
1.360 brouard 16966: }
1.227 brouard 16967: fprintf(ficrest,"\n");
16968: }
1.208 brouard 16969: } /* End vpopbased */
1.269 brouard 16970: free_vector(epj,1,nlstate+1);
1.208 brouard 16971: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
16972: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 16973: printf("done selection\n");fflush(stdout);
16974: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 16975:
1.335 brouard 16976: } /* End k selection or end covariate selection for nres */
1.227 brouard 16977:
16978: printf("done State-specific expectancies\n");fflush(stdout);
16979: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
16980:
1.335 brouard 16981: /* variance-covariance of forward period prevalence */
1.269 brouard 16982: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 16983:
1.227 brouard 16984:
1.290 brouard 16985: free_vector(weight,firstobs,lastobs);
1.351 brouard 16986: free_imatrix(Tvardk,0,NCOVMAX,1,2);
1.227 brouard 16987: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 16988: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
16989: free_matrix(anint,1,maxwav,firstobs,lastobs);
16990: free_matrix(mint,1,maxwav,firstobs,lastobs);
16991: free_ivector(cod,firstobs,lastobs);
1.227 brouard 16992: free_ivector(tab,1,NCOVMAX);
16993: fclose(ficresstdeij);
16994: fclose(ficrescveij);
16995: fclose(ficresvij);
16996: fclose(ficrest);
16997: fclose(ficpar);
16998:
16999:
1.126 brouard 17000: /*---------- End : free ----------------*/
1.219 brouard 17001: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 17002: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
17003: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 17004: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
17005: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 17006: } /* mle==-3 arrives here for freeing */
1.227 brouard 17007: /* endfree:*/
1.359 brouard 17008: if(mle!=-3) free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /* Could be elsewhere ?*/
1.227 brouard 17009: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
17010: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
17011: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.341 brouard 17012: /* if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs); */
17013: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs);
1.290 brouard 17014: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
17015: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
17016: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 17017: free_matrix(matcov,1,npar,1,npar);
17018: free_matrix(hess,1,npar,1,npar);
17019: /*free_vector(delti,1,npar);*/
17020: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
17021: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 17022: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 17023: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
17024:
17025: free_ivector(ncodemax,1,NCOVMAX);
17026: free_ivector(ncodemaxwundef,1,NCOVMAX);
17027: free_ivector(Dummy,-1,NCOVMAX);
17028: free_ivector(Fixed,-1,NCOVMAX);
1.349 brouard 17029: free_ivector(DummyV,-1,NCOVMAX);
17030: free_ivector(FixedV,-1,NCOVMAX);
1.227 brouard 17031: free_ivector(Typevar,-1,NCOVMAX);
17032: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 17033: free_ivector(TvarsQ,1,NCOVMAX);
17034: free_ivector(TvarsQind,1,NCOVMAX);
17035: free_ivector(TvarsD,1,NCOVMAX);
1.330 brouard 17036: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 17037: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 17038: free_ivector(TvarFD,1,NCOVMAX);
17039: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 17040: free_ivector(TvarF,1,NCOVMAX);
17041: free_ivector(TvarFind,1,NCOVMAX);
17042: free_ivector(TvarV,1,NCOVMAX);
17043: free_ivector(TvarVind,1,NCOVMAX);
17044: free_ivector(TvarA,1,NCOVMAX);
17045: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 17046: free_ivector(TvarFQ,1,NCOVMAX);
17047: free_ivector(TvarFQind,1,NCOVMAX);
17048: free_ivector(TvarVD,1,NCOVMAX);
17049: free_ivector(TvarVDind,1,NCOVMAX);
17050: free_ivector(TvarVQ,1,NCOVMAX);
17051: free_ivector(TvarVQind,1,NCOVMAX);
1.349 brouard 17052: free_ivector(TvarAVVA,1,NCOVMAX);
17053: free_ivector(TvarAVVAind,1,NCOVMAX);
17054: free_ivector(TvarVVA,1,NCOVMAX);
17055: free_ivector(TvarVVAind,1,NCOVMAX);
1.339 brouard 17056: free_ivector(TvarVV,1,NCOVMAX);
17057: free_ivector(TvarVVind,1,NCOVMAX);
17058:
1.230 brouard 17059: free_ivector(Tvarsel,1,NCOVMAX);
17060: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 17061: free_ivector(Tposprod,1,NCOVMAX);
17062: free_ivector(Tprod,1,NCOVMAX);
17063: free_ivector(Tvaraff,1,NCOVMAX);
1.338 brouard 17064: free_ivector(invalidvarcomb,0,ncovcombmax);
1.227 brouard 17065: free_ivector(Tage,1,NCOVMAX);
17066: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 17067: free_ivector(TmodelInvind,1,NCOVMAX);
17068: free_ivector(TmodelInvQind,1,NCOVMAX);
1.332 brouard 17069:
1.359 brouard 17070: /* free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /\* Could be elsewhere ?*\/ */
1.332 brouard 17071:
1.227 brouard 17072: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
17073: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 17074: fflush(fichtm);
17075: fflush(ficgp);
17076:
1.227 brouard 17077:
1.126 brouard 17078: if((nberr >0) || (nbwarn>0)){
1.216 brouard 17079: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
17080: fprintf(ficlog,"End of Imach with %d errors and/or warnings %d. Please look at the log file for details.\n",nberr,nbwarn);
1.126 brouard 17081: }else{
17082: printf("End of Imach\n");
17083: fprintf(ficlog,"End of Imach\n");
17084: }
17085: printf("See log file on %s\n",filelog);
17086: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 17087: /*(void) gettimeofday(&end_time,&tzp);*/
17088: rend_time = time(NULL);
17089: end_time = *localtime(&rend_time);
17090: /* tml = *localtime(&end_time.tm_sec); */
17091: strcpy(strtend,asctime(&end_time));
1.126 brouard 17092: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
17093: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 17094: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 17095:
1.157 brouard 17096: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
17097: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
17098: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 17099: /* printf("Total time was %d uSec.\n", total_usecs);*/
17100: /* if(fileappend(fichtm,optionfilehtm)){ */
17101: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
17102: fclose(fichtm);
17103: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
17104: fclose(fichtmcov);
17105: fclose(ficgp);
17106: fclose(ficlog);
17107: /*------ End -----------*/
1.227 brouard 17108:
1.281 brouard 17109:
17110: /* Executes gnuplot */
1.227 brouard 17111:
17112: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 17113: #ifdef WIN32
1.227 brouard 17114: if (_chdir(pathcd) != 0)
17115: printf("Can't move to directory %s!\n",path);
17116: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 17117: #else
1.227 brouard 17118: if(chdir(pathcd) != 0)
17119: printf("Can't move to directory %s!\n", path);
17120: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 17121: #endif
1.126 brouard 17122: printf("Current directory %s!\n",pathcd);
17123: /*strcat(plotcmd,CHARSEPARATOR);*/
17124: sprintf(plotcmd,"gnuplot");
1.157 brouard 17125: #ifdef _WIN32
1.126 brouard 17126: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
17127: #endif
17128: if(!stat(plotcmd,&info)){
1.158 brouard 17129: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 17130: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 17131: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 17132: }else
17133: strcpy(pplotcmd,plotcmd);
1.157 brouard 17134: #ifdef __unix
1.126 brouard 17135: strcpy(plotcmd,GNUPLOTPROGRAM);
17136: if(!stat(plotcmd,&info)){
1.158 brouard 17137: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 17138: }else
17139: strcpy(pplotcmd,plotcmd);
17140: #endif
17141: }else
17142: strcpy(pplotcmd,plotcmd);
17143:
17144: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 17145: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 17146: strcpy(pplotcmd,plotcmd);
1.227 brouard 17147:
1.126 brouard 17148: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 17149: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 17150: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 17151: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 17152: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 17153: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 17154: strcpy(plotcmd,pplotcmd);
17155: }
1.126 brouard 17156: }
1.158 brouard 17157: printf(" Successful, please wait...");
1.126 brouard 17158: while (z[0] != 'q') {
17159: /* chdir(path); */
1.154 brouard 17160: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 17161: scanf("%s",z);
17162: /* if (z[0] == 'c') system("./imach"); */
17163: if (z[0] == 'e') {
1.158 brouard 17164: #ifdef __APPLE__
1.152 brouard 17165: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 17166: #elif __linux
17167: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 17168: #else
1.152 brouard 17169: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 17170: #endif
17171: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
17172: system(pplotcmd);
1.126 brouard 17173: }
17174: else if (z[0] == 'g') system(plotcmd);
17175: else if (z[0] == 'q') exit(0);
17176: }
1.227 brouard 17177: end:
1.126 brouard 17178: while (z[0] != 'q') {
1.195 brouard 17179: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 17180: scanf("%s",z);
17181: }
1.283 brouard 17182: printf("End\n");
1.282 brouard 17183: exit(0);
1.126 brouard 17184: }
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>
![[BACK]](/icons/small/back.gif){kind=icon}
Return to imach.c CVS log ![[TXT]](/icons/small/text.gif){kind=icon}
![[DIR]](/icons/small/dir.gif){kind=icon}
Up to [Local Repository] / imach / src