Annotation of imach/src/imach.c, revision 1.360
1.360 ! brouard 1: /* $Id: imach.c,v 1.359 2024/04/24 21:21:17 brouard Exp $
1.126 brouard 2: $State: Exp $
1.360 ! brouard 3: $Log: imach.c,v $
! 4: Revision 1.359 2024/04/24 21:21:17 brouard
! 5: Summary: First IMaCh version using Brent Praxis software based on Buckhardt and Gegenfürtner C codes
! 6:
1.359 brouard 7: Revision 1.6 2024/04/24 21:10:29 brouard
8: Summary: First IMaCh version using Brent Praxis software based on Buckhardt and Gegenfürtner C codes
1.358 brouard 9:
1.359 brouard 10: Revision 1.5 2023/10/09 09:10:01 brouard
11: Summary: trying to reconsider
1.357 brouard 12:
1.359 brouard 13: Revision 1.4 2023/06/22 12:50:51 brouard
14: Summary: stil on going
1.357 brouard 15:
1.359 brouard 16: Revision 1.3 2023/06/22 11:28:07 brouard
17: *** empty log message ***
1.356 brouard 18:
1.359 brouard 19: Revision 1.2 2023/06/22 11:22:40 brouard
20: Summary: with svd but not working yet
1.355 brouard 21:
1.354 brouard 22: Revision 1.353 2023/05/08 18:48:22 brouard
23: *** empty log message ***
24:
1.353 brouard 25: Revision 1.352 2023/04/29 10:46:21 brouard
26: *** empty log message ***
27:
1.352 brouard 28: Revision 1.351 2023/04/29 10:43:47 brouard
29: Summary: 099r45
30:
1.351 brouard 31: Revision 1.350 2023/04/24 11:38:06 brouard
32: *** empty log message ***
33:
1.350 brouard 34: Revision 1.349 2023/01/31 09:19:37 brouard
35: Summary: Improvements in models with age*Vn*Vm
36:
1.348 brouard 37: Revision 1.347 2022/09/18 14:36:44 brouard
38: Summary: version 0.99r42
39:
1.347 brouard 40: Revision 1.346 2022/09/16 13:52:36 brouard
41: * src/imach.c (Module): 0.99r41 Was an error when product of timevarying and fixed. Using FixedV[of name] now. Thank you Feinuo
42:
1.346 brouard 43: Revision 1.345 2022/09/16 13:40:11 brouard
44: Summary: Version 0.99r41
45:
46: * imach.c (Module): 0.99r41 Was an error when product of timevarying and fixed. Using FixedV[of name] now. Thank you Feinuo
47:
1.345 brouard 48: Revision 1.344 2022/09/14 19:33:30 brouard
49: Summary: version 0.99r40
50:
51: * imach.c (Module): Fixing names of variables in T_ (thanks to Feinuo)
52:
1.344 brouard 53: Revision 1.343 2022/09/14 14:22:16 brouard
54: Summary: version 0.99r39
55:
56: * imach.c (Module): Version 0.99r39 with colored dummy covariates
57: (fixed or time varying), using new last columns of
58: ILK_parameter.txt file.
59:
1.343 brouard 60: Revision 1.342 2022/09/11 19:54:09 brouard
61: Summary: 0.99r38
62:
63: * imach.c (Module): Adding timevarying products of any kinds,
64: should work before shifting cotvar from ncovcol+nqv columns in
65: order to have a correspondance between the column of cotvar and
66: the id of column.
67: (Module): Some cleaning and adding covariates in ILK.txt
68:
1.342 brouard 69: Revision 1.341 2022/09/11 07:58:42 brouard
70: Summary: Version 0.99r38
71:
72: After adding change in cotvar.
73:
1.341 brouard 74: Revision 1.340 2022/09/11 07:53:11 brouard
75: Summary: Version imach 0.99r37
76:
77: * imach.c (Module): Adding timevarying products of any kinds,
78: should work before shifting cotvar from ncovcol+nqv columns in
79: order to have a correspondance between the column of cotvar and
80: the id of column.
81:
1.340 brouard 82: Revision 1.339 2022/09/09 17:55:22 brouard
83: Summary: version 0.99r37
84:
85: * imach.c (Module): Many improvements for fixing products of fixed
86: timevarying as well as fixed * fixed, and test with quantitative
87: covariate.
88:
1.339 brouard 89: Revision 1.338 2022/09/04 17:40:33 brouard
90: Summary: 0.99r36
91:
92: * imach.c (Module): Now the easy runs i.e. without result or
93: model=1+age only did not work. The defautl combination should be 1
94: and not 0 because everything hasn't been tranformed yet.
95:
1.338 brouard 96: Revision 1.337 2022/09/02 14:26:02 brouard
97: Summary: version 0.99r35
98:
99: * src/imach.c: Version 0.99r35 because it outputs same results with
100: 1+age+V1+V1*age for females and 1+age for females only
101: (education=1 noweight)
102:
1.337 brouard 103: Revision 1.336 2022/08/31 09:52:36 brouard
104: *** empty log message ***
105:
1.336 brouard 106: Revision 1.335 2022/08/31 08:23:16 brouard
107: Summary: improvements...
108:
1.335 brouard 109: Revision 1.334 2022/08/25 09:08:41 brouard
110: Summary: In progress for quantitative
111:
1.334 brouard 112: Revision 1.333 2022/08/21 09:10:30 brouard
113: * src/imach.c (Module): Version 0.99r33 A lot of changes in
114: reassigning covariates: my first idea was that people will always
115: use the first covariate V1 into the model but in fact they are
116: producing data with many covariates and can use an equation model
117: with some of the covariate; it means that in a model V2+V3 instead
118: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
119: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
120: the equation model is restricted to two variables only (V2, V3)
121: and the combination for V2 should be codtabm(k,1) instead of
122: (codtabm(k,2), and the code should be
123: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
124: made. All of these should be simplified once a day like we did in
125: hpxij() for example by using precov[nres] which is computed in
126: decoderesult for each nres of each resultline. Loop should be done
127: on the equation model globally by distinguishing only product with
128: age (which are changing with age) and no more on type of
129: covariates, single dummies, single covariates.
130:
1.333 brouard 131: Revision 1.332 2022/08/21 09:06:25 brouard
132: Summary: Version 0.99r33
133:
134: * src/imach.c (Module): Version 0.99r33 A lot of changes in
135: reassigning covariates: my first idea was that people will always
136: use the first covariate V1 into the model but in fact they are
137: producing data with many covariates and can use an equation model
138: with some of the covariate; it means that in a model V2+V3 instead
139: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
140: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
141: the equation model is restricted to two variables only (V2, V3)
142: and the combination for V2 should be codtabm(k,1) instead of
143: (codtabm(k,2), and the code should be
144: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
145: made. All of these should be simplified once a day like we did in
146: hpxij() for example by using precov[nres] which is computed in
147: decoderesult for each nres of each resultline. Loop should be done
148: on the equation model globally by distinguishing only product with
149: age (which are changing with age) and no more on type of
150: covariates, single dummies, single covariates.
151:
1.332 brouard 152: Revision 1.331 2022/08/07 05:40:09 brouard
153: *** empty log message ***
154:
1.331 brouard 155: Revision 1.330 2022/08/06 07:18:25 brouard
156: Summary: last 0.99r31
157:
158: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function
159:
1.330 brouard 160: Revision 1.329 2022/08/03 17:29:54 brouard
161: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
162:
1.329 brouard 163: Revision 1.328 2022/07/27 17:40:48 brouard
164: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
165:
1.328 brouard 166: Revision 1.327 2022/07/27 14:47:35 brouard
167: Summary: Still a problem for one-step probabilities in case of quantitative variables
168:
1.327 brouard 169: Revision 1.326 2022/07/26 17:33:55 brouard
170: Summary: some test with nres=1
171:
1.326 brouard 172: Revision 1.325 2022/07/25 14:27:23 brouard
173: Summary: r30
174:
175: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
176: coredumped, revealed by Feiuno, thank you.
177:
1.325 brouard 178: Revision 1.324 2022/07/23 17:44:26 brouard
179: *** empty log message ***
180:
1.324 brouard 181: Revision 1.323 2022/07/22 12:30:08 brouard
182: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
183:
1.323 brouard 184: Revision 1.322 2022/07/22 12:27:48 brouard
185: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
186:
1.322 brouard 187: Revision 1.321 2022/07/22 12:04:24 brouard
188: Summary: r28
189:
190: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
191:
1.321 brouard 192: Revision 1.320 2022/06/02 05:10:11 brouard
193: *** empty log message ***
194:
1.320 brouard 195: Revision 1.319 2022/06/02 04:45:11 brouard
196: * imach.c (Module): Adding the Wald tests from the log to the main
197: htm for better display of the maximum likelihood estimators.
198:
1.319 brouard 199: Revision 1.318 2022/05/24 08:10:59 brouard
200: * imach.c (Module): Some attempts to find a bug of wrong estimates
201: of confidencce intervals with product in the equation modelC
202:
1.318 brouard 203: Revision 1.317 2022/05/15 15:06:23 brouard
204: * imach.c (Module): Some minor improvements
205:
1.317 brouard 206: Revision 1.316 2022/05/11 15:11:31 brouard
207: Summary: r27
208:
1.316 brouard 209: Revision 1.315 2022/05/11 15:06:32 brouard
210: *** empty log message ***
211:
1.315 brouard 212: Revision 1.314 2022/04/13 17:43:09 brouard
213: * imach.c (Module): Adding link to text data files
214:
1.314 brouard 215: Revision 1.313 2022/04/11 15:57:42 brouard
216: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
217:
1.313 brouard 218: Revision 1.312 2022/04/05 21:24:39 brouard
219: *** empty log message ***
220:
1.312 brouard 221: Revision 1.311 2022/04/05 21:03:51 brouard
222: Summary: Fixed quantitative covariates
223:
224: Fixed covariates (dummy or quantitative)
225: with missing values have never been allowed but are ERRORS and
226: program quits. Standard deviations of fixed covariates were
227: wrongly computed. Mean and standard deviations of time varying
228: covariates are still not computed.
229:
1.311 brouard 230: Revision 1.310 2022/03/17 08:45:53 brouard
231: Summary: 99r25
232:
233: Improving detection of errors: result lines should be compatible with
234: the model.
235:
1.310 brouard 236: Revision 1.309 2021/05/20 12:39:14 brouard
237: Summary: Version 0.99r24
238:
1.309 brouard 239: Revision 1.308 2021/03/31 13:11:57 brouard
240: Summary: Version 0.99r23
241:
242:
243: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
244:
1.308 brouard 245: Revision 1.307 2021/03/08 18:11:32 brouard
246: Summary: 0.99r22 fixed bug on result:
247:
1.307 brouard 248: Revision 1.306 2021/02/20 15:44:02 brouard
249: Summary: Version 0.99r21
250:
251: * imach.c (Module): Fix bug on quitting after result lines!
252: (Module): Version 0.99r21
253:
1.306 brouard 254: Revision 1.305 2021/02/20 15:28:30 brouard
255: * imach.c (Module): Fix bug on quitting after result lines!
256:
1.305 brouard 257: Revision 1.304 2021/02/12 11:34:20 brouard
258: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
259:
1.304 brouard 260: Revision 1.303 2021/02/11 19:50:15 brouard
261: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
262:
1.303 brouard 263: Revision 1.302 2020/02/22 21:00:05 brouard
264: * (Module): imach.c Update mle=-3 (for computing Life expectancy
265: and life table from the data without any state)
266:
1.302 brouard 267: Revision 1.301 2019/06/04 13:51:20 brouard
268: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
269:
1.301 brouard 270: Revision 1.300 2019/05/22 19:09:45 brouard
271: Summary: version 0.99r19 of May 2019
272:
1.300 brouard 273: Revision 1.299 2019/05/22 18:37:08 brouard
274: Summary: Cleaned 0.99r19
275:
1.299 brouard 276: Revision 1.298 2019/05/22 18:19:56 brouard
277: *** empty log message ***
278:
1.298 brouard 279: Revision 1.297 2019/05/22 17:56:10 brouard
280: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
281:
1.297 brouard 282: Revision 1.296 2019/05/20 13:03:18 brouard
283: Summary: Projection syntax simplified
284:
285:
286: We can now start projections, forward or backward, from the mean date
287: of inteviews up to or down to a number of years of projection:
288: prevforecast=1 yearsfproj=15.3 mobil_average=0
289: or
290: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
291: or
292: prevbackcast=1 yearsbproj=12.3 mobil_average=1
293: or
294: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
295:
1.296 brouard 296: Revision 1.295 2019/05/18 09:52:50 brouard
297: Summary: doxygen tex bug
298:
1.295 brouard 299: Revision 1.294 2019/05/16 14:54:33 brouard
300: Summary: There was some wrong lines added
301:
1.294 brouard 302: Revision 1.293 2019/05/09 15:17:34 brouard
303: *** empty log message ***
304:
1.293 brouard 305: Revision 1.292 2019/05/09 14:17:20 brouard
306: Summary: Some updates
307:
1.292 brouard 308: Revision 1.291 2019/05/09 13:44:18 brouard
309: Summary: Before ncovmax
310:
1.291 brouard 311: Revision 1.290 2019/05/09 13:39:37 brouard
312: Summary: 0.99r18 unlimited number of individuals
313:
314: 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.
315:
1.290 brouard 316: Revision 1.289 2018/12/13 09:16:26 brouard
317: Summary: Bug for young ages (<-30) will be in r17
318:
1.289 brouard 319: Revision 1.288 2018/05/02 20:58:27 brouard
320: Summary: Some bugs fixed
321:
1.288 brouard 322: Revision 1.287 2018/05/01 17:57:25 brouard
323: Summary: Bug fixed by providing frequencies only for non missing covariates
324:
1.287 brouard 325: Revision 1.286 2018/04/27 14:27:04 brouard
326: Summary: some minor bugs
327:
1.286 brouard 328: Revision 1.285 2018/04/21 21:02:16 brouard
329: Summary: Some bugs fixed, valgrind tested
330:
1.285 brouard 331: Revision 1.284 2018/04/20 05:22:13 brouard
332: Summary: Computing mean and stdeviation of fixed quantitative variables
333:
1.284 brouard 334: Revision 1.283 2018/04/19 14:49:16 brouard
335: Summary: Some minor bugs fixed
336:
1.283 brouard 337: Revision 1.282 2018/02/27 22:50:02 brouard
338: *** empty log message ***
339:
1.282 brouard 340: Revision 1.281 2018/02/27 19:25:23 brouard
341: Summary: Adding second argument for quitting
342:
1.281 brouard 343: Revision 1.280 2018/02/21 07:58:13 brouard
344: Summary: 0.99r15
345:
346: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
347:
1.280 brouard 348: Revision 1.279 2017/07/20 13:35:01 brouard
349: Summary: temporary working
350:
1.279 brouard 351: Revision 1.278 2017/07/19 14:09:02 brouard
352: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
353:
1.278 brouard 354: Revision 1.277 2017/07/17 08:53:49 brouard
355: Summary: BOM files can be read now
356:
1.277 brouard 357: Revision 1.276 2017/06/30 15:48:31 brouard
358: Summary: Graphs improvements
359:
1.276 brouard 360: Revision 1.275 2017/06/30 13:39:33 brouard
361: Summary: Saito's color
362:
1.275 brouard 363: Revision 1.274 2017/06/29 09:47:08 brouard
364: Summary: Version 0.99r14
365:
1.274 brouard 366: Revision 1.273 2017/06/27 11:06:02 brouard
367: Summary: More documentation on projections
368:
1.273 brouard 369: Revision 1.272 2017/06/27 10:22:40 brouard
370: Summary: Color of backprojection changed from 6 to 5(yellow)
371:
1.272 brouard 372: Revision 1.271 2017/06/27 10:17:50 brouard
373: Summary: Some bug with rint
374:
1.271 brouard 375: Revision 1.270 2017/05/24 05:45:29 brouard
376: *** empty log message ***
377:
1.270 brouard 378: Revision 1.269 2017/05/23 08:39:25 brouard
379: Summary: Code into subroutine, cleanings
380:
1.269 brouard 381: Revision 1.268 2017/05/18 20:09:32 brouard
382: Summary: backprojection and confidence intervals of backprevalence
383:
1.268 brouard 384: Revision 1.267 2017/05/13 10:25:05 brouard
385: Summary: temporary save for backprojection
386:
1.267 brouard 387: Revision 1.266 2017/05/13 07:26:12 brouard
388: Summary: Version 0.99r13 (improvements and bugs fixed)
389:
1.266 brouard 390: Revision 1.265 2017/04/26 16:22:11 brouard
391: Summary: imach 0.99r13 Some bugs fixed
392:
1.265 brouard 393: Revision 1.264 2017/04/26 06:01:29 brouard
394: Summary: Labels in graphs
395:
1.264 brouard 396: Revision 1.263 2017/04/24 15:23:15 brouard
397: Summary: to save
398:
1.263 brouard 399: Revision 1.262 2017/04/18 16:48:12 brouard
400: *** empty log message ***
401:
1.262 brouard 402: Revision 1.261 2017/04/05 10:14:09 brouard
403: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
404:
1.261 brouard 405: Revision 1.260 2017/04/04 17:46:59 brouard
406: Summary: Gnuplot indexations fixed (humm)
407:
1.260 brouard 408: Revision 1.259 2017/04/04 13:01:16 brouard
409: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
410:
1.259 brouard 411: Revision 1.258 2017/04/03 10:17:47 brouard
412: Summary: Version 0.99r12
413:
414: Some cleanings, conformed with updated documentation.
415:
1.258 brouard 416: Revision 1.257 2017/03/29 16:53:30 brouard
417: Summary: Temp
418:
1.257 brouard 419: Revision 1.256 2017/03/27 05:50:23 brouard
420: Summary: Temporary
421:
1.256 brouard 422: Revision 1.255 2017/03/08 16:02:28 brouard
423: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
424:
1.255 brouard 425: Revision 1.254 2017/03/08 07:13:00 brouard
426: Summary: Fixing data parameter line
427:
1.254 brouard 428: Revision 1.253 2016/12/15 11:59:41 brouard
429: Summary: 0.99 in progress
430:
1.253 brouard 431: Revision 1.252 2016/09/15 21:15:37 brouard
432: *** empty log message ***
433:
1.252 brouard 434: Revision 1.251 2016/09/15 15:01:13 brouard
435: Summary: not working
436:
1.251 brouard 437: Revision 1.250 2016/09/08 16:07:27 brouard
438: Summary: continue
439:
1.250 brouard 440: Revision 1.249 2016/09/07 17:14:18 brouard
441: Summary: Starting values from frequencies
442:
1.249 brouard 443: Revision 1.248 2016/09/07 14:10:18 brouard
444: *** empty log message ***
445:
1.248 brouard 446: Revision 1.247 2016/09/02 11:11:21 brouard
447: *** empty log message ***
448:
1.247 brouard 449: Revision 1.246 2016/09/02 08:49:22 brouard
450: *** empty log message ***
451:
1.246 brouard 452: Revision 1.245 2016/09/02 07:25:01 brouard
453: *** empty log message ***
454:
1.245 brouard 455: Revision 1.244 2016/09/02 07:17:34 brouard
456: *** empty log message ***
457:
1.244 brouard 458: Revision 1.243 2016/09/02 06:45:35 brouard
459: *** empty log message ***
460:
1.243 brouard 461: Revision 1.242 2016/08/30 15:01:20 brouard
462: Summary: Fixing a lots
463:
1.242 brouard 464: Revision 1.241 2016/08/29 17:17:25 brouard
465: Summary: gnuplot problem in Back projection to fix
466:
1.241 brouard 467: Revision 1.240 2016/08/29 07:53:18 brouard
468: Summary: Better
469:
1.240 brouard 470: Revision 1.239 2016/08/26 15:51:03 brouard
471: Summary: Improvement in Powell output in order to copy and paste
472:
473: Author:
474:
1.239 brouard 475: Revision 1.238 2016/08/26 14:23:35 brouard
476: Summary: Starting tests of 0.99
477:
1.238 brouard 478: Revision 1.237 2016/08/26 09:20:19 brouard
479: Summary: to valgrind
480:
1.237 brouard 481: Revision 1.236 2016/08/25 10:50:18 brouard
482: *** empty log message ***
483:
1.236 brouard 484: Revision 1.235 2016/08/25 06:59:23 brouard
485: *** empty log message ***
486:
1.235 brouard 487: Revision 1.234 2016/08/23 16:51:20 brouard
488: *** empty log message ***
489:
1.234 brouard 490: Revision 1.233 2016/08/23 07:40:50 brouard
491: Summary: not working
492:
1.233 brouard 493: Revision 1.232 2016/08/22 14:20:21 brouard
494: Summary: not working
495:
1.232 brouard 496: Revision 1.231 2016/08/22 07:17:15 brouard
497: Summary: not working
498:
1.231 brouard 499: Revision 1.230 2016/08/22 06:55:53 brouard
500: Summary: Not working
501:
1.230 brouard 502: Revision 1.229 2016/07/23 09:45:53 brouard
503: Summary: Completing for func too
504:
1.229 brouard 505: Revision 1.228 2016/07/22 17:45:30 brouard
506: Summary: Fixing some arrays, still debugging
507:
1.227 brouard 508: Revision 1.226 2016/07/12 18:42:34 brouard
509: Summary: temp
510:
1.226 brouard 511: Revision 1.225 2016/07/12 08:40:03 brouard
512: Summary: saving but not running
513:
1.225 brouard 514: Revision 1.224 2016/07/01 13:16:01 brouard
515: Summary: Fixes
516:
1.224 brouard 517: Revision 1.223 2016/02/19 09:23:35 brouard
518: Summary: temporary
519:
1.223 brouard 520: Revision 1.222 2016/02/17 08:14:50 brouard
521: Summary: Probably last 0.98 stable version 0.98r6
522:
1.222 brouard 523: Revision 1.221 2016/02/15 23:35:36 brouard
524: Summary: minor bug
525:
1.220 brouard 526: Revision 1.219 2016/02/15 00:48:12 brouard
527: *** empty log message ***
528:
1.219 brouard 529: Revision 1.218 2016/02/12 11:29:23 brouard
530: Summary: 0.99 Back projections
531:
1.218 brouard 532: Revision 1.217 2015/12/23 17:18:31 brouard
533: Summary: Experimental backcast
534:
1.217 brouard 535: Revision 1.216 2015/12/18 17:32:11 brouard
536: Summary: 0.98r4 Warning and status=-2
537:
538: Version 0.98r4 is now:
539: - displaying an error when status is -1, date of interview unknown and date of death known;
540: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
541: Older changes concerning s=-2, dating from 2005 have been supersed.
542:
1.216 brouard 543: Revision 1.215 2015/12/16 08:52:24 brouard
544: Summary: 0.98r4 working
545:
1.215 brouard 546: Revision 1.214 2015/12/16 06:57:54 brouard
547: Summary: temporary not working
548:
1.214 brouard 549: Revision 1.213 2015/12/11 18:22:17 brouard
550: Summary: 0.98r4
551:
1.213 brouard 552: Revision 1.212 2015/11/21 12:47:24 brouard
553: Summary: minor typo
554:
1.212 brouard 555: Revision 1.211 2015/11/21 12:41:11 brouard
556: Summary: 0.98r3 with some graph of projected cross-sectional
557:
558: Author: Nicolas Brouard
559:
1.211 brouard 560: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 561: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 562: Summary: Adding ftolpl parameter
563: Author: N Brouard
564:
565: We had difficulties to get smoothed confidence intervals. It was due
566: to the period prevalence which wasn't computed accurately. The inner
567: parameter ftolpl is now an outer parameter of the .imach parameter
568: file after estepm. If ftolpl is small 1.e-4 and estepm too,
569: computation are long.
570:
1.209 brouard 571: Revision 1.208 2015/11/17 14:31:57 brouard
572: Summary: temporary
573:
1.208 brouard 574: Revision 1.207 2015/10/27 17:36:57 brouard
575: *** empty log message ***
576:
1.207 brouard 577: Revision 1.206 2015/10/24 07:14:11 brouard
578: *** empty log message ***
579:
1.206 brouard 580: Revision 1.205 2015/10/23 15:50:53 brouard
581: Summary: 0.98r3 some clarification for graphs on likelihood contributions
582:
1.205 brouard 583: Revision 1.204 2015/10/01 16:20:26 brouard
584: Summary: Some new graphs of contribution to likelihood
585:
1.204 brouard 586: Revision 1.203 2015/09/30 17:45:14 brouard
587: Summary: looking at better estimation of the hessian
588:
589: Also a better criteria for convergence to the period prevalence And
590: therefore adding the number of years needed to converge. (The
591: prevalence in any alive state shold sum to one
592:
1.203 brouard 593: Revision 1.202 2015/09/22 19:45:16 brouard
594: Summary: Adding some overall graph on contribution to likelihood. Might change
595:
1.202 brouard 596: Revision 1.201 2015/09/15 17:34:58 brouard
597: Summary: 0.98r0
598:
599: - Some new graphs like suvival functions
600: - Some bugs fixed like model=1+age+V2.
601:
1.201 brouard 602: Revision 1.200 2015/09/09 16:53:55 brouard
603: Summary: Big bug thanks to Flavia
604:
605: Even model=1+age+V2. did not work anymore
606:
1.200 brouard 607: Revision 1.199 2015/09/07 14:09:23 brouard
608: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
609:
1.199 brouard 610: Revision 1.198 2015/09/03 07:14:39 brouard
611: Summary: 0.98q5 Flavia
612:
1.198 brouard 613: Revision 1.197 2015/09/01 18:24:39 brouard
614: *** empty log message ***
615:
1.197 brouard 616: Revision 1.196 2015/08/18 23:17:52 brouard
617: Summary: 0.98q5
618:
1.196 brouard 619: Revision 1.195 2015/08/18 16:28:39 brouard
620: Summary: Adding a hack for testing purpose
621:
622: After reading the title, ftol and model lines, if the comment line has
623: a q, starting with #q, the answer at the end of the run is quit. It
624: permits to run test files in batch with ctest. The former workaround was
625: $ echo q | imach foo.imach
626:
1.195 brouard 627: Revision 1.194 2015/08/18 13:32:00 brouard
628: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
629:
1.194 brouard 630: Revision 1.193 2015/08/04 07:17:42 brouard
631: Summary: 0.98q4
632:
1.193 brouard 633: Revision 1.192 2015/07/16 16:49:02 brouard
634: Summary: Fixing some outputs
635:
1.192 brouard 636: Revision 1.191 2015/07/14 10:00:33 brouard
637: Summary: Some fixes
638:
1.191 brouard 639: Revision 1.190 2015/05/05 08:51:13 brouard
640: Summary: Adding digits in output parameters (7 digits instead of 6)
641:
642: Fix 1+age+.
643:
1.190 brouard 644: Revision 1.189 2015/04/30 14:45:16 brouard
645: Summary: 0.98q2
646:
1.189 brouard 647: Revision 1.188 2015/04/30 08:27:53 brouard
648: *** empty log message ***
649:
1.188 brouard 650: Revision 1.187 2015/04/29 09:11:15 brouard
651: *** empty log message ***
652:
1.187 brouard 653: Revision 1.186 2015/04/23 12:01:52 brouard
654: Summary: V1*age is working now, version 0.98q1
655:
656: Some codes had been disabled in order to simplify and Vn*age was
657: working in the optimization phase, ie, giving correct MLE parameters,
658: but, as usual, outputs were not correct and program core dumped.
659:
1.186 brouard 660: Revision 1.185 2015/03/11 13:26:42 brouard
661: Summary: Inclusion of compile and links command line for Intel Compiler
662:
1.185 brouard 663: Revision 1.184 2015/03/11 11:52:39 brouard
664: Summary: Back from Windows 8. Intel Compiler
665:
1.184 brouard 666: Revision 1.183 2015/03/10 20:34:32 brouard
667: Summary: 0.98q0, trying with directest, mnbrak fixed
668:
669: We use directest instead of original Powell test; probably no
670: incidence on the results, but better justifications;
671: We fixed Numerical Recipes mnbrak routine which was wrong and gave
672: wrong results.
673:
1.183 brouard 674: Revision 1.182 2015/02/12 08:19:57 brouard
675: Summary: Trying to keep directest which seems simpler and more general
676: Author: Nicolas Brouard
677:
1.182 brouard 678: Revision 1.181 2015/02/11 23:22:24 brouard
679: Summary: Comments on Powell added
680:
681: Author:
682:
1.181 brouard 683: Revision 1.180 2015/02/11 17:33:45 brouard
684: Summary: Finishing move from main to function (hpijx and prevalence_limit)
685:
1.180 brouard 686: Revision 1.179 2015/01/04 09:57:06 brouard
687: Summary: back to OS/X
688:
1.179 brouard 689: Revision 1.178 2015/01/04 09:35:48 brouard
690: *** empty log message ***
691:
1.178 brouard 692: Revision 1.177 2015/01/03 18:40:56 brouard
693: Summary: Still testing ilc32 on OSX
694:
1.177 brouard 695: Revision 1.176 2015/01/03 16:45:04 brouard
696: *** empty log message ***
697:
1.176 brouard 698: Revision 1.175 2015/01/03 16:33:42 brouard
699: *** empty log message ***
700:
1.175 brouard 701: Revision 1.174 2015/01/03 16:15:49 brouard
702: Summary: Still in cross-compilation
703:
1.174 brouard 704: Revision 1.173 2015/01/03 12:06:26 brouard
705: Summary: trying to detect cross-compilation
706:
1.173 brouard 707: Revision 1.172 2014/12/27 12:07:47 brouard
708: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
709:
1.172 brouard 710: Revision 1.171 2014/12/23 13:26:59 brouard
711: Summary: Back from Visual C
712:
713: Still problem with utsname.h on Windows
714:
1.171 brouard 715: Revision 1.170 2014/12/23 11:17:12 brouard
716: Summary: Cleaning some \%% back to %%
717:
718: The escape was mandatory for a specific compiler (which one?), but too many warnings.
719:
1.170 brouard 720: Revision 1.169 2014/12/22 23:08:31 brouard
721: Summary: 0.98p
722:
723: Outputs some informations on compiler used, OS etc. Testing on different platforms.
724:
1.169 brouard 725: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 726: Summary: update
1.169 brouard 727:
1.168 brouard 728: Revision 1.167 2014/12/22 13:50:56 brouard
729: Summary: Testing uname and compiler version and if compiled 32 or 64
730:
731: Testing on Linux 64
732:
1.167 brouard 733: Revision 1.166 2014/12/22 11:40:47 brouard
734: *** empty log message ***
735:
1.166 brouard 736: Revision 1.165 2014/12/16 11:20:36 brouard
737: Summary: After compiling on Visual C
738:
739: * imach.c (Module): Merging 1.61 to 1.162
740:
1.165 brouard 741: Revision 1.164 2014/12/16 10:52:11 brouard
742: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
743:
744: * imach.c (Module): Merging 1.61 to 1.162
745:
1.164 brouard 746: Revision 1.163 2014/12/16 10:30:11 brouard
747: * imach.c (Module): Merging 1.61 to 1.162
748:
1.163 brouard 749: Revision 1.162 2014/09/25 11:43:39 brouard
750: Summary: temporary backup 0.99!
751:
1.162 brouard 752: Revision 1.1 2014/09/16 11:06:58 brouard
753: Summary: With some code (wrong) for nlopt
754:
755: Author:
756:
757: Revision 1.161 2014/09/15 20:41:41 brouard
758: Summary: Problem with macro SQR on Intel compiler
759:
1.161 brouard 760: Revision 1.160 2014/09/02 09:24:05 brouard
761: *** empty log message ***
762:
1.160 brouard 763: Revision 1.159 2014/09/01 10:34:10 brouard
764: Summary: WIN32
765: Author: Brouard
766:
1.159 brouard 767: Revision 1.158 2014/08/27 17:11:51 brouard
768: *** empty log message ***
769:
1.158 brouard 770: Revision 1.157 2014/08/27 16:26:55 brouard
771: Summary: Preparing windows Visual studio version
772: Author: Brouard
773:
774: In order to compile on Visual studio, time.h is now correct and time_t
775: and tm struct should be used. difftime should be used but sometimes I
776: just make the differences in raw time format (time(&now).
777: Trying to suppress #ifdef LINUX
778: Add xdg-open for __linux in order to open default browser.
779:
1.157 brouard 780: Revision 1.156 2014/08/25 20:10:10 brouard
781: *** empty log message ***
782:
1.156 brouard 783: Revision 1.155 2014/08/25 18:32:34 brouard
784: Summary: New compile, minor changes
785: Author: Brouard
786:
1.155 brouard 787: Revision 1.154 2014/06/20 17:32:08 brouard
788: Summary: Outputs now all graphs of convergence to period prevalence
789:
1.154 brouard 790: Revision 1.153 2014/06/20 16:45:46 brouard
791: Summary: If 3 live state, convergence to period prevalence on same graph
792: Author: Brouard
793:
1.153 brouard 794: Revision 1.152 2014/06/18 17:54:09 brouard
795: Summary: open browser, use gnuplot on same dir than imach if not found in the path
796:
1.152 brouard 797: Revision 1.151 2014/06/18 16:43:30 brouard
798: *** empty log message ***
799:
1.151 brouard 800: Revision 1.150 2014/06/18 16:42:35 brouard
801: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
802: Author: brouard
803:
1.150 brouard 804: Revision 1.149 2014/06/18 15:51:14 brouard
805: Summary: Some fixes in parameter files errors
806: Author: Nicolas Brouard
807:
1.149 brouard 808: Revision 1.148 2014/06/17 17:38:48 brouard
809: Summary: Nothing new
810: Author: Brouard
811:
812: Just a new packaging for OS/X version 0.98nS
813:
1.148 brouard 814: Revision 1.147 2014/06/16 10:33:11 brouard
815: *** empty log message ***
816:
1.147 brouard 817: Revision 1.146 2014/06/16 10:20:28 brouard
818: Summary: Merge
819: Author: Brouard
820:
821: Merge, before building revised version.
822:
1.146 brouard 823: Revision 1.145 2014/06/10 21:23:15 brouard
824: Summary: Debugging with valgrind
825: Author: Nicolas Brouard
826:
827: Lot of changes in order to output the results with some covariates
828: After the Edimburgh REVES conference 2014, it seems mandatory to
829: improve the code.
830: No more memory valgrind error but a lot has to be done in order to
831: continue the work of splitting the code into subroutines.
832: Also, decodemodel has been improved. Tricode is still not
833: optimal. nbcode should be improved. Documentation has been added in
834: the source code.
835:
1.144 brouard 836: Revision 1.143 2014/01/26 09:45:38 brouard
837: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
838:
839: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
840: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
841:
1.143 brouard 842: Revision 1.142 2014/01/26 03:57:36 brouard
843: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
844:
845: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
846:
1.142 brouard 847: Revision 1.141 2014/01/26 02:42:01 brouard
848: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
849:
1.141 brouard 850: Revision 1.140 2011/09/02 10:37:54 brouard
851: Summary: times.h is ok with mingw32 now.
852:
1.140 brouard 853: Revision 1.139 2010/06/14 07:50:17 brouard
854: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
855: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
856:
1.139 brouard 857: Revision 1.138 2010/04/30 18:19:40 brouard
858: *** empty log message ***
859:
1.138 brouard 860: Revision 1.137 2010/04/29 18:11:38 brouard
861: (Module): Checking covariates for more complex models
862: than V1+V2. A lot of change to be done. Unstable.
863:
1.137 brouard 864: Revision 1.136 2010/04/26 20:30:53 brouard
865: (Module): merging some libgsl code. Fixing computation
866: of likelione (using inter/intrapolation if mle = 0) in order to
867: get same likelihood as if mle=1.
868: Some cleaning of code and comments added.
869:
1.136 brouard 870: Revision 1.135 2009/10/29 15:33:14 brouard
871: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
872:
1.135 brouard 873: Revision 1.134 2009/10/29 13:18:53 brouard
874: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
875:
1.134 brouard 876: Revision 1.133 2009/07/06 10:21:25 brouard
877: just nforces
878:
1.133 brouard 879: Revision 1.132 2009/07/06 08:22:05 brouard
880: Many tings
881:
1.132 brouard 882: Revision 1.131 2009/06/20 16:22:47 brouard
883: Some dimensions resccaled
884:
1.131 brouard 885: Revision 1.130 2009/05/26 06:44:34 brouard
886: (Module): Max Covariate is now set to 20 instead of 8. A
887: lot of cleaning with variables initialized to 0. Trying to make
888: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
889:
1.130 brouard 890: Revision 1.129 2007/08/31 13:49:27 lievre
891: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
892:
1.129 lievre 893: Revision 1.128 2006/06/30 13:02:05 brouard
894: (Module): Clarifications on computing e.j
895:
1.128 brouard 896: Revision 1.127 2006/04/28 18:11:50 brouard
897: (Module): Yes the sum of survivors was wrong since
898: imach-114 because nhstepm was no more computed in the age
899: loop. Now we define nhstepma in the age loop.
900: (Module): In order to speed up (in case of numerous covariates) we
901: compute health expectancies (without variances) in a first step
902: and then all the health expectancies with variances or standard
903: deviation (needs data from the Hessian matrices) which slows the
904: computation.
905: In the future we should be able to stop the program is only health
906: expectancies and graph are needed without standard deviations.
907:
1.127 brouard 908: Revision 1.126 2006/04/28 17:23:28 brouard
909: (Module): Yes the sum of survivors was wrong since
910: imach-114 because nhstepm was no more computed in the age
911: loop. Now we define nhstepma in the age loop.
912: Version 0.98h
913:
1.126 brouard 914: Revision 1.125 2006/04/04 15:20:31 lievre
915: Errors in calculation of health expectancies. Age was not initialized.
916: Forecasting file added.
917:
918: Revision 1.124 2006/03/22 17:13:53 lievre
919: Parameters are printed with %lf instead of %f (more numbers after the comma).
920: The log-likelihood is printed in the log file
921:
922: Revision 1.123 2006/03/20 10:52:43 brouard
923: * imach.c (Module): <title> changed, corresponds to .htm file
924: name. <head> headers where missing.
925:
926: * imach.c (Module): Weights can have a decimal point as for
927: English (a comma might work with a correct LC_NUMERIC environment,
928: otherwise the weight is truncated).
929: Modification of warning when the covariates values are not 0 or
930: 1.
931: Version 0.98g
932:
933: Revision 1.122 2006/03/20 09:45:41 brouard
934: (Module): Weights can have a decimal point as for
935: English (a comma might work with a correct LC_NUMERIC environment,
936: otherwise the weight is truncated).
937: Modification of warning when the covariates values are not 0 or
938: 1.
939: Version 0.98g
940:
941: Revision 1.121 2006/03/16 17:45:01 lievre
942: * imach.c (Module): Comments concerning covariates added
943:
944: * imach.c (Module): refinements in the computation of lli if
945: status=-2 in order to have more reliable computation if stepm is
946: not 1 month. Version 0.98f
947:
948: Revision 1.120 2006/03/16 15:10:38 lievre
949: (Module): refinements in the computation of lli if
950: status=-2 in order to have more reliable computation if stepm is
951: not 1 month. Version 0.98f
952:
953: Revision 1.119 2006/03/15 17:42:26 brouard
954: (Module): Bug if status = -2, the loglikelihood was
955: computed as likelihood omitting the logarithm. Version O.98e
956:
957: Revision 1.118 2006/03/14 18:20:07 brouard
958: (Module): varevsij Comments added explaining the second
959: table of variances if popbased=1 .
960: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
961: (Module): Function pstamp added
962: (Module): Version 0.98d
963:
964: Revision 1.117 2006/03/14 17:16:22 brouard
965: (Module): varevsij Comments added explaining the second
966: table of variances if popbased=1 .
967: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
968: (Module): Function pstamp added
969: (Module): Version 0.98d
970:
971: Revision 1.116 2006/03/06 10:29:27 brouard
972: (Module): Variance-covariance wrong links and
973: varian-covariance of ej. is needed (Saito).
974:
975: Revision 1.115 2006/02/27 12:17:45 brouard
976: (Module): One freematrix added in mlikeli! 0.98c
977:
978: Revision 1.114 2006/02/26 12:57:58 brouard
979: (Module): Some improvements in processing parameter
980: filename with strsep.
981:
982: Revision 1.113 2006/02/24 14:20:24 brouard
983: (Module): Memory leaks checks with valgrind and:
984: datafile was not closed, some imatrix were not freed and on matrix
985: allocation too.
986:
987: Revision 1.112 2006/01/30 09:55:26 brouard
988: (Module): Back to gnuplot.exe instead of wgnuplot.exe
989:
990: Revision 1.111 2006/01/25 20:38:18 brouard
991: (Module): Lots of cleaning and bugs added (Gompertz)
992: (Module): Comments can be added in data file. Missing date values
993: can be a simple dot '.'.
994:
995: Revision 1.110 2006/01/25 00:51:50 brouard
996: (Module): Lots of cleaning and bugs added (Gompertz)
997:
998: Revision 1.109 2006/01/24 19:37:15 brouard
999: (Module): Comments (lines starting with a #) are allowed in data.
1000:
1001: Revision 1.108 2006/01/19 18:05:42 lievre
1002: Gnuplot problem appeared...
1003: To be fixed
1004:
1005: Revision 1.107 2006/01/19 16:20:37 brouard
1006: Test existence of gnuplot in imach path
1007:
1008: Revision 1.106 2006/01/19 13:24:36 brouard
1009: Some cleaning and links added in html output
1010:
1011: Revision 1.105 2006/01/05 20:23:19 lievre
1012: *** empty log message ***
1013:
1014: Revision 1.104 2005/09/30 16:11:43 lievre
1015: (Module): sump fixed, loop imx fixed, and simplifications.
1016: (Module): If the status is missing at the last wave but we know
1017: that the person is alive, then we can code his/her status as -2
1018: (instead of missing=-1 in earlier versions) and his/her
1019: contributions to the likelihood is 1 - Prob of dying from last
1020: health status (= 1-p13= p11+p12 in the easiest case of somebody in
1021: the healthy state at last known wave). Version is 0.98
1022:
1023: Revision 1.103 2005/09/30 15:54:49 lievre
1024: (Module): sump fixed, loop imx fixed, and simplifications.
1025:
1026: Revision 1.102 2004/09/15 17:31:30 brouard
1027: Add the possibility to read data file including tab characters.
1028:
1029: Revision 1.101 2004/09/15 10:38:38 brouard
1030: Fix on curr_time
1031:
1032: Revision 1.100 2004/07/12 18:29:06 brouard
1033: Add version for Mac OS X. Just define UNIX in Makefile
1034:
1035: Revision 1.99 2004/06/05 08:57:40 brouard
1036: *** empty log message ***
1037:
1038: Revision 1.98 2004/05/16 15:05:56 brouard
1039: New version 0.97 . First attempt to estimate force of mortality
1040: directly from the data i.e. without the need of knowing the health
1041: state at each age, but using a Gompertz model: log u =a + b*age .
1042: This is the basic analysis of mortality and should be done before any
1043: other analysis, in order to test if the mortality estimated from the
1044: cross-longitudinal survey is different from the mortality estimated
1045: from other sources like vital statistic data.
1046:
1047: The same imach parameter file can be used but the option for mle should be -3.
1048:
1.324 brouard 1049: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 1050: former routines in order to include the new code within the former code.
1051:
1052: The output is very simple: only an estimate of the intercept and of
1053: the slope with 95% confident intervals.
1054:
1055: Current limitations:
1056: A) Even if you enter covariates, i.e. with the
1057: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
1058: B) There is no computation of Life Expectancy nor Life Table.
1059:
1060: Revision 1.97 2004/02/20 13:25:42 lievre
1061: Version 0.96d. Population forecasting command line is (temporarily)
1062: suppressed.
1063:
1064: Revision 1.96 2003/07/15 15:38:55 brouard
1065: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
1066: rewritten within the same printf. Workaround: many printfs.
1067:
1068: Revision 1.95 2003/07/08 07:54:34 brouard
1069: * imach.c (Repository):
1070: (Repository): Using imachwizard code to output a more meaningful covariance
1071: matrix (cov(a12,c31) instead of numbers.
1072:
1073: Revision 1.94 2003/06/27 13:00:02 brouard
1074: Just cleaning
1075:
1076: Revision 1.93 2003/06/25 16:33:55 brouard
1077: (Module): On windows (cygwin) function asctime_r doesn't
1078: exist so I changed back to asctime which exists.
1079: (Module): Version 0.96b
1080:
1081: Revision 1.92 2003/06/25 16:30:45 brouard
1082: (Module): On windows (cygwin) function asctime_r doesn't
1083: exist so I changed back to asctime which exists.
1084:
1085: Revision 1.91 2003/06/25 15:30:29 brouard
1086: * imach.c (Repository): Duplicated warning errors corrected.
1087: (Repository): Elapsed time after each iteration is now output. It
1088: helps to forecast when convergence will be reached. Elapsed time
1089: is stamped in powell. We created a new html file for the graphs
1090: concerning matrix of covariance. It has extension -cov.htm.
1091:
1092: Revision 1.90 2003/06/24 12:34:15 brouard
1093: (Module): Some bugs corrected for windows. Also, when
1094: mle=-1 a template is output in file "or"mypar.txt with the design
1095: of the covariance matrix to be input.
1096:
1097: Revision 1.89 2003/06/24 12:30:52 brouard
1098: (Module): Some bugs corrected for windows. Also, when
1099: mle=-1 a template is output in file "or"mypar.txt with the design
1100: of the covariance matrix to be input.
1101:
1102: Revision 1.88 2003/06/23 17:54:56 brouard
1103: * 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.
1104:
1105: Revision 1.87 2003/06/18 12:26:01 brouard
1106: Version 0.96
1107:
1108: Revision 1.86 2003/06/17 20:04:08 brouard
1109: (Module): Change position of html and gnuplot routines and added
1110: routine fileappend.
1111:
1112: Revision 1.85 2003/06/17 13:12:43 brouard
1113: * imach.c (Repository): Check when date of death was earlier that
1114: current date of interview. It may happen when the death was just
1115: prior to the death. In this case, dh was negative and likelihood
1116: was wrong (infinity). We still send an "Error" but patch by
1117: assuming that the date of death was just one stepm after the
1118: interview.
1119: (Repository): Because some people have very long ID (first column)
1120: we changed int to long in num[] and we added a new lvector for
1121: memory allocation. But we also truncated to 8 characters (left
1122: truncation)
1123: (Repository): No more line truncation errors.
1124:
1125: Revision 1.84 2003/06/13 21:44:43 brouard
1126: * imach.c (Repository): Replace "freqsummary" at a correct
1127: place. It differs from routine "prevalence" which may be called
1128: many times. Probs is memory consuming and must be used with
1129: parcimony.
1130: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
1131:
1132: Revision 1.83 2003/06/10 13:39:11 lievre
1133: *** empty log message ***
1134:
1135: Revision 1.82 2003/06/05 15:57:20 brouard
1136: Add log in imach.c and fullversion number is now printed.
1137:
1138: */
1139: /*
1140: Interpolated Markov Chain
1141:
1142: Short summary of the programme:
1143:
1.227 brouard 1144: This program computes Healthy Life Expectancies or State-specific
1145: (if states aren't health statuses) Expectancies from
1146: cross-longitudinal data. Cross-longitudinal data consist in:
1147:
1148: -1- a first survey ("cross") where individuals from different ages
1149: are interviewed on their health status or degree of disability (in
1150: the case of a health survey which is our main interest)
1151:
1152: -2- at least a second wave of interviews ("longitudinal") which
1153: measure each change (if any) in individual health status. Health
1154: expectancies are computed from the time spent in each health state
1155: according to a model. More health states you consider, more time is
1156: necessary to reach the Maximum Likelihood of the parameters involved
1157: in the model. The simplest model is the multinomial logistic model
1158: where pij is the probability to be observed in state j at the second
1159: wave conditional to be observed in state i at the first
1160: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1161: etc , where 'age' is age and 'sex' is a covariate. If you want to
1162: have a more complex model than "constant and age", you should modify
1163: the program where the markup *Covariates have to be included here
1164: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1165: convergence.
1166:
1167: The advantage of this computer programme, compared to a simple
1168: multinomial logistic model, is clear when the delay between waves is not
1169: identical for each individual. Also, if a individual missed an
1170: intermediate interview, the information is lost, but taken into
1171: account using an interpolation or extrapolation.
1172:
1173: hPijx is the probability to be observed in state i at age x+h
1174: conditional to the observed state i at age x. The delay 'h' can be
1175: split into an exact number (nh*stepm) of unobserved intermediate
1176: states. This elementary transition (by month, quarter,
1177: semester or year) is modelled as a multinomial logistic. The hPx
1178: matrix is simply the matrix product of nh*stepm elementary matrices
1179: and the contribution of each individual to the likelihood is simply
1180: hPijx.
1181:
1182: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1183: of the life expectancies. It also computes the period (stable) prevalence.
1184:
1185: Back prevalence and projections:
1.227 brouard 1186:
1187: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1188: double agemaxpar, double ftolpl, int *ncvyearp, double
1189: dateprev1,double dateprev2, int firstpass, int lastpass, int
1190: mobilavproj)
1191:
1192: Computes the back prevalence limit for any combination of
1193: covariate values k at any age between ageminpar and agemaxpar and
1194: returns it in **bprlim. In the loops,
1195:
1196: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1197: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1198:
1199: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1200: Computes for any combination of covariates k and any age between bage and fage
1201: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1202: oldm=oldms;savm=savms;
1.227 brouard 1203:
1.267 brouard 1204: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1205: Computes the transition matrix starting at age 'age' over
1206: 'nhstepm*hstepm*stepm' months (i.e. until
1207: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1208: nhstepm*hstepm matrices.
1209:
1210: Returns p3mat[i][j][h] after calling
1211: p3mat[i][j][h]=matprod2(newm,
1212: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1213: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1214: oldm);
1.226 brouard 1215:
1216: Important routines
1217:
1218: - func (or funcone), computes logit (pij) distinguishing
1219: o fixed variables (single or product dummies or quantitative);
1220: o varying variables by:
1221: (1) wave (single, product dummies, quantitative),
1222: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1223: % fixed dummy (treated) or quantitative (not done because time-consuming);
1224: % varying dummy (not done) or quantitative (not done);
1225: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1226: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1227: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1228: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1229: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1230:
1.226 brouard 1231:
1232:
1.324 brouard 1233: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1234: Institut national d'études démographiques, Paris.
1.126 brouard 1235: This software have been partly granted by Euro-REVES, a concerted action
1236: from the European Union.
1237: It is copyrighted identically to a GNU software product, ie programme and
1238: software can be distributed freely for non commercial use. Latest version
1239: can be accessed at http://euroreves.ined.fr/imach .
1240:
1241: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1242: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1243:
1244: **********************************************************************/
1245: /*
1246: main
1247: read parameterfile
1248: read datafile
1249: concatwav
1250: freqsummary
1251: if (mle >= 1)
1252: mlikeli
1253: print results files
1254: if mle==1
1255: computes hessian
1256: read end of parameter file: agemin, agemax, bage, fage, estepm
1257: begin-prev-date,...
1258: open gnuplot file
1259: open html file
1.145 brouard 1260: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1261: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1262: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1263: freexexit2 possible for memory heap.
1264:
1265: h Pij x | pij_nom ficrestpij
1266: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1267: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1268: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1269:
1270: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1271: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1272: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1273: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1274: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1275:
1.126 brouard 1276: forecasting if prevfcast==1 prevforecast call prevalence()
1277: health expectancies
1278: Variance-covariance of DFLE
1279: prevalence()
1280: movingaverage()
1281: varevsij()
1282: if popbased==1 varevsij(,popbased)
1283: total life expectancies
1284: Variance of period (stable) prevalence
1285: end
1286: */
1287:
1.187 brouard 1288: /* #define DEBUG */
1289: /* #define DEBUGBRENT */
1.203 brouard 1290: /* #define DEBUGLINMIN */
1291: /* #define DEBUGHESS */
1292: #define DEBUGHESSIJ
1.224 brouard 1293: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1294: #define POWELL /* Instead of NLOPT */
1.224 brouard 1295: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1296: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1297: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1298: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.359 brouard 1299: /* #define POWELLORIGINCONJUGATE /\* Don't use conjugate but biggest decrease if valuable *\/ */
1300: /* #define NOTMINFIT */
1.126 brouard 1301:
1302: #include <math.h>
1303: #include <stdio.h>
1304: #include <stdlib.h>
1305: #include <string.h>
1.226 brouard 1306: #include <ctype.h>
1.159 brouard 1307:
1308: #ifdef _WIN32
1309: #include <io.h>
1.172 brouard 1310: #include <windows.h>
1311: #include <tchar.h>
1.159 brouard 1312: #else
1.126 brouard 1313: #include <unistd.h>
1.159 brouard 1314: #endif
1.126 brouard 1315:
1316: #include <limits.h>
1317: #include <sys/types.h>
1.171 brouard 1318:
1319: #if defined(__GNUC__)
1320: #include <sys/utsname.h> /* Doesn't work on Windows */
1321: #endif
1322:
1.126 brouard 1323: #include <sys/stat.h>
1324: #include <errno.h>
1.159 brouard 1325: /* extern int errno; */
1.126 brouard 1326:
1.157 brouard 1327: /* #ifdef LINUX */
1328: /* #include <time.h> */
1329: /* #include "timeval.h" */
1330: /* #else */
1331: /* #include <sys/time.h> */
1332: /* #endif */
1333:
1.126 brouard 1334: #include <time.h>
1335:
1.136 brouard 1336: #ifdef GSL
1337: #include <gsl/gsl_errno.h>
1338: #include <gsl/gsl_multimin.h>
1339: #endif
1340:
1.167 brouard 1341:
1.162 brouard 1342: #ifdef NLOPT
1343: #include <nlopt.h>
1344: typedef struct {
1345: double (* function)(double [] );
1346: } myfunc_data ;
1347: #endif
1348:
1.126 brouard 1349: /* #include <libintl.h> */
1350: /* #define _(String) gettext (String) */
1351:
1.349 brouard 1352: #define MAXLINE 16384 /* Was 256 and 1024 and 2048. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1353:
1354: #define GNUPLOTPROGRAM "gnuplot"
1.343 brouard 1355: #define GNUPLOTVERSION 5.1
1356: double gnuplotversion=GNUPLOTVERSION;
1.126 brouard 1357: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1358: #define FILENAMELENGTH 256
1.126 brouard 1359:
1360: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1361: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1362:
1.349 brouard 1363: #define MAXPARM 216 /**< Maximum number of parameters for the optimization was 128 */
1.144 brouard 1364: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1365:
1366: #define NINTERVMAX 8
1.144 brouard 1367: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1368: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1369: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1370: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1371: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1372: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1373: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1374: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1375: /* #define AGESUP 130 */
1.288 brouard 1376: /* #define AGESUP 150 */
1377: #define AGESUP 200
1.268 brouard 1378: #define AGEINF 0
1.218 brouard 1379: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1380: #define AGEBASE 40
1.194 brouard 1381: #define AGEOVERFLOW 1.e20
1.164 brouard 1382: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1383: #ifdef _WIN32
1384: #define DIRSEPARATOR '\\'
1385: #define CHARSEPARATOR "\\"
1386: #define ODIRSEPARATOR '/'
1387: #else
1.126 brouard 1388: #define DIRSEPARATOR '/'
1389: #define CHARSEPARATOR "/"
1390: #define ODIRSEPARATOR '\\'
1391: #endif
1392:
1.360 ! brouard 1393: /* $Id: imach.c,v 1.359 2024/04/24 21:21:17 brouard Exp $ */
1.126 brouard 1394: /* $State: Exp $ */
1.196 brouard 1395: #include "version.h"
1396: char version[]=__IMACH_VERSION__;
1.360 ! brouard 1397: 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";
! 1398: char fullversion[]="$Revision: 1.359 $ $Date: 2024/04/24 21:21:17 $";
1.126 brouard 1399: char strstart[80];
1400: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1401: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.342 brouard 1402: int debugILK=0; /* debugILK is set by a #d in a comment line */
1.187 brouard 1403: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 brouard 1404: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
1405: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
1.335 brouard 1406: 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 1407: 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 1408: int cptcovs=0; /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
1409: int cptcovsnq=0; /**< cptcovsnq number of SIMPLE covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1410: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1.349 brouard 1411: int cptcovprodage=0; /**< Number of fixed covariates with age: V3*age or V2*V3*age 1 */
1412: int cptcovprodvage=0; /**< Number of varying covariates with age: V7*age or V7*V6*age */
1413: int cptcovdageprod=0; /**< Number of doubleproducts with age, since 0.99r44 only: age*Vn*Vm for gnuplot printing*/
1.145 brouard 1414: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.335 brouard 1415: 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 1416: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1417: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.339 brouard 1418: int ncovvt=0; /* Total number of effective (wave) varying covariates (dummy or quantitative or products [without age]) in the model */
1.349 brouard 1419: 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 */
1420: 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 */
1421: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (single or product, dummy or quantitative) in the model */
1422: 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 1423: int nsd=0; /**< Total number of single dummy variables (output) */
1424: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1425: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1426: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1427: int ntveff=0; /**< ntveff number of effective time varying variables */
1428: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1429: int cptcov=0; /* Working variable */
1.334 brouard 1430: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs+1 declared globally ;*/
1.290 brouard 1431: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1432: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1433: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1434: int nlstate=2; /* Number of live states */
1435: int ndeath=1; /* Number of dead states */
1.130 brouard 1436: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.339 brouard 1437: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable*/
1438: int ncovcolt=0; /* ncovcolt=ncovcol+nqv+ntv+nqtv; total of covariates in the data, not in the model equation*/
1.126 brouard 1439: int popbased=0;
1440:
1441: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1442: int maxwav=0; /* Maxim number of waves */
1443: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1444: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1.359 brouard 1445: int gipmx = 0;
1446: double gsw = 0; /* Global variables on the number of contributions
1.126 brouard 1447: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1448: int mle=1, weightopt=0;
1.126 brouard 1449: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1450: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1451: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1452: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1453: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1454: int selected(int kvar); /* Is covariate kvar selected for printing results */
1455:
1.130 brouard 1456: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1457: double **matprod2(); /* test */
1.126 brouard 1458: double **oldm, **newm, **savm; /* Working pointers to matrices */
1459: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1460: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1461:
1.136 brouard 1462: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1463: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1464: FILE *ficlog, *ficrespow;
1.130 brouard 1465: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1466: double fretone; /* Only one call to likelihood */
1.130 brouard 1467: long ipmx=0; /* Number of contributions */
1.126 brouard 1468: double sw; /* Sum of weights */
1469: char filerespow[FILENAMELENGTH];
1470: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1471: FILE *ficresilk;
1472: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1473: FILE *ficresprobmorprev;
1474: FILE *fichtm, *fichtmcov; /* Html File */
1475: FILE *ficreseij;
1476: char filerese[FILENAMELENGTH];
1477: FILE *ficresstdeij;
1478: char fileresstde[FILENAMELENGTH];
1479: FILE *ficrescveij;
1480: char filerescve[FILENAMELENGTH];
1481: FILE *ficresvij;
1482: char fileresv[FILENAMELENGTH];
1.269 brouard 1483:
1.126 brouard 1484: char title[MAXLINE];
1.234 brouard 1485: char model[MAXLINE]; /**< The model line */
1.217 brouard 1486: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1487: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1488: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1489: char command[FILENAMELENGTH];
1490: int outcmd=0;
1491:
1.217 brouard 1492: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1493: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1494: char filelog[FILENAMELENGTH]; /* Log file */
1495: char filerest[FILENAMELENGTH];
1496: char fileregp[FILENAMELENGTH];
1497: char popfile[FILENAMELENGTH];
1498:
1499: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1500:
1.157 brouard 1501: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1502: /* struct timezone tzp; */
1503: /* extern int gettimeofday(); */
1504: struct tm tml, *gmtime(), *localtime();
1505:
1506: extern time_t time();
1507:
1508: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1509: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1.349 brouard 1510: time_t rlast_btime; /* raw time */
1.157 brouard 1511: struct tm tm;
1512:
1.126 brouard 1513: char strcurr[80], strfor[80];
1514:
1515: char *endptr;
1516: long lval;
1517: double dval;
1518:
1519: #define NR_END 1
1520: #define FREE_ARG char*
1521: #define FTOL 1.0e-10
1522:
1523: #define NRANSI
1.240 brouard 1524: #define ITMAX 200
1525: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1526:
1527: #define TOL 2.0e-4
1528:
1529: #define CGOLD 0.3819660
1530: #define ZEPS 1.0e-10
1531: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1532:
1533: #define GOLD 1.618034
1534: #define GLIMIT 100.0
1535: #define TINY 1.0e-20
1536:
1537: static double maxarg1,maxarg2;
1538: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1539: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1540:
1541: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1542: #define rint(a) floor(a+0.5)
1.166 brouard 1543: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1544: #define mytinydouble 1.0e-16
1.166 brouard 1545: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1546: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1547: /* static double dsqrarg; */
1548: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1549: static double sqrarg;
1550: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1551: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1552: int agegomp= AGEGOMP;
1553:
1554: int imx;
1555: int stepm=1;
1556: /* Stepm, step in month: minimum step interpolation*/
1557:
1558: int estepm;
1559: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1560:
1561: int m,nb;
1562: long *num;
1.197 brouard 1563: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1564: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1565: covariate for which somebody answered excluding
1566: undefined. Usually 2: 0 and 1. */
1567: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1568: covariate for which somebody answered including
1569: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1570: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1571: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1572: double ***mobaverage, ***mobaverages; /* New global variable */
1.332 brouard 1573: 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 1574: double *ageexmed,*agecens;
1575: double dateintmean=0;
1.296 brouard 1576: double anprojd, mprojd, jprojd; /* For eventual projections */
1577: double anprojf, mprojf, jprojf;
1.126 brouard 1578:
1.296 brouard 1579: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1580: double anbackf, mbackf, jbackf;
1581: double jintmean,mintmean,aintmean;
1.126 brouard 1582: double *weight;
1583: int **s; /* Status */
1.141 brouard 1584: double *agedc;
1.145 brouard 1585: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1586: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1587: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1588: double **coqvar; /* Fixed quantitative covariate nqv */
1.341 brouard 1589: double ***cotvar; /* Time varying covariate start at ncovcol + nqv + (1 to ntv) */
1.225 brouard 1590: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1591: double idx;
1592: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1593: /* Some documentation */
1594: /* Design original data
1595: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1596: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1597: * ntv=3 nqtv=1
1.330 brouard 1598: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1599: * For time varying covariate, quanti or dummies
1600: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1.341 brouard 1601: * cotvar[wav][ncovcol+nqv+ iv(1 to nqtv)][i]= [(1 to nqtv)][i]=(V12) quanti
1.319 brouard 1602: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1603: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1.332 brouard 1604: * covar[Vk,i], value of the Vkth fixed covariate dummy or quanti for individual i:
1.319 brouard 1605: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1606: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1607: * k= 1 2 3 4 5 6 7 8 9 10 11
1608: */
1609: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1610: /* 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
1611: # States 1=Coresidence, 2 Living alone, 3 Institution
1612: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1613: */
1.349 brouard 1614: /* V5+V4+ V3+V4*V3 +V5*age+V2 +V1*V2+V1*age+V1+V4*V3*age */
1615: /* kmodel 1 2 3 4 5 6 7 8 9 10 */
1616: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 3 *//*0 for simple covariate (dummy, quantitative,*/
1617: /* fixed or varying), 1 for age product, 2 for*/
1618: /* product without age, 3 for age and double product */
1619: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 3 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1620: /*(single or product without age), 2 dummy*/
1621: /* with age product, 3 quant with age product*/
1622: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 6 */
1623: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1624: /*TnsdVar[Tvar] 1 2 3 */
1625: /*Tvaraff[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1626: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1627: /*TvarsDind[nsd] 2 3 9 */ /* position K of single dummy cova */
1628: /* nsq 1 2 */ /* Counting single quantit tv */
1629: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1630: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1631: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1632: /* cptcovage 1 2 3 */ /* Counting cov*age in the model equation */
1633: /* Tage[cptcovage]=k 5 8 10 */ /* Position in the model of ith cov*age */
1.350 brouard 1634: /* 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"*/
1635: /* 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 1636: /* p Tvard[2][1]@21 = {7, 2, 6, 3, 7, 3, 6, 4, 7, 4, 0 <repeats 11 times>} */
1.350 brouard 1637: /* 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}*/
1638: /* 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 1639: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 brouard 1640: /* 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 1641: /* 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 1642: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1643: /* Type */
1644: /* V 1 2 3 4 5 */
1645: /* F F V V V */
1646: /* D Q D D Q */
1647: /* */
1648: int *TvarsD;
1.330 brouard 1649: int *TnsdVar;
1.234 brouard 1650: int *TvarsDind;
1651: int *TvarsQ;
1652: int *TvarsQind;
1653:
1.318 brouard 1654: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1655: int nresult=0;
1.258 brouard 1656: int parameterline=0; /* # of the parameter (type) line */
1.334 brouard 1657: int TKresult[MAXRESULTLINESPONE]; /* TKresult[nres]=k for each resultline nres give the corresponding combination of dummies */
1658: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model corresponds to the k3 position in the resultline */
1659: int modelresult[MAXRESULTLINESPONE][NCOVMAX];/* modelresult[k3]=k1: k1th position in the model corresponds to the k3 position in the resultline */
1660: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1661: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1662: double TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.334 brouard 1663: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1664: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.318 brouard 1665: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1.332 brouard 1666: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.318 brouard 1667:
1668: /* 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
1669: # States 1=Coresidence, 2 Living alone, 3 Institution
1670: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1671: */
1.234 brouard 1672: /* 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 1673: 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 */
1674: 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 */
1675: 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 */
1676: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1677: 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 */
1678: 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 1679: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1680: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1681: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1682: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1683: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1684: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1685: 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 */
1686: 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 1687: int *TvarVV; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1688: int *TvarVVind; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1.349 brouard 1689: int *TvarVVA; /* We count ncovvt time varying covariates (single or products with age) and put their name into TvarVVA */
1690: int *TvarVVAind; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1691: int *TvarAVVA; /* We count ALL ncovta time varying covariates (single or products with age) and put their name into TvarVVA */
1692: int *TvarAVVAind; /* We count ALL ncovta time varying covariates (single or products without age) and put their name into TvarVV */
1.339 brouard 1693: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
1.349 brouard 1694: /* model V1+V3+age*V1+age*V3+V1*V3 + V1*V3*age */
1695: /* Tvar={1, 3, 1, 3, 6, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
1696: /* TvarVV={3,1,3,1,3}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1697: /* 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 1698: int *Tvarsel; /**< Selected covariates for output */
1699: double *Tvalsel; /**< Selected modality value of covariate for output */
1.349 brouard 1700: 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 1701: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1702: 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 1703: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1704: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1705: int *Tage;
1.227 brouard 1706: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1707: 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 1708: 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*/
1709: 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 1710: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1711: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1712: int **Tvard;
1.330 brouard 1713: int **Tvardk;
1.227 brouard 1714: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1715: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1716: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1717: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1718: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1719: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1720: double *lsurv, *lpop, *tpop;
1721:
1.231 brouard 1722: #define FD 1; /* Fixed dummy covariate */
1723: #define FQ 2; /* Fixed quantitative covariate */
1724: #define FP 3; /* Fixed product covariate */
1725: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1726: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1727: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1728: #define VD 10; /* Varying dummy covariate */
1729: #define VQ 11; /* Varying quantitative covariate */
1730: #define VP 12; /* Varying product covariate */
1731: #define VPDD 13; /* Varying product dummy*dummy covariate */
1732: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1733: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1734: #define APFD 16; /* Age product * fixed dummy covariate */
1735: #define APFQ 17; /* Age product * fixed quantitative covariate */
1736: #define APVD 18; /* Age product * varying dummy covariate */
1737: #define APVQ 19; /* Age product * varying quantitative covariate */
1738:
1739: #define FTYPE 1; /* Fixed covariate */
1740: #define VTYPE 2; /* Varying covariate (loop in wave) */
1741: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1742:
1743: struct kmodel{
1744: int maintype; /* main type */
1745: int subtype; /* subtype */
1746: };
1747: struct kmodel modell[NCOVMAX];
1748:
1.143 brouard 1749: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1750: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1751:
1752: /**************** split *************************/
1753: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1754: {
1755: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1756: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1757: */
1758: char *ss; /* pointer */
1.186 brouard 1759: int l1=0, l2=0; /* length counters */
1.126 brouard 1760:
1761: l1 = strlen(path ); /* length of path */
1762: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1763: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1764: if ( ss == NULL ) { /* no directory, so determine current directory */
1765: strcpy( name, path ); /* we got the fullname name because no directory */
1766: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1767: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1768: /* get current working directory */
1769: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1770: #ifdef WIN32
1771: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1772: #else
1773: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1774: #endif
1.126 brouard 1775: return( GLOCK_ERROR_GETCWD );
1776: }
1777: /* got dirc from getcwd*/
1778: printf(" DIRC = %s \n",dirc);
1.205 brouard 1779: } else { /* strip directory from path */
1.126 brouard 1780: ss++; /* after this, the filename */
1781: l2 = strlen( ss ); /* length of filename */
1782: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1783: strcpy( name, ss ); /* save file name */
1784: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1785: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1786: printf(" DIRC2 = %s \n",dirc);
1787: }
1788: /* We add a separator at the end of dirc if not exists */
1789: l1 = strlen( dirc ); /* length of directory */
1790: if( dirc[l1-1] != DIRSEPARATOR ){
1791: dirc[l1] = DIRSEPARATOR;
1792: dirc[l1+1] = 0;
1793: printf(" DIRC3 = %s \n",dirc);
1794: }
1795: ss = strrchr( name, '.' ); /* find last / */
1796: if (ss >0){
1797: ss++;
1798: strcpy(ext,ss); /* save extension */
1799: l1= strlen( name);
1800: l2= strlen(ss)+1;
1801: strncpy( finame, name, l1-l2);
1802: finame[l1-l2]= 0;
1803: }
1804:
1805: return( 0 ); /* we're done */
1806: }
1807:
1808:
1809: /******************************************/
1810:
1811: void replace_back_to_slash(char *s, char*t)
1812: {
1813: int i;
1814: int lg=0;
1815: i=0;
1816: lg=strlen(t);
1817: for(i=0; i<= lg; i++) {
1818: (s[i] = t[i]);
1819: if (t[i]== '\\') s[i]='/';
1820: }
1821: }
1822:
1.132 brouard 1823: char *trimbb(char *out, char *in)
1.137 brouard 1824: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1825: char *s;
1826: s=out;
1827: while (*in != '\0'){
1.137 brouard 1828: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1829: in++;
1830: }
1831: *out++ = *in++;
1832: }
1833: *out='\0';
1834: return s;
1835: }
1836:
1.351 brouard 1837: char *trimbtab(char *out, char *in)
1838: { /* Trim blanks or tabs in line but keeps first blanks if line starts with blanks */
1839: char *s;
1840: s=out;
1841: while (*in != '\0'){
1842: while( (*in == ' ' || *in == '\t')){ /* && *(in+1) != '\0'){*/
1843: in++;
1844: }
1845: *out++ = *in++;
1846: }
1847: *out='\0';
1848: return s;
1849: }
1850:
1.187 brouard 1851: /* char *substrchaine(char *out, char *in, char *chain) */
1852: /* { */
1853: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1854: /* char *s, *t; */
1855: /* t=in;s=out; */
1856: /* while ((*in != *chain) && (*in != '\0')){ */
1857: /* *out++ = *in++; */
1858: /* } */
1859:
1860: /* /\* *in matches *chain *\/ */
1861: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1862: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1863: /* } */
1864: /* in--; chain--; */
1865: /* while ( (*in != '\0')){ */
1866: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1867: /* *out++ = *in++; */
1868: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1869: /* } */
1870: /* *out='\0'; */
1871: /* out=s; */
1872: /* return out; */
1873: /* } */
1874: char *substrchaine(char *out, char *in, char *chain)
1875: {
1876: /* Substract chain 'chain' from 'in', return and output 'out' */
1.349 brouard 1877: /* in="V1+V1*age+age*age+V2", chain="+age*age" out="V1+V1*age+V2" */
1.187 brouard 1878:
1879: char *strloc;
1880:
1.349 brouard 1881: strcpy (out, in); /* out="V1+V1*age+age*age+V2" */
1882: strloc = strstr(out, chain); /* strloc points to out at "+age*age+V2" */
1883: 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 1884: if(strloc != NULL){
1.349 brouard 1885: /* will affect out */ /* strloc+strlen(chain)=|+V2 = "V1+V1*age+age*age|+V2" */ /* Will also work in Unicodek */
1886: 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)*/
1887: /* equivalent to strcpy (strloc, strloc +strlen(chain)) if no overlap; Copies from "+V2" to V1+V1*age+ */
1.187 brouard 1888: }
1.349 brouard 1889: 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 1890: return out;
1891: }
1892:
1893:
1.145 brouard 1894: char *cutl(char *blocc, char *alocc, char *in, char occ)
1895: {
1.187 brouard 1896: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.349 brouard 1897: and alocc starts after first occurence of char 'occ' : ex cutl(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1898: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1899: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1900: */
1.160 brouard 1901: char *s, *t;
1.145 brouard 1902: t=in;s=in;
1903: while ((*in != occ) && (*in != '\0')){
1904: *alocc++ = *in++;
1905: }
1906: if( *in == occ){
1907: *(alocc)='\0';
1908: s=++in;
1909: }
1910:
1911: if (s == t) {/* occ not found */
1912: *(alocc-(in-s))='\0';
1913: in=s;
1914: }
1915: while ( *in != '\0'){
1916: *blocc++ = *in++;
1917: }
1918:
1919: *blocc='\0';
1920: return t;
1921: }
1.137 brouard 1922: char *cutv(char *blocc, char *alocc, char *in, char occ)
1923: {
1.187 brouard 1924: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1925: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1926: gives blocc="abcdef2ghi" and alocc="j".
1927: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1928: */
1929: char *s, *t;
1930: t=in;s=in;
1931: while (*in != '\0'){
1932: while( *in == occ){
1933: *blocc++ = *in++;
1934: s=in;
1935: }
1936: *blocc++ = *in++;
1937: }
1938: if (s == t) /* occ not found */
1939: *(blocc-(in-s))='\0';
1940: else
1941: *(blocc-(in-s)-1)='\0';
1942: in=s;
1943: while ( *in != '\0'){
1944: *alocc++ = *in++;
1945: }
1946:
1947: *alocc='\0';
1948: return s;
1949: }
1950:
1.126 brouard 1951: int nbocc(char *s, char occ)
1952: {
1953: int i,j=0;
1954: int lg=20;
1955: i=0;
1956: lg=strlen(s);
1957: for(i=0; i<= lg; i++) {
1.234 brouard 1958: if (s[i] == occ ) j++;
1.126 brouard 1959: }
1960: return j;
1961: }
1962:
1.349 brouard 1963: int nboccstr(char *textin, char *chain)
1964: {
1965: /* Counts the number of occurence of "chain" in string textin */
1966: /* in="+V7*V4+age*V2+age*V3+age*V4" chain="age" */
1967: char *strloc;
1968:
1969: int i,j=0;
1970:
1971: i=0;
1972:
1973: strloc=textin; /* strloc points to "^+V7*V4+age+..." in textin */
1974: for(;;) {
1975: strloc= strstr(strloc,chain); /* strloc points to first character of chain in textin if found. Example strloc points^ to "+V7*V4+^age" in textin */
1976: if(strloc != NULL){
1977: strloc = strloc+strlen(chain); /* strloc points to "+V7*V4+age^" in textin */
1978: j++;
1979: }else
1980: break;
1981: }
1982: return j;
1983:
1984: }
1.137 brouard 1985: /* void cutv(char *u,char *v, char*t, char occ) */
1986: /* { */
1987: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1988: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1989: /* gives u="abcdef2ghi" and v="j" *\/ */
1990: /* int i,lg,j,p=0; */
1991: /* i=0; */
1992: /* lg=strlen(t); */
1993: /* for(j=0; j<=lg-1; j++) { */
1994: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1995: /* } */
1.126 brouard 1996:
1.137 brouard 1997: /* for(j=0; j<p; j++) { */
1998: /* (u[j] = t[j]); */
1999: /* } */
2000: /* u[p]='\0'; */
1.126 brouard 2001:
1.137 brouard 2002: /* for(j=0; j<= lg; j++) { */
2003: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
2004: /* } */
2005: /* } */
1.126 brouard 2006:
1.160 brouard 2007: #ifdef _WIN32
2008: char * strsep(char **pp, const char *delim)
2009: {
2010: char *p, *q;
2011:
2012: if ((p = *pp) == NULL)
2013: return 0;
2014: if ((q = strpbrk (p, delim)) != NULL)
2015: {
2016: *pp = q + 1;
2017: *q = '\0';
2018: }
2019: else
2020: *pp = 0;
2021: return p;
2022: }
2023: #endif
2024:
1.126 brouard 2025: /********************** nrerror ********************/
2026:
2027: void nrerror(char error_text[])
2028: {
2029: fprintf(stderr,"ERREUR ...\n");
2030: fprintf(stderr,"%s\n",error_text);
2031: exit(EXIT_FAILURE);
2032: }
2033: /*********************** vector *******************/
2034: double *vector(int nl, int nh)
2035: {
2036: double *v;
2037: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
2038: if (!v) nrerror("allocation failure in vector");
2039: return v-nl+NR_END;
2040: }
2041:
2042: /************************ free vector ******************/
2043: void free_vector(double*v, int nl, int nh)
2044: {
2045: free((FREE_ARG)(v+nl-NR_END));
2046: }
2047:
2048: /************************ivector *******************************/
2049: int *ivector(long nl,long nh)
2050: {
2051: int *v;
2052: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
2053: if (!v) nrerror("allocation failure in ivector");
2054: return v-nl+NR_END;
2055: }
2056:
2057: /******************free ivector **************************/
2058: void free_ivector(int *v, long nl, long nh)
2059: {
2060: free((FREE_ARG)(v+nl-NR_END));
2061: }
2062:
2063: /************************lvector *******************************/
2064: long *lvector(long nl,long nh)
2065: {
2066: long *v;
2067: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
2068: if (!v) nrerror("allocation failure in ivector");
2069: return v-nl+NR_END;
2070: }
2071:
2072: /******************free lvector **************************/
2073: void free_lvector(long *v, long nl, long nh)
2074: {
2075: free((FREE_ARG)(v+nl-NR_END));
2076: }
2077:
2078: /******************* imatrix *******************************/
2079: int **imatrix(long nrl, long nrh, long ncl, long nch)
2080: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
2081: {
2082: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
2083: int **m;
2084:
2085: /* allocate pointers to rows */
2086: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
2087: if (!m) nrerror("allocation failure 1 in matrix()");
2088: m += NR_END;
2089: m -= nrl;
2090:
2091:
2092: /* allocate rows and set pointers to them */
2093: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
2094: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2095: m[nrl] += NR_END;
2096: m[nrl] -= ncl;
2097:
2098: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
2099:
2100: /* return pointer to array of pointers to rows */
2101: return m;
2102: }
2103:
2104: /****************** free_imatrix *************************/
2105: void free_imatrix(m,nrl,nrh,ncl,nch)
2106: int **m;
2107: long nch,ncl,nrh,nrl;
2108: /* free an int matrix allocated by imatrix() */
2109: {
2110: free((FREE_ARG) (m[nrl]+ncl-NR_END));
2111: free((FREE_ARG) (m+nrl-NR_END));
2112: }
2113:
2114: /******************* matrix *******************************/
2115: double **matrix(long nrl, long nrh, long ncl, long nch)
2116: {
2117: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
2118: double **m;
2119:
2120: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2121: if (!m) nrerror("allocation failure 1 in matrix()");
2122: m += NR_END;
2123: m -= nrl;
2124:
2125: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2126: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2127: m[nrl] += NR_END;
2128: m[nrl] -= ncl;
2129:
2130: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2131: return m;
1.145 brouard 2132: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
2133: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
2134: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 2135: */
2136: }
2137:
2138: /*************************free matrix ************************/
2139: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
2140: {
2141: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2142: free((FREE_ARG)(m+nrl-NR_END));
2143: }
2144:
2145: /******************* ma3x *******************************/
2146: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
2147: {
2148: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
2149: double ***m;
2150:
2151: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2152: if (!m) nrerror("allocation failure 1 in matrix()");
2153: m += NR_END;
2154: m -= nrl;
2155:
2156: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2157: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2158: m[nrl] += NR_END;
2159: m[nrl] -= ncl;
2160:
2161: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2162:
2163: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
2164: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
2165: m[nrl][ncl] += NR_END;
2166: m[nrl][ncl] -= nll;
2167: for (j=ncl+1; j<=nch; j++)
2168: m[nrl][j]=m[nrl][j-1]+nlay;
2169:
2170: for (i=nrl+1; i<=nrh; i++) {
2171: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
2172: for (j=ncl+1; j<=nch; j++)
2173: m[i][j]=m[i][j-1]+nlay;
2174: }
2175: return m;
2176: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
2177: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
2178: */
2179: }
2180:
2181: /*************************free ma3x ************************/
2182: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
2183: {
2184: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
2185: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2186: free((FREE_ARG)(m+nrl-NR_END));
2187: }
2188:
2189: /*************** function subdirf ***********/
2190: char *subdirf(char fileres[])
2191: {
2192: /* Caution optionfilefiname is hidden */
2193: strcpy(tmpout,optionfilefiname);
2194: strcat(tmpout,"/"); /* Add to the right */
2195: strcat(tmpout,fileres);
2196: return tmpout;
2197: }
2198:
2199: /*************** function subdirf2 ***********/
2200: char *subdirf2(char fileres[], char *preop)
2201: {
1.314 brouard 2202: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
2203: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 2204: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 2205: /* Caution optionfilefiname is hidden */
2206: strcpy(tmpout,optionfilefiname);
2207: strcat(tmpout,"/");
2208: strcat(tmpout,preop);
2209: strcat(tmpout,fileres);
2210: return tmpout;
2211: }
2212:
2213: /*************** function subdirf3 ***********/
2214: char *subdirf3(char fileres[], char *preop, char *preop2)
2215: {
2216:
2217: /* Caution optionfilefiname is hidden */
2218: strcpy(tmpout,optionfilefiname);
2219: strcat(tmpout,"/");
2220: strcat(tmpout,preop);
2221: strcat(tmpout,preop2);
2222: strcat(tmpout,fileres);
2223: return tmpout;
2224: }
1.213 brouard 2225:
2226: /*************** function subdirfext ***********/
2227: char *subdirfext(char fileres[], char *preop, char *postop)
2228: {
2229:
2230: strcpy(tmpout,preop);
2231: strcat(tmpout,fileres);
2232: strcat(tmpout,postop);
2233: return tmpout;
2234: }
1.126 brouard 2235:
1.213 brouard 2236: /*************** function subdirfext3 ***********/
2237: char *subdirfext3(char fileres[], char *preop, char *postop)
2238: {
2239:
2240: /* Caution optionfilefiname is hidden */
2241: strcpy(tmpout,optionfilefiname);
2242: strcat(tmpout,"/");
2243: strcat(tmpout,preop);
2244: strcat(tmpout,fileres);
2245: strcat(tmpout,postop);
2246: return tmpout;
2247: }
2248:
1.162 brouard 2249: char *asc_diff_time(long time_sec, char ascdiff[])
2250: {
2251: long sec_left, days, hours, minutes;
2252: days = (time_sec) / (60*60*24);
2253: sec_left = (time_sec) % (60*60*24);
2254: hours = (sec_left) / (60*60) ;
2255: sec_left = (sec_left) %(60*60);
2256: minutes = (sec_left) /60;
2257: sec_left = (sec_left) % (60);
2258: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2259: return ascdiff;
2260: }
2261:
1.126 brouard 2262: /***************** f1dim *************************/
2263: extern int ncom;
2264: extern double *pcom,*xicom;
2265: extern double (*nrfunc)(double []);
2266:
2267: double f1dim(double x)
2268: {
2269: int j;
2270: double f;
2271: double *xt;
2272:
2273: xt=vector(1,ncom);
2274: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2275: f=(*nrfunc)(xt);
2276: free_vector(xt,1,ncom);
2277: return f;
2278: }
2279:
2280: /*****************brent *************************/
2281: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2282: {
2283: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2284: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2285: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2286: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2287: * returned function value.
2288: */
1.126 brouard 2289: int iter;
2290: double a,b,d,etemp;
1.159 brouard 2291: double fu=0,fv,fw,fx;
1.164 brouard 2292: double ftemp=0.;
1.126 brouard 2293: double p,q,r,tol1,tol2,u,v,w,x,xm;
2294: double e=0.0;
2295:
2296: a=(ax < cx ? ax : cx);
2297: b=(ax > cx ? ax : cx);
2298: x=w=v=bx;
2299: fw=fv=fx=(*f)(x);
2300: for (iter=1;iter<=ITMAX;iter++) {
2301: xm=0.5*(a+b);
2302: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2303: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2304: printf(".");fflush(stdout);
2305: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2306: #ifdef DEBUGBRENT
1.126 brouard 2307: 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);
2308: 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);
2309: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2310: #endif
2311: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2312: *xmin=x;
2313: return fx;
2314: }
2315: ftemp=fu;
2316: if (fabs(e) > tol1) {
2317: r=(x-w)*(fx-fv);
2318: q=(x-v)*(fx-fw);
2319: p=(x-v)*q-(x-w)*r;
2320: q=2.0*(q-r);
2321: if (q > 0.0) p = -p;
2322: q=fabs(q);
2323: etemp=e;
2324: e=d;
2325: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2326: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2327: else {
1.224 brouard 2328: d=p/q;
2329: u=x+d;
2330: if (u-a < tol2 || b-u < tol2)
2331: d=SIGN(tol1,xm-x);
1.126 brouard 2332: }
2333: } else {
2334: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2335: }
2336: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2337: fu=(*f)(u);
2338: if (fu <= fx) {
2339: if (u >= x) a=x; else b=x;
2340: SHFT(v,w,x,u)
1.183 brouard 2341: SHFT(fv,fw,fx,fu)
2342: } else {
2343: if (u < x) a=u; else b=u;
2344: if (fu <= fw || w == x) {
1.224 brouard 2345: v=w;
2346: w=u;
2347: fv=fw;
2348: fw=fu;
1.183 brouard 2349: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2350: v=u;
2351: fv=fu;
1.183 brouard 2352: }
2353: }
1.126 brouard 2354: }
2355: nrerror("Too many iterations in brent");
2356: *xmin=x;
2357: return fx;
2358: }
2359:
2360: /****************** mnbrak ***********************/
2361:
2362: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2363: double (*func)(double))
1.183 brouard 2364: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2365: the downhill direction (defined by the function as evaluated at the initial points) and returns
2366: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2367: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2368: */
1.126 brouard 2369: double ulim,u,r,q, dum;
2370: double fu;
1.187 brouard 2371:
2372: double scale=10.;
2373: int iterscale=0;
2374:
2375: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2376: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2377:
2378:
2379: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2380: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2381: /* *bx = *ax - (*ax - *bx)/scale; */
2382: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2383: /* } */
2384:
1.126 brouard 2385: if (*fb > *fa) {
2386: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2387: SHFT(dum,*fb,*fa,dum)
2388: }
1.126 brouard 2389: *cx=(*bx)+GOLD*(*bx-*ax);
2390: *fc=(*func)(*cx);
1.183 brouard 2391: #ifdef DEBUG
1.224 brouard 2392: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2393: 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 2394: #endif
1.224 brouard 2395: 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 2396: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2397: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2398: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2399: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2400: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2401: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2402: fu=(*func)(u);
1.163 brouard 2403: #ifdef DEBUG
2404: /* f(x)=A(x-u)**2+f(u) */
2405: double A, fparabu;
2406: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2407: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2408: 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);
2409: 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 2410: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2411: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2412: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2413: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2414: #endif
1.184 brouard 2415: #ifdef MNBRAKORIGINAL
1.183 brouard 2416: #else
1.191 brouard 2417: /* if (fu > *fc) { */
2418: /* #ifdef DEBUG */
2419: /* printf("mnbrak4 fu > fc \n"); */
2420: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2421: /* #endif */
2422: /* /\* 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 *\\/ *\/ */
2423: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2424: /* dum=u; /\* Shifting c and u *\/ */
2425: /* u = *cx; */
2426: /* *cx = dum; */
2427: /* dum = fu; */
2428: /* fu = *fc; */
2429: /* *fc =dum; */
2430: /* } else { /\* end *\/ */
2431: /* #ifdef DEBUG */
2432: /* printf("mnbrak3 fu < fc \n"); */
2433: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2434: /* #endif */
2435: /* dum=u; /\* Shifting c and u *\/ */
2436: /* u = *cx; */
2437: /* *cx = dum; */
2438: /* dum = fu; */
2439: /* fu = *fc; */
2440: /* *fc =dum; */
2441: /* } */
1.224 brouard 2442: #ifdef DEBUGMNBRAK
2443: double A, fparabu;
2444: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2445: fparabu= *fa - A*(*ax-u)*(*ax-u);
2446: 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);
2447: 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 2448: #endif
1.191 brouard 2449: dum=u; /* Shifting c and u */
2450: u = *cx;
2451: *cx = dum;
2452: dum = fu;
2453: fu = *fc;
2454: *fc =dum;
1.183 brouard 2455: #endif
1.162 brouard 2456: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2457: #ifdef DEBUG
1.224 brouard 2458: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2459: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2460: #endif
1.126 brouard 2461: fu=(*func)(u);
2462: if (fu < *fc) {
1.183 brouard 2463: #ifdef DEBUG
1.224 brouard 2464: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2465: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2466: #endif
2467: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2468: SHFT(*fb,*fc,fu,(*func)(u))
2469: #ifdef DEBUG
2470: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2471: #endif
2472: }
1.162 brouard 2473: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2474: #ifdef DEBUG
1.224 brouard 2475: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2476: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2477: #endif
1.126 brouard 2478: u=ulim;
2479: fu=(*func)(u);
1.183 brouard 2480: } else { /* u could be left to b (if r > q parabola has a maximum) */
2481: #ifdef DEBUG
1.224 brouard 2482: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2483: 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 2484: #endif
1.126 brouard 2485: u=(*cx)+GOLD*(*cx-*bx);
2486: fu=(*func)(u);
1.224 brouard 2487: #ifdef DEBUG
2488: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2489: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2490: #endif
1.183 brouard 2491: } /* end tests */
1.126 brouard 2492: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2493: SHFT(*fa,*fb,*fc,fu)
2494: #ifdef DEBUG
1.224 brouard 2495: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2496: 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 2497: #endif
2498: } /* 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 2499: }
2500:
2501: /*************** linmin ************************/
1.162 brouard 2502: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2503: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2504: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2505: the value of func at the returned location p . This is actually all accomplished by calling the
2506: routines mnbrak and brent .*/
1.126 brouard 2507: int ncom;
2508: double *pcom,*xicom;
2509: double (*nrfunc)(double []);
2510:
1.224 brouard 2511: #ifdef LINMINORIGINAL
1.126 brouard 2512: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2513: #else
2514: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2515: #endif
1.126 brouard 2516: {
2517: double brent(double ax, double bx, double cx,
2518: double (*f)(double), double tol, double *xmin);
2519: double f1dim(double x);
2520: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2521: double *fc, double (*func)(double));
2522: int j;
2523: double xx,xmin,bx,ax;
2524: double fx,fb,fa;
1.187 brouard 2525:
1.203 brouard 2526: #ifdef LINMINORIGINAL
2527: #else
2528: double scale=10., axs, xxs; /* Scale added for infinity */
2529: #endif
2530:
1.126 brouard 2531: ncom=n;
2532: pcom=vector(1,n);
2533: xicom=vector(1,n);
2534: nrfunc=func;
2535: for (j=1;j<=n;j++) {
2536: pcom[j]=p[j];
1.202 brouard 2537: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2538: }
1.187 brouard 2539:
1.203 brouard 2540: #ifdef LINMINORIGINAL
2541: xx=1.;
2542: #else
2543: axs=0.0;
2544: xxs=1.;
2545: do{
2546: xx= xxs;
2547: #endif
1.187 brouard 2548: ax=0.;
2549: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2550: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2551: /* 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)) */
2552: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2553: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2554: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2555: /* 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 2556: #ifdef LINMINORIGINAL
2557: #else
2558: if (fx != fx){
1.224 brouard 2559: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2560: printf("|");
2561: fprintf(ficlog,"|");
1.203 brouard 2562: #ifdef DEBUGLINMIN
1.224 brouard 2563: 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 2564: #endif
2565: }
1.224 brouard 2566: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2567: #endif
2568:
1.191 brouard 2569: #ifdef DEBUGLINMIN
2570: 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 2571: 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 2572: #endif
1.224 brouard 2573: #ifdef LINMINORIGINAL
2574: #else
1.317 brouard 2575: if(fb == fx){ /* Flat function in the direction */
2576: xmin=xx;
1.224 brouard 2577: *flat=1;
1.317 brouard 2578: }else{
1.224 brouard 2579: *flat=0;
2580: #endif
2581: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2582: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2583: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2584: /* fmin = f(p[j] + xmin * xi[j]) */
2585: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2586: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2587: #ifdef DEBUG
1.224 brouard 2588: 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);
2589: 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);
2590: #endif
2591: #ifdef LINMINORIGINAL
2592: #else
2593: }
1.126 brouard 2594: #endif
1.191 brouard 2595: #ifdef DEBUGLINMIN
2596: printf("linmin end ");
1.202 brouard 2597: fprintf(ficlog,"linmin end ");
1.191 brouard 2598: #endif
1.126 brouard 2599: for (j=1;j<=n;j++) {
1.203 brouard 2600: #ifdef LINMINORIGINAL
2601: xi[j] *= xmin;
2602: #else
2603: #ifdef DEBUGLINMIN
2604: if(xxs <1.0)
2605: printf(" before xi[%d]=%12.8f", j,xi[j]);
2606: #endif
2607: 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) */
2608: #ifdef DEBUGLINMIN
2609: if(xxs <1.0)
2610: 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 );
2611: #endif
2612: #endif
1.187 brouard 2613: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2614: }
1.191 brouard 2615: #ifdef DEBUGLINMIN
1.203 brouard 2616: printf("\n");
1.191 brouard 2617: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2618: 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 2619: for (j=1;j<=n;j++) {
1.202 brouard 2620: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2621: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2622: if(j % ncovmodel == 0){
1.191 brouard 2623: printf("\n");
1.202 brouard 2624: fprintf(ficlog,"\n");
2625: }
1.191 brouard 2626: }
1.203 brouard 2627: #else
1.191 brouard 2628: #endif
1.126 brouard 2629: free_vector(xicom,1,n);
2630: free_vector(pcom,1,n);
2631: }
2632:
1.359 brouard 2633: /**** praxis gegen ****/
2634:
2635: /* This has been tested by Visual C from Microsoft and works */
2636: /* meaning tha valgrind could be wrong */
2637: /*********************************************************************/
2638: /* f u n c t i o n p r a x i s */
2639: /* */
2640: /* praxis is a general purpose routine for the minimization of a */
2641: /* function in several variables. the algorithm used is a modifi- */
2642: /* cation of conjugate gradient search method by powell. the changes */
2643: /* are due to r.p. brent, who gives an algol-w program, which served */
2644: /* as a basis for this function. */
2645: /* */
2646: /* references: */
2647: /* - powell, m.j.d., 1964. an efficient method for finding */
2648: /* the minimum of a function in several variables without */
2649: /* calculating derivatives, computer journal, 7, 155-162 */
2650: /* - brent, r.p., 1973. algorithms for minimization without */
2651: /* derivatives, prentice hall, englewood cliffs. */
2652: /* */
2653: /* problems, suggestions or improvements are always wellcome */
2654: /* karl gegenfurtner 07/08/87 */
2655: /* c - version */
2656: /*********************************************************************/
2657: /* */
2658: /* usage: min = praxis(tol, macheps, h, n, prin, x, func) */
2659: /* macheps has been suppressed because it is replaced by DBL_EPSILON */
2660: /* and if it was an argument of praxis (as it is in original brent) */
2661: /* it should be declared external */
2662: /* usage: min = praxis(tol, h, n, prin, x, func) */
2663: /* was min = praxis(fun, x, n); */
2664: /* */
2665: /* fun the function to be minimized. fun is called from */
2666: /* praxis with x and n as arguments */
2667: /* x a double array containing the initial guesses for */
2668: /* the minimum, which will contain the solution on */
2669: /* return */
2670: /* n an integer specifying the number of unknown */
2671: /* parameters */
2672: /* min praxis returns the least calculated value of fun */
2673: /* */
2674: /* some additional global variables control some more aspects of */
2675: /* the inner workings of praxis. setting them is optional, they */
2676: /* are all set to some reasonable default values given below. */
2677: /* */
2678: /* prin controls the printed output from the routine. */
2679: /* 0 -> no output */
2680: /* 1 -> print only starting and final values */
2681: /* 2 -> detailed map of the minimization process */
2682: /* 3 -> print also eigenvalues and vectors of the */
2683: /* search directions */
2684: /* the default value is 1 */
2685: /* tol is the tolerance allowed for the precision of the */
2686: /* solution. praxis returns if the criterion */
2687: /* 2 * ||x[k]-x[k-1]|| <= sqrt(macheps) * ||x[k]|| + tol */
2688: /* is fulfilled more than ktm times. */
2689: /* the default value depends on the machine precision */
2690: /* ktm see just above. default is 1, and a value of 4 leads */
2691: /* to a very(!) cautious stopping criterion. */
2692: /* h0 or step is a steplength parameter and should be set equal */
2693: /* to the expected distance from the solution. */
2694: /* exceptionally small or large values of step lead to */
2695: /* slower convergence on the first few iterations */
2696: /* the default value for step is 1.0 */
2697: /* scbd is a scaling parameter. 1.0 is the default and */
2698: /* indicates no scaling. if the scales for the different */
2699: /* parameters are very different, scbd should be set to */
2700: /* a value of about 10.0. */
2701: /* illc should be set to true (1) if the problem is known to */
2702: /* be ill-conditioned. the default is false (0). this */
2703: /* variable is automatically set, when praxis finds */
2704: /* the problem to be ill-conditioned during iterations. */
2705: /* maxfun is the maximum number of calls to fun allowed. praxis */
2706: /* will return after maxfun calls to fun even when the */
2707: /* minimum is not yet found. the default value of 0 */
2708: /* indicates no limit on the number of calls. */
2709: /* this return condition is only checked every n */
2710: /* iterations. */
2711: /* */
2712: /*********************************************************************/
2713:
2714: #include <math.h>
2715: #include <stdio.h>
2716: #include <stdlib.h>
2717: #include <float.h> /* for DBL_EPSILON */
2718: /* #include "machine.h" */
2719:
2720:
2721: /* extern void minfit(int n, double eps, double tol, double **ab, double q[]); */
2722: /* extern void minfit(int n, double eps, double tol, double ab[N][N], double q[]); */
2723: /* control parameters */
2724: /* control parameters */
2725: #define SQREPSILON 1.0e-19
2726: /* #define EPSILON 1.0e-8 */ /* in main */
2727:
2728: double tol = SQREPSILON,
2729: scbd = 1.0,
2730: step = 1.0;
2731: int ktm = 1,
2732: /* prin = 2, */
2733: maxfun = 0,
2734: illc = 0;
2735:
2736: /* some global variables */
2737: static int i, j, k, k2, nl, nf, kl, kt;
2738: /* static double s; */
2739: double sl, dn, dmin,
2740: fx, f1, lds, ldt, sf, df,
2741: qf1, qd0, qd1, qa, qb, qc,
2742: m2, m4, small_windows, vsmall, large,
2743: vlarge, ldfac, t2;
2744: /* static double d[N], y[N], z[N], */
2745: /* q0[N], q1[N], v[N][N]; */
2746:
2747: static double *d, *y, *z;
2748: static double *q0, *q1, **v;
2749: double *tflin; /* used in flin: return (*fun)(tflin, n); */
2750: double *e; /* used in minfit, don't konw how to free memory and thus made global */
2751: /* static double s, sl, dn, dmin, */
2752: /* fx, f1, lds, ldt, sf, df, */
2753: /* qf1, qd0, qd1, qa, qb, qc, */
2754: /* m2, m4, small, vsmall, large, */
2755: /* vlarge, ldfac, t2; */
2756: /* static double d[N], y[N], z[N], */
2757: /* q0[N], q1[N], v[N][N]; */
2758:
2759: /* these will be set by praxis to point to it's arguments */
2760: static int prin; /* added */
2761: static int n;
2762: static double *x;
2763: static double (*fun)();
2764: /* static double (*fun)(double *x, int n); */
2765:
2766: /* these will be set by praxis to the global control parameters */
2767: /* static double h, macheps, t; */
2768: extern double macheps;
2769: static double h;
2770: static double t;
2771:
2772: static double
2773: drandom() /* return random no between 0 and 1 */
2774: {
2775: return (double)(rand()%(8192*2))/(double)(8192*2);
2776: }
2777:
2778: static void sort() /* d and v in descending order */
2779: {
2780: int k, i, j;
2781: double s;
2782:
2783: for (i=1; i<=n-1; i++) {
2784: k = i; s = d[i];
2785: for (j=i+1; j<=n; j++) {
2786: if (d[j] > s) {
2787: k = j;
2788: s = d[j];
2789: }
2790: }
2791: if (k > i) {
2792: d[k] = d[i];
2793: d[i] = s;
2794: for (j=1; j<=n; j++) {
2795: s = v[j][i];
2796: v[j][i] = v[j][k];
2797: v[j][k] = s;
2798: }
2799: }
2800: }
2801: }
2802:
2803: double randbrent ( int *naught )
2804: {
2805: double ran1, ran3[127], half;
2806: int ran2, q, r, i, j;
2807: int init=0; /* false */
2808: double rr;
2809: /* REAL*8 RAN1,RAN3(127),HALF */
2810:
2811: /* INTEGER RAN2,Q,R */
2812: /* LOGICAL INIT */
2813: /* DATA INIT/.FALSE./ */
2814: /* IF (INIT) GO TO 3 */
2815: if(!init){
2816: /* R = MOD(NAUGHT,8190) + 1 *//* 1804289383 rand () */
2817: r = *naught % 8190 + 1;/* printf(" naught r %d %d",*naught,r); */
2818: ran2=127;
2819: for(i=ran2; i>0; i--){
2820: /* RAN2 = 128 */
2821: /* DO 2 I=1,127 */
2822: ran2 = ran2-1;
2823: /* RAN2 = RAN2 - 1 */
2824: ran1 = -pow(2.0,55);
2825: /* RAN1 = -2.D0**55 */
2826: /* DO 1 J=1,7 */
2827: for(j=1; j<=7;j++){
2828: /* R = MOD(1756*R,8191) */
2829: r = (1756*r) % 8191;/* printf(" i=%d (1756*r)%8191=%d",j,r); */
2830: q=r/32;
2831: /* Q = R/32 */
2832: /* 1 RAN1 = (RAN1 + Q)*(1.0D0/256) */
2833: ran1 =(ran1+q)*(1.0/256);
2834: }
2835: /* 2 RAN3(RAN2) = RAN1 */
2836: ran3[ran2] = ran1; /* printf(" ran2=%d ran1=%.7g \n",ran2,ran1); */
2837: }
2838: /* INIT = .TRUE. */
2839: init=1;
2840: /* 3 IF (RAN2.EQ.1) RAN2 = 128 */
2841: }
2842: if(ran2 == 0) ran2 = 126;
2843: else ran2 = ran2 -1;
2844: /* RAN2 = RAN2 - 1 */
2845: /* RAN1 = RAN1 + RAN3(RAN2) */
2846: ran1 = ran1 + ran3[ran2];/* printf("BIS ran2=%d ran1=%.7g \n",ran2,ran1); */
2847: half= 0.5;
2848: /* HALF = .5D0 */
2849: /* IF (RAN1.GE.0.D0) HALF = -HALF */
2850: if(ran1 >= 0.) half =-half;
2851: ran1 = ran1 +half;
2852: ran3[ran2] = ran1;
2853: rr= ran1+0.5;
2854: /* RAN1 = RAN1 + HALF */
2855: /* RAN3(RAN2) = RAN1 */
2856: /* RANDOM = RAN1 + .5D0 */
2857: /* r = ( ( double ) ( *seed ) ) * 4.656612875E-10; */
2858: return rr;
2859: }
2860: static void matprint(char *s, double **v, int m, int n)
2861: /* char *s; */
2862: /* double v[N][N]; */
2863: {
2864: #define INCX 8
2865: int i;
2866:
2867: int i2hi;
2868: int ihi;
2869: int ilo;
2870: int i2lo;
2871: int jlo=1;
2872: int j;
2873: int j2hi;
2874: int jhi;
2875: int j2lo;
2876: ilo=1;
2877: ihi=n;
2878: jlo=1;
2879: jhi=n;
2880:
2881: printf ("\n" );
2882: printf ("%s\n", s );
2883: for ( j2lo = jlo; j2lo <= jhi; j2lo = j2lo + INCX )
2884: {
2885: j2hi = j2lo + INCX - 1;
2886: if ( n < j2hi )
2887: {
2888: j2hi = n;
2889: }
2890: if ( jhi < j2hi )
2891: {
2892: j2hi = jhi;
2893: }
2894:
2895: /* fprintf ( ficlog, "\n" ); */
2896: printf ("\n" );
2897: /*
2898: For each column J in the current range...
2899:
2900: Write the header.
2901: */
2902: /* fprintf ( ficlog, " Col: "); */
2903: printf ("Col:");
2904: for ( j = j2lo; j <= j2hi; j++ )
2905: {
2906: /* fprintf ( ficlog, " %7d ", j - 1 ); */
2907: /* printf (" %9d ", j - 1 ); */
2908: printf (" %9d ", j );
2909: }
2910: /* fprintf ( ficlog, "\n" ); */
2911: /* fprintf ( ficlog, " Row\n" ); */
2912: /* fprintf ( ficlog, "\n" ); */
2913: printf ("\n" );
2914: printf (" Row\n" );
2915: printf ("\n" );
2916: /*
2917: Determine the range of the rows in this strip.
2918: */
2919: if ( 1 < ilo ){
2920: i2lo = ilo;
2921: }else{
2922: i2lo = 1;
2923: }
2924: if ( m < ihi ){
2925: i2hi = m;
2926: }else{
2927: i2hi = ihi;
2928: }
2929:
2930: for ( i = i2lo; i <= i2hi; i++ ){
2931: /*
2932: Print out (up to) 5 entries in row I, that lie in the current strip.
2933: */
2934: /* fprintf ( ficlog, "%5d:", i - 1 ); */
2935: /* printf ("%5d:", i - 1 ); */
2936: printf ("%5d:", i );
2937: for ( j = j2lo; j <= j2hi; j++ )
2938: {
2939: /* fprintf ( ficlog, " %14g", a[i-1+(j-1)*m] ); */
2940: /* printf ("%14.7g ", a[i-1+(j-1)*m] ); */
2941: /* printf("%14.7f ", v[i-1][j-1]); */
2942: printf("%14.7f ", v[i][j]);
2943: /* fprintf ( stdout, " %14g", a[i-1+(j-1)*m] ); */
2944: }
2945: /* fprintf ( ficlog, "\n" ); */
2946: printf ("\n" );
2947: }
2948: }
2949:
2950: /* printf("%s\n", s); */
2951: /* for (k=0; k<n; k++) { */
2952: /* for (i=0; i<n; i++) { */
2953: /* /\* printf("%20.10e ", v[k][i]); *\/ */
2954: /* } */
2955: /* printf("\n"); */
2956: /* } */
2957: #undef INCX
2958: }
2959:
2960: void vecprint(char *s, double *x, int n)
2961: /* char *s; */
2962: /* double x[N]; */
2963: {
2964: int i=0;
2965:
2966: printf(" %s", s);
2967: /* for (i=0; i<n; i++) */
2968: for (i=1; i<=n; i++)
2969: printf (" %14.7g", x[i] );
2970: /* printf(" %8d: %14g\n", i, x[i]); */
2971: printf ("\n" );
2972: }
2973:
2974: static void print() /* print a line of traces */
2975: {
2976:
2977:
2978: printf("\n");
2979: /* printf("... chi square reduced to ... %20.10e\n", fx); */
2980: /* printf("... after %u function calls ...\n", nf); */
2981: /* printf("... including %u linear searches ...\n", nl); */
2982: printf("%10d %10d%14.7g",nl, nf, fx);
2983: vecprint("... current values of x ...", x, n);
2984: }
2985: /* static void print2(int n, double *x, int prin, double fx, int nf, int nl) */ /* print a line of traces */
2986: static void print2() /* print a line of traces */
2987: {
2988: int i; double fmin=0.;
2989:
2990: /* printf("\n"); */
2991: /* printf("... chi square reduced to ... %20.10e\n", fx); */
2992: /* printf("... after %u function calls ...\n", nf); */
2993: /* printf("... including %u linear searches ...\n", nl); */
2994: /* printf("%10d %10d%14.7g",nl, nf, fx); */
2995: printf ( "\n" );
2996: printf ( " Linear searches %d", nl );
2997: /* printf ( " Linear searches %d\n", nl ); */
2998: /* printf ( " Function evaluations %d\n", nf ); */
2999: /* printf ( " Function value FX = %g\n", fx ); */
3000: printf ( " Function evaluations %d", nf );
3001: printf ( " Function value FX = %.12lf\n", fx );
3002: #ifdef DEBUGPRAX
3003: printf("n=%d prin=%d\n",n,prin);
3004: #endif
3005: if(fx <= fmin) printf(" UNDEFINED "); else printf("%14.7g",log(fx-fmin));
3006: if ( n <= 4 || 2 < prin )
3007: {
3008: /* for(i=1;i<=n;i++)printf("%14.7g",x[i-1]); */
3009: for(i=1;i<=n;i++)printf("%14.7g",x[i]);
3010: /* r8vec_print ( n, x, " X:" ); */
3011: }
3012: printf("\n");
3013: }
3014:
3015:
3016: /* #ifdef MSDOS */
3017: /* static double tflin[N]; */
3018: /* #endif */
3019:
3020: static double flin(double l, int j)
3021: /* double l; */
3022: {
3023: int i;
3024: /* #ifndef MSDOS */
3025: /* double tflin[N]; */
3026: /* #endif */
3027: /* double *tflin; */ /* Be careful to put tflin on a vector n */
3028:
3029: /* j is used from 0 to n-1 and can be -1 for parabolic search */
3030:
3031: /* if (j != -1) { /\* linear search *\/ */
3032: if (j > 0) { /* linear search */
3033: /* for (i=0; i<n; i++){ */
3034: for (i=1; i<=n; i++){
3035: tflin[i] = x[i] + l *v[i][j];
3036: #ifdef DEBUGPRAX
3037: /* 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); */
3038: 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);
3039: #endif
3040: }
3041: }
3042: else { /* search along parabolic space curve */
3043: qa = l*(l-qd1)/(qd0*(qd0+qd1));
3044: qb = (l+qd0)*(qd1-l)/(qd0*qd1);
3045: qc = l*(l+qd0)/(qd1*(qd0+qd1));
3046: #ifdef DEBUGPRAX
3047: 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);
3048: #endif
3049: /* for (i=0; i<n; i++){ */
3050: for (i=1; i<=n; i++){
3051: tflin[i] = qa*q0[i]+qb*x[i]+qc*q1[i];
3052: #ifdef DEBUGPRAX
3053: /* 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]); */
3054: 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]);
3055: #endif
3056: }
3057: }
3058: nf++;
3059:
3060: #ifdef NR_SHIFT
3061: return (*fun)((tflin-1), n);
3062: #else
3063: /* return (*fun)(tflin, n);*/
3064: return (*fun)(tflin);
3065: #endif
3066: }
3067:
3068: void minny(int j, int nits, double *d2, double *x1, double f1, int fk)
3069: /* double *d2, *x1, f1; */
3070: {
3071: /* here j is from 0 to n-1 and can be -1 for parabolic search */
3072: /* MINIMIZES F FROM X IN THE DIRECTION V(*,J) */
3073: /* UNLESS J<1, WHEN A QUADRATIC SEARCH IS DONE */
3074: /* IN THE PLANE DEFINED BY Q0, Q1 AND X. */
3075: /* D2 AN APPROXIMATION TO HALF F'' (OR ZERO), */
3076: /* X1 AN ESTIMATE OF DISTANCE TO MINIMUM, */
3077: /* RETURNED AS THE DISTANCE FOUND. */
3078: /* IF FK = TRUE THEN F1 IS FLIN(X1), OTHERWISE */
3079: /* X1 AND F1 ARE IGNORED ON ENTRY UNLESS FINAL */
3080: /* FX > F1. NITS CONTROLS THE NUMBER OF TIMES */
3081: /* AN ATTEMPT IS MADE TO HALVE THE INTERVAL. */
3082: /* SIDE EFFECTS: USES AND ALTERS X, FX, NF, NL. */
3083: /* IF J < 1 USES VARIABLES Q... . */
3084: /* USES H, N, T, M2, M4, LDT, DMIN, MACHEPS; */
3085: int k, i, dz;
3086: double x2, xm, f0, f2, fm, d1, t2, sf1, sx1;
3087: double s;
3088: double macheps;
3089: macheps=pow(16.0,-13.0);
3090: sf1 = f1; sx1 = *x1;
3091: k = 0; xm = 0.0; fm = f0 = fx; dz = *d2 < macheps;
3092: /* h=1.0;*/ /* To be revised */
3093: #ifdef DEBUGPRAX
3094: /* printf("min macheps=%14g h=%14g step=%14g t=%14g fx=%14g\n",macheps,h, step,t, fx); */
3095: /* Where is fx coming from */
3096: printf(" min macheps=%14g h=%14g t=%14g fx=%.9lf dirj=%d\n",macheps, h, t, fx, j);
3097: matprint(" min vectors:",v,n,n);
3098: #endif
3099: /* find step size */
3100: s = 0.;
3101: /* for (i=0; i<n; i++) s += x[i]*x[i]; */
3102: for (i=1; i<=n; i++) s += x[i]*x[i];
3103: s = sqrt(s);
3104: if (dz)
3105: t2 = m4*sqrt(fabs(fx)/dmin + s*ldt) + m2*ldt;
3106: else
3107: t2 = m4*sqrt(fabs(fx)/(*d2) + s*ldt) + m2*ldt;
3108: s = s*m4 + t;
3109: if (dz && t2 > s) t2 = s;
3110: if (t2 < small_windows) t2 = small_windows;
3111: if (t2 > 0.01*h) t2 = 0.01 * h;
3112: if (fk && f1 <= fm) {
3113: xm = *x1;
3114: fm = f1;
3115: }
3116: #ifdef DEBUGPRAX
3117: printf(" additional flin X1=%14.7f t2=%14.7f *f1=%14.7f fm=%14.7f fk=%d\n",*x1,t2,f1,fm,fk);
3118: #endif
3119: if (!fk || fabs(*x1) < t2) {
3120: *x1 = (*x1 >= 0 ? t2 : -t2);
3121: /* *x1 = (*x1 > 0 ? t2 : -t2); */ /* kind of error */
3122: #ifdef DEBUGPRAX
3123: printf(" additional flin X1=%16.10e dirj=%d fk=%d\n",*x1, j, fk);
3124: #endif
3125: f1 = flin(*x1, j);
3126: #ifdef DEBUGPRAX
3127: printf(" after flin f1=%18.12e dirj=%d fk=%d\n",f1, j,fk);
3128: #endif
3129: }
3130: if (f1 <= fm) {
3131: xm = *x1;
3132: fm = f1;
3133: }
3134: L0: /*L0 loop or next */
3135: /*
3136: Evaluate FLIN at another point and estimate the second derivative.
3137: */
3138: if (dz) {
3139: x2 = (f0 < f1 ? -(*x1) : 2*(*x1));
3140: #ifdef DEBUGPRAX
3141: printf(" additional second flin x2=%14.8e x1=%14.8e f0=%14.8e f1=%18.12e dirj=%d\n",x2,*x1,f0,f1,j);
3142: #endif
3143: f2 = flin(x2, j);
3144: #ifdef DEBUGPRAX
3145: 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);
3146: #endif
3147: if (f2 <= fm) {
3148: xm = x2;
3149: fm = f2;
3150: }
3151: /* d2 is the curvature or double difference f1 doesn't seem to be accurately computed */
3152: *d2 = (x2*(f1-f0) - (*x1)*(f2-f0))/((*x1)*x2*((*x1)-x2));
3153: #ifdef DEBUGPRAX
3154: double d11,d12;
3155: d11=(f1-f0)/(*x1);d12=(f2-f0)/x2;
3156: 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)));
3157: 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);
3158: double ff1=7.783920622852e+04;
3159: double f1mf0=9.0344736236e-05;
3160: *d2 = (f1mf0)/ (*x1)/((*x1)-x2) - (f2-f0)/x2/((*x1)-x2);
3161: /* *d2 = (ff1-f0)/ (*x1)/((*x1)-x2) - (f2-f0)/x2/((*x1)-x2); */
3162: printf(" simpliff computing *d2=%16.10e f1mf0=%18.12e,f1=f0+f1mf0=%18.12e\n",*d2,f1mf0,f0+f1mf0);
3163: *d2 = ((f1-f0)/ (*x1) - (f2-f0)/x2)/((*x1)-x2);
3164: printf(" overlifi computing *d2=%16.10e\n",*d2);
3165: #endif
3166: *d2 = ((f1-f0)/ (*x1) - (f2-f0)/x2)/((*x1)-x2);
3167: }
3168: #ifdef DEBUGPRAX
3169: printf(" additional second flin xm=%14.8e fm=%14.8e *d2=%14.8e\n",xm, fm,*d2);
3170: #endif
3171: /*
3172: Estimate the first derivative at 0.
3173: */
3174: d1 = (f1-f0)/(*x1) - *x1**d2; dz = 1;
3175: /*
3176: Predict the minimum.
3177: */
3178: if (*d2 <= small_windows) {
3179: x2 = (d1 < 0 ? h : -h);
3180: }
3181: else {
3182: x2 = - 0.5*d1/(*d2);
3183: }
3184: #ifdef DEBUGPRAX
3185: 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);
3186: #endif
3187: if (fabs(x2) > h)
3188: x2 = (x2 > 0 ? h : -h);
3189: L1: /* L1 or try loop */
3190: #ifdef DEBUGPRAX
3191: printf(" AT predicted minimum flin x2=%14.8e x1=%14.8e K=%14d NITS=%14d dirj=%d\n",x2,*x1,k,nits,j);
3192: #endif
3193: f2 = flin(x2, j); /* x[i]+x2*v[i][j] */
3194: #ifdef DEBUGPRAX
3195: printf(" after flin f0=%14.8e f1=%14.8e f2=%14.8e fm=%14.8e\n",f0,f1,f2, fm);
3196: #endif
3197: if ((k < nits) && (f2 > f0)) {
3198: #ifdef DEBUGPRAX
3199: printf(" NO SUCCESS SO TRY AGAIN;\n");
3200: #endif
3201: k++;
3202: if ((f0 < f1) && (*x1*x2 > 0.0))
3203: goto L0; /* or next */
3204: x2 *= 0.5;
3205: goto L1;
3206: }
3207: nl++;
3208: #ifdef DEBUGPRAX
3209: 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);
3210: #endif
3211: if (f2 > fm) x2 = xm; else fm = f2;
3212: if (fabs(x2*(x2-*x1)) > small_windows) {
3213: *d2 = (x2*(f1-f0) - *x1*(fm-f0))/(*x1*x2*(*x1-x2));
3214: }
3215: else {
3216: if (k > 0) *d2 = 0;
3217: }
3218: #ifdef DEBUGPRAX
3219: printf(" bebe end of min x1=%14.8e fx=%14.8e d2=%14.8e\n",*x1, fx, *d2);
3220: #endif
3221: if (*d2 <= small_windows) *d2 = small_windows;
3222: *x1 = x2; fx = fm;
3223: if (sf1 < fx) {
3224: fx = sf1;
3225: *x1 = sx1;
3226: }
3227: /*
3228: Update X for linear search.
3229: */
3230: #ifdef DEBUGPRAX
3231: printf(" end of min x1=%14.8e fx=%14.8e d2=%14.8e\n",*x1, fx, *d2);
3232: #endif
3233:
3234: /* if (j != -1) */
3235: /* for (i=0; i<n; i++) */
3236: /* x[i] += (*x1)*v[i][j]; */
3237: if (j > 0)
3238: for (i=1; i<=n; i++)
3239: x[i] += (*x1)*v[i][j];
3240: }
3241:
3242: void quad() /* look for a minimum along the curve q0, q1, q2 */
3243: {
3244: int i;
3245: double l, s;
3246:
3247: s = fx; fx = qf1; qf1 = s; qd1 = 0.0;
3248: /* for (i=0; i<n; i++) { */
3249: for (i=1; i<=n; i++) {
3250: s = x[i]; l = q1[i]; x[i] = l; q1[i] = s;
3251: qd1 = qd1 + (s-l)*(s-l);
3252: }
3253: s = 0.0; qd1 = sqrt(qd1); l = qd1;
3254: #ifdef DEBUGPRAX
3255: printf(" QUAD after sqrt qd1=%14.8e \n",qd1);
3256: #endif
3257:
3258: if (qd0>0.0 && qd1>0.0 &&nl>=3*n*n) {
3259: #ifdef DEBUGPRAX
3260: printf(" QUAD before min value=%14.8e \n",qf1);
3261: #endif
3262: /* min(-1, 2, &s, &l, qf1, 1); */
3263: minny(0, 2, &s, &l, qf1, 1);
3264: qa = l*(l-qd1)/(qd0*(qd0+qd1));
3265: qb = (l+qd0)*(qd1-l)/(qd0*qd1);
3266: qc = l*(l+qd0)/(qd1*(qd0+qd1));
3267: }
3268: else {
3269: fx = qf1; qa = qb = 0.0; qc = 1.0;
3270: }
3271: #ifdef DEBUGPRAX
3272: printf("after eventual min qd0=%14.8e qd1=%14.8e nl=%d\n",qd0, qd1,nl);
3273: #endif
3274: qd0 = qd1;
3275: /* for (i=0; i<n; i++) { */
3276: for (i=1; i<=n; i++) {
3277: s = q0[i]; q0[i] = x[i];
3278: x[i] = qa*s + qb*x[i] + qc*q1[i];
3279: }
3280: #ifdef DEBUGQUAD
3281: vecprint ( " X after QUAD:" , x, n );
3282: #endif
3283: }
3284:
3285: /* void minfit(int n, double eps, double tol, double ab[N][N], double q[]) */
3286: void minfit(int n, double eps, double tol, double **ab, double q[])
3287: /* int n; */
3288: /* double eps, tol, ab[N][N], q[N]; */
3289: {
3290: int l, kt, l2, i, j, k;
3291: double c, f, g, h, s, x, y, z;
3292: /* double eps; */
3293: /* #ifndef MSDOS */
3294: /* double e[N]; /\* plenty of stack on a vax *\/ */
3295: /* #endif */
3296: /* double *e; */
3297: /* e=vector(0,n-1); /\* should be freed somewhere but gotos *\/ */
3298:
3299: /* householder's reduction to bidiagonal form */
3300:
3301: if(n==1){
3302: /* q[1-1]=ab[1-1][1-1]; */
3303: /* ab[1-1][1-1]=1.0; */
3304: q[1]=ab[1][1];
3305: ab[1][1]=1.0;
3306: return; /* added from hardt */
3307: }
3308: /* eps=macheps; */ /* added */
3309: x = g = 0.0;
3310: #ifdef DEBUGPRAX
3311: matprint (" HOUSE holder:", ab, n, n);
3312: #endif
3313:
3314: /* for (i=0; i<n; i++) { /\* FOR I := 1 UNTIL N DO *\/ */
3315: for (i=1; i<=n; i++) { /* FOR I := 1 UNTIL N DO */
3316: e[i] = g; s = 0.0; l = i+1;
3317: /* for (j=i; j<n; j++) /\* FOR J := I UNTIL N DO S := S*AB(J,I)**2; *\/ /\* not correct *\/ */
3318: for (j=i; j<=n; j++) /* FOR J := I UNTIL N DO S := S*AB(J,I)**2; */ /* not correct */
3319: s += ab[j][i] * ab[j][i];
3320: #ifdef DEBUGPRAXFIN
3321: printf("i=%d s=%d %.7g tol=%.7g",i,s,tol);
3322: #endif
3323: if (s < tol) {
3324: g = 0.0;
3325: }
3326: else {
3327: /* f = ab[i][i]; */
3328: f = ab[i][i];
3329: if (f < 0.0)
3330: g = sqrt(s);
3331: else
3332: g = -sqrt(s);
3333: /* h = f*g - s; ab[i][i] = f - g; */
3334: h = f*g - s; ab[i][i] = f - g;
3335: /* for (j=l; j<n; j++) { */ /* FOR J := L UNTIL N DO */ /* wrong */
3336: for (j=l; j<=n; j++) {
3337: f = 0.0;
3338: /* for (k=i; k<n; k++) /\* FOR K := I UNTIL N DO *\/ /\* wrong *\/ */
3339: for (k=i; k<=n; k++) /* FOR K := I UNTIL N DO */
3340: /* f += ab[k][i] * ab[k][j]; */
3341: f += ab[k][i] * ab[k][j];
3342: f /= h;
3343: for (k=i; k<=n; k++) /* FOR K := I UNTIL N DO */
3344: /* for (k=i; k<n; k++)/\* FOR K := I UNTIL N DO *\/ /\* wrong *\/ */
3345: ab[k][j] += f * ab[k][i];
3346: /* ab[k][j] += f * ab[k][i]; */
3347: #ifdef DEBUGPRAX
3348: printf("Holder J=%d F=%.7g",j,f);
3349: #endif
3350: }
3351: } /* end s */
3352: /* q[i] = g; s = 0.0; */
3353: q[i] = g; s = 0.0;
3354: #ifdef DEBUGPRAX
3355: printf(" I Q=%d %.7g",i,q[i]);
3356: #endif
3357:
3358: /* if (i < n) */
3359: /* if (i <= n) /\* I is always lower or equal to n wasn't in golub reinsch*\/ */
3360: /* for (j=l; j<n; j++) */
3361: for (j=l; j<=n; j++)
3362: s += ab[i][j] * ab[i][j];
3363: /* s += ab[i][j] * ab[i][j]; */
3364: if (s < tol) {
3365: g = 0.0;
3366: }
3367: else {
3368: if(i<n)
3369: /* f = ab[i][i+1]; */ /* Brent golub overflow */
3370: f = ab[i][i+1];
3371: if (f < 0.0)
3372: g = sqrt(s);
3373: else
3374: g = - sqrt(s);
3375: h = f*g - s;
3376: /* h = f*g - s; ab[i][i+1] = f - g; */ /* Overflow for i=n Error in Golub too but not Burkardt*/
3377: /* for (j=l; j<n; j++) */
3378: /* e[j] = ab[i][j]/h; */
3379: if(i<n){
3380: ab[i][i+1] = f - g;
3381: for (j=l; j<=n; j++)
3382: e[j] = ab[i][j]/h;
3383: /* for (j=l; j<n; j++) { */
3384: for (j=l; j<=n; j++) {
3385: s = 0.0;
3386: /* for (k=l; k<n; k++) s += ab[j][k]*ab[i][k]; */
3387: for (k=l; k<=n; k++) s += ab[j][k]*ab[i][k];
3388: /* for (k=l; k<n; k++) ab[j][k] += s * e[k]; */
3389: for (k=l; k<=n; k++) ab[j][k] += s * e[k];
3390: } /* END J */
3391: } /* END i <n */
3392: } /* end s */
3393: /* y = fabs(q[i]) + fabs(e[i]); */
3394: y = fabs(q[i]) + fabs(e[i]);
3395: if (y > x) x = y;
3396: #ifdef DEBUGPRAX
3397: printf(" I Y=%d %.7g",i,y);
3398: #endif
3399: #ifdef DEBUGPRAX
3400: printf(" i=%d e(i) %.7g",i,e[i]);
3401: #endif
3402: } /* end i */
3403: /*
3404: Accumulation of right hand transformations */
3405: /* for (i=n-1; i >= 0; i--) { */ /* FOR I := N STEP -1 UNTIL 1 DO */
3406: /* We should avoid the overflow in Golub */
3407: /* ab[n-1][n-1] = 1.0; */
3408: /* g = e[n-1]; */
3409: ab[n][n] = 1.0;
3410: g = e[n];
3411: l = n;
3412:
3413: /* for (i=n; i >= 1; i--) { */
3414: for (i=n-1; i >= 1; i--) { /* n-1 loops, different from brent and golub*/
3415: if (g != 0.0) {
3416: /* h = ab[i-1][i]*g; */
3417: h = ab[i][i+1]*g;
3418: for (j=l; j<=n; j++) ab[j][i] = ab[i][j] / h;
3419: for (j=l; j<=n; j++) {
3420: /* h = ab[i][i+1]*g; */
3421: /* for (j=l; j<n; j++) ab[j][i] = ab[i][j] / h; */
3422: /* for (j=l; j<n; j++) { */
3423: s = 0.0;
3424: /* for (k=l; k<n; k++) s += ab[i][k] * ab[k][j]; */
3425: /* for (k=l; k<n; k++) ab[k][j] += s * ab[k][i]; */
3426: for (k=l; k<=n; k++) s += ab[i][k] * ab[k][j];
3427: for (k=l; k<=n; k++) ab[k][j] += s * ab[k][i];
3428: }/* END J */
3429: }/* END G */
3430: /* for (j=l; j<n; j++) */
3431: /* ab[i][j] = ab[j][i] = 0.0; */
3432: /* ab[i][i] = 1.0; g = e[i]; l = i; */
3433: for (j=l; j<=n; j++)
3434: ab[i][j] = ab[j][i] = 0.0;
3435: ab[i][i] = 1.0; g = e[i]; l = i;
3436: }/* END I */
3437: #ifdef DEBUGPRAX
3438: matprint (" HOUSE accumulation:",ab,n, n );
3439: #endif
3440:
3441: /* diagonalization to bidiagonal form */
3442: eps *= x;
3443: /* for (k=n-1; k>= 0; k--) { */
3444: for (k=n; k>= 1; k--) {
3445: kt = 0;
3446: TestFsplitting:
3447: #ifdef DEBUGPRAX
3448: printf(" TestFsplitting: k=%d kt=%d\n",k,kt);
3449: /* for(i=1;i<=n;i++)printf(" e(%d)=%.14f",i,e[i]);printf("\n"); */
3450: #endif
3451: kt = kt+1;
3452: /* TestFsplitting: */
3453: /* if (++kt > 30) { */
3454: if (kt > 30) {
3455: e[k] = 0.0;
3456: fprintf(stderr, "\n+++ MINFIT - Fatal error\n");
3457: fprintf ( stderr, " The QR algorithm failed to converge.\n" );
3458: }
3459: /* for (l2=k; l2>=0; l2--) { */
3460: for (l2=k; l2>=1; l2--) {
3461: l = l2;
3462: #ifdef DEBUGPRAX
3463: printf(" l e(l)< eps %d %.7g %.7g ",l,e[l], eps);
3464: #endif
3465: /* if (fabs(e[l]) <= eps) */
3466: if (fabs(e[l]) <= eps)
3467: goto TestFconvergence;
3468: /* if (fabs(q[l-1]) <= eps)*/ /* missing if ( 1 < l ){ *//* printf(" q(l-1)< eps %d %.7g %.7g ",l-1,q[l-2], eps); */
3469: if (fabs(q[l-1]) <= eps)
3470: break; /* goto Cancellation; */
3471: }
3472: Cancellation:
3473: #ifdef DEBUGPRAX
3474: printf(" Cancellation:\n");
3475: #endif
3476: c = 0.0; s = 1.0;
3477: for (i=l; i<=k; i++) {
3478: f = s * e[i]; e[i] *= c;
3479: /* f = s * e[i]; e[i] *= c; */
3480: if (fabs(f) <= eps)
3481: goto TestFconvergence;
3482: /* g = q[i]; */
3483: g = q[i];
3484: if (fabs(f) < fabs(g)) {
3485: double fg = f/g;
3486: h = fabs(g)*sqrt(1.0+fg*fg);
3487: }
3488: else {
3489: double gf = g/f;
3490: h = (f!=0.0 ? fabs(f)*sqrt(1.0+gf*gf) : 0.0);
3491: }
3492: /* COMMENT: THE ABOVE REPLACES Q(I):=H:=LONGSQRT(G*G+F*F) */
3493: /* WHICH MAY GIVE INCORRECT RESULTS IF THE */
3494: /* SQUARES UNDERFLOW OR IF F = G = 0; */
3495:
3496: /* q[i] = h; */
3497: q[i] = h;
3498: if (h == 0.0) { h = 1.0; g = 1.0; }
3499: c = g/h; s = -f/h;
3500: }
3501: TestFconvergence:
3502: #ifdef DEBUGPRAX
3503: printf(" TestFconvergence: l=%d k=%d\n",l,k);
3504: #endif
3505: /* z = q[k]; */
3506: z = q[k];
3507: if (l == k)
3508: goto Convergence;
3509: /* shift from bottom 2x2 minor */
3510: /* x = q[l]; y = q[k-l]; g = e[k-1]; h = e[k]; */ /* Error */
3511: x = q[l]; y = q[k-1]; g = e[k-1]; h = e[k];
3512: f = ((y-z)*(y+z) + (g-h)*(g+h)) / (2.0*h*y);
3513: g = sqrt(f*f+1.0);
3514: if (f <= 0.0)
3515: f = ((x-z)*(x+z) + h*(y/(f-g)-h))/x;
3516: else
3517: f = ((x-z)*(x+z) + h*(y/(f+g)-h))/x;
3518: /* next qr transformation */
3519: s = c = 1.0;
3520: for (i=l+1; i<=k; i++) {
3521: #ifdef DEBUGPRAXQR
3522: 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]);
3523: #endif
3524: /* g = e[i]; y = q[i]; h = s*g; g *= c; */
3525: g = e[i]; y = q[i]; h = s*g; g *= c;
3526: if (fabs(f) < fabs(h)) {
3527: double fh = f/h;
3528: z = fabs(h) * sqrt(1.0 + fh*fh);
3529: }
3530: else {
3531: double hf = h/f;
3532: z = (f!=0.0 ? fabs(f)*sqrt(1.0+hf*hf) : 0.0);
3533: }
3534: /* e[i-1] = z; */
3535: e[i-1] = z;
3536: #ifdef DEBUGPRAXQR
3537: 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]);
3538: #endif
3539: if (z == 0.0)
3540: f = z = 1.0;
3541: c = f/z; s = h/z;
3542: f = x*c + g*s; g = - x*s + g*c; h = y*s;
3543: y *= c;
3544: /* for (j=0; j<n; j++) { */
3545: /* x = ab[j][i-1]; z = ab[j][i]; */
3546: /* ab[j][i-1] = x*c + z*s; */
3547: /* ab[j][i] = - x*s + z*c; */
3548: /* } */
3549: for (j=1; j<=n; j++) {
3550: x = ab[j][i-1]; z = ab[j][i];
3551: ab[j][i-1] = x*c + z*s;
3552: ab[j][i] = - x*s + z*c;
3553: }
3554: if (fabs(f) < fabs(h)) {
3555: double fh = f/h;
3556: z = fabs(h) * sqrt(1.0 + fh*fh);
3557: }
3558: else {
3559: double hf = h/f;
3560: z = (f!=0.0 ? fabs(f)*sqrt(1.0+hf*hf) : 0.0);
3561: }
3562: #ifdef DEBUGPRAXQR
3563: printf(" qr transformation z f h=%.7g %.7g %.7g i=%d k=%d\n",z,f,h, i, k);
3564: #endif
3565: q[i-1] = z;
3566: if (z == 0.0)
3567: z = f = 1.0;
3568: c = f/z; s = h/z;
3569: f = c*g + s*y; /* f can be very small */
3570: x = - s*g + c*y;
3571: }
3572: /* e[l] = 0.0; e[k] = f; q[k] = x; */
3573: e[l] = 0.0; e[k] = f; q[k] = x;
3574: #ifdef DEBUGPRAXQR
3575: 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);
3576: #endif
3577: goto TestFsplitting;
3578: Convergence:
3579: #ifdef DEBUGPRAX
3580: printf(" Convergence:\n");
3581: #endif
3582: if (z < 0.0) {
3583: /* q[k] = - z; */
3584: /* for (j=0; j<n; j++) ab[j][k] = - ab[j][k]; */
3585: q[k] = - z;
3586: for (j=1; j<=n; j++) ab[j][k] = - ab[j][k];
3587: }/* END Z */
3588: }/* END K */
3589: } /* END MINFIT */
3590:
3591:
3592: double praxis(double tol, double macheps, double h0, int _n, int _prin, double *_x, double (*_fun)(double *_x))
3593: /* double praxis(double tol, double macheps, double h0, int _n, int _prin, double *_x, double (*_fun)(double *_x, int _n)) */
3594: /* double praxis(double (*_fun)(), double _x[], int _n) */
3595: /* double (*_fun)(); */
3596: /* double _x[N]; */
3597: /* double (*_fun)(); */
3598: /* double _x[N]; */
3599: {
3600: /* init global extern variables and parameters */
3601: /* double *d, *y, *z, */
3602: /* *q0, *q1, **v; */
3603: /* double *tflin; /\* used in flin: return (*fun)(tflin, n); *\/ */
3604: /* double *e; /\* used in minfit, don't konw how to free memory and thus made global *\/ */
3605:
3606:
3607: int seed; /* added */
3608: int biter=0;
3609: double r;
3610: double randbrent( int (*));
3611: double s, sf;
3612:
3613: h = h0; /* step; */
3614: t = tol;
3615: scbd = 1.0;
3616: illc = 0;
3617: ktm = 1;
3618:
3619: macheps = DBL_EPSILON;
3620: /* prin=4; */
3621: #ifdef DEBUGPRAX
3622: printf("Praxis macheps=%14g h=%14g step=%14g tol=%14g\n",macheps,h, h0,tol);
3623: #endif
3624: n = _n;
3625: x = _x;
3626: prin = _prin;
3627: fun = _fun;
3628: d=vector(1, n);
3629: y=vector(1, n);
3630: z=vector(1, n);
3631: q0=vector(1, n);
3632: q1=vector(1, n);
3633: e=vector(1, n);
3634: tflin=vector(1, n);
3635: v=matrix(1, n, 1, n);
3636: for(i=1;i<=n;i++){d[i]=y[i]=z[i]=q0[0]=e[i]=tflin[i]=0.;}
3637: small_windows = (macheps) * (macheps); vsmall = small_windows*small_windows;
3638: large = 1.0/small_windows; vlarge = 1.0/vsmall;
3639: m2 = sqrt(macheps); m4 = sqrt(m2);
3640: seed = 123456789; /* added */
3641: ldfac = (illc ? 0.1 : 0.01);
3642: for(i=1;i<=n;i++) z[i]=0.; /* Was missing in Gegenfurtner as well as Brent's algol or fortran */
3643: nl = kt = 0; nf = 1;
3644: #ifdef NR_SHIFT
3645: fx = (*fun)((x-1), n);
3646: #else
3647: fx = (*fun)(x);
3648: #endif
3649: qf1 = fx;
3650: t2 = small_windows + fabs(t); t = t2; dmin = small_windows;
3651: #ifdef DEBUGPRAX
3652: printf("praxis2 macheps=%14g h=%14g step=%14g small=%14g t=%14g\n",macheps,h, h0,small_windows, t);
3653: #endif
3654: if (h < 100.0*t) h = 100.0*t;
3655: #ifdef DEBUGPRAX
3656: printf("praxis3 macheps=%14g h=%14g step=%14g small=%14g t=%14g\n",macheps,h, h0,small_windows, t);
3657: #endif
3658: ldt = h;
3659: /* for (i=0; i<n; i++) for (j=0; j<n; j++) */
3660: for (i=1; i<=n; i++) for (j=1; j<=n; j++)
3661: v[i][j] = (i == j ? 1.0 : 0.0);
3662: d[1] = 0.0; qd0 = 0.0;
3663: /* for (i=0; i<n; i++) q1[i] = x[i]; */
3664: for (i=1; i<=n; i++) q1[i] = x[i];
3665: if (prin > 1) {
3666: printf("\n------------- enter function praxis -----------\n");
3667: printf("... current parameter settings ...\n");
3668: printf("... scaling ... %20.10e\n", scbd);
3669: printf("... tol ... %20.10e\n", t);
3670: printf("... maxstep ... %20.10e\n", h);
3671: printf("... illc ... %20u\n", illc);
3672: printf("... ktm ... %20u\n", ktm);
3673: printf("... maxfun ... %20u\n", maxfun);
3674: }
3675: if (prin) print2();
3676:
3677: mloop:
3678: biter++; /* Added to count the loops */
3679: /* sf = d[0]; */
3680: /* s = d[0] = 0.0; */
3681: printf("\n Big iteration %d \n",biter);
3682: fprintf(ficlog,"\n Big iteration %d \n",biter);
3683: sf = d[1];
3684: s = d[1] = 0.0;
3685:
3686: /* minimize along first direction V(*,1) */
3687: #ifdef DEBUGPRAX
3688: printf(" Minimize along the first direction V(*,1). illc=%d\n",illc);
3689: /* fprintf(ficlog," Minimize along the first direction V(*,1).\n"); */
3690: #endif
3691: #ifdef DEBUGPRAX2
3692: printf("praxis4 macheps=%14g h=%14g step=%14g small=%14g t=%14g\n",macheps,h, h0,small_windows, t);
3693: #endif
3694: /* min(0, 2, &d[0], &s, fx, 0); /\* mac heps not global *\/ */
3695: minny(1, 2, &d[1], &s, fx, 0); /* mac heps not global */
3696: #ifdef DEBUGPRAX
3697: printf("praxis5 macheps=%14g h=%14g looks at sign of s=%14g fx=%14g\n",macheps,h, s,fx);
3698: #endif
3699: if (s <= 0.0)
3700: /* for (i=0; i < n; i++) */
3701: for (i=1; i <= n; i++)
3702: v[i][1] = -v[i][1];
3703: /* if ((sf <= (0.9 * d[0])) || ((0.9 * sf) >= d[0])) */
3704: if ((sf <= (0.9 * d[1])) || ((0.9 * sf) >= d[1]))
3705: /* for (i=1; i<n; i++) */
3706: for (i=2; i<=n; i++)
3707: d[i] = 0.0;
3708: /* for (k=1; k<n; k++) { */
3709: for (k=2; k<=n; k++) {
3710: /*
3711: The inner loop starts here.
3712: */
3713: #ifdef DEBUGPRAX
3714: printf(" The inner loop here from k=%d to n=%d.\n",k,n);
3715: /* fprintf(ficlog," The inner loop here from k=%d to n=%d.\n",k,n); */
3716: #endif
3717: /* for (i=0; i<n; i++) */
3718: for (i=1; i<=n; i++)
3719: y[i] = x[i];
3720: sf = fx;
3721: #ifdef DEBUGPRAX
3722: printf(" illc=%d and kt=%d and ktm=%d\n", illc, kt, ktm);
3723: #endif
3724: illc = illc || (kt > 0);
3725: next:
3726: kl = k;
3727: df = 0.0;
3728: if (illc) { /* random step to get off resolution valley */
3729: #ifdef DEBUGPRAX
3730: printf(" A random step follows, to avoid resolution valleys.\n");
3731: matprint(" before rand, vectors:",v,n,n);
3732: #endif
3733: for (i=1; i<=n; i++) {
3734: #ifdef NOBRENTRAND
3735: r = drandom();
3736: #else
3737: seed=i;
3738: /* seed=i+1; */
3739: #ifdef DEBUGRAND
3740: printf(" Random seed=%d, brent i=%d",seed,i); /* YYYY i=5 j=1 vji= -0.0001170073 */
3741: #endif
3742: r = randbrent ( &seed );
3743: #endif
3744: #ifdef DEBUGRAND
3745: printf(" Random r=%.7g \n",r);
3746: #endif
3747: z[i] = (0.1 * ldt + t2 * pow(10.0,(double)kt)) * (r - 0.5);
3748: /* z[i] = (0.1 * ldt + t2 * pow(10.0,(double)kt)) * (drandom() - 0.5); */
3749:
3750: s = z[i];
3751: for (j=1; j <= n; j++)
3752: x[j] += s * v[j][i];
3753: }
3754: #ifdef DEBUGRAND
3755: matprint(" after rand, vectors:",v,n,n);
3756: #endif
3757: #ifdef NR_SHIFT
3758: fx = (*fun)((x-1), n);
3759: #else
3760: fx = (*fun)(x, n);
3761: #endif
3762: /* fx = (*func) ( (x-1) ); *//* This for func which is computed from x[1] and not from x[0] xm1=(x-1)*/
3763: nf++;
3764: }
3765: /* minimize along non-conjugate directions */
3766: #ifdef DEBUGPRAX
3767: printf(" Minimize along the 'non-conjugate' directions (dots printed) V(*,%d),...,V(*,%d).\n",k,n);
3768: /* fprintf(ficlog," Minimize along the 'non-conjugate' directions (dots printed) V(*,%d),...,V(*,%d).\n",k,n); */
3769: #endif
3770: /* for (k2=k; k2<n; k2++) { /\* Be careful here k2 <=n ? *\/ */
3771: for (k2=k; k2<=n; k2++) { /* Be careful here k2 <=n ? */
3772: sl = fx;
3773: s = 0.0;
3774: #ifdef DEBUGPRAX
3775: printf(" Minimize along the 'NON-CONJUGATE' true direction k2=%14d fx=%14.7f\n",k2, fx);
3776: matprint(" before min vectors:",v,n,n);
3777: #endif
3778: /* min(k2, 2, &d[k2], &s, fx, 0); */
3779: /* jsearch=k2-1; */
3780: /* min(jsearch, 2, &d[jsearch], &s, fx, 0); */
3781: minny(k2, 2, &d[k2], &s, fx, 0);
3782: #ifdef DEBUGPRAX
3783: 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);
3784: #endif
3785: if (illc) {
3786: /* double szk = s + z[k2]; */
3787: /* s = d[k2] * szk*szk; */
3788: double szk = s + z[k2];
3789: s = d[k2] * szk*szk;
3790: }
3791: else
3792: s = sl - fx;
3793: /* if (df < s) { */
3794: if (df <= s) {
3795: df = s;
3796: kl = k2;
3797: #ifdef DEBUGPRAX
3798: printf(" df=%.7g and choose kl=%d \n",df,kl); /* UUUU */
3799: #endif
3800: }
3801: } /* end loop k2 */
3802: /*
3803: If there was not much improvement on the first try, set
3804: ILLC = true and start the inner loop again.
3805: */
3806: #ifdef DEBUGPRAX
3807: printf(" If there was not much improvement on the first try, set ILLC = true and start the inner loop again. illc=%d\n",illc);
3808: /* fprintf(ficlog," If there was not much improvement on the first try, set ILLC = true and start the inner loop again.\n"); */
3809: #endif
3810: if (!illc && (df < fabs(100.0 * (macheps) * fx))) {
3811: #ifdef DEBUGPRAX
3812: 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);
3813: #endif
3814: illc = 1;
3815: goto next;
3816: }
3817: #ifdef DEBUGPRAX
3818: 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);
3819: #endif
3820:
3821: /* if ((k == 1) && (prin > 1)){ /\* be careful k=2 *\/ */
3822: if ((k == 2) && (prin > 1)){ /* be careful k=2 */
3823: #ifdef DEBUGPRAX
3824: printf(" NEW D The second difference array d:\n" );
3825: /* fprintf(ficlog, " NEW D The second difference array d:\n" ); */
3826: #endif
3827: vecprint(" NEW D The second difference array d:",d,n);
3828: }
3829: /* minimize along conjugate directions */
3830: /*
3831: Minimize along the "conjugate" directions V(*,1),...,V(*,K-1).
3832: */
3833: #ifdef DEBUGPRAX
3834: printf("Minimize along the 'conjugate' directions V(*,1),...,V(*,K-1=%d).\n",k-1);
3835: /* fprintf(ficlog,"Minimize along the 'conjugate' directions V(*,1),...,V(*,K-1=%d).\n",k-1); */
3836: #endif
3837: /* for (k2=0; k2<=k-1; k2++) { */
3838: for (k2=1; k2<=k-1; k2++) {
3839: s = 0.0;
3840: /* min(k2-1, 2, &d[k2-1], &s, fx, 0); */
3841: minny(k2, 2, &d[k2], &s, fx, 0);
3842: }
3843: f1 = fx;
3844: fx = sf;
3845: lds = 0.0;
3846: /* for (i=0; i<n; i++) { */
3847: for (i=1; i<=n; i++) {
3848: sl = x[i];
3849: x[i] = y[i];
3850: y[i] = sl - y[i];
3851: sl = y[i];
3852: lds = lds + sl*sl;
3853: }
3854: lds = sqrt(lds);
3855: #ifdef DEBUGPRAX
3856: printf("Minimization done 'conjugate', shifted all points, computed lds=%.8f\n",lds);
3857: #endif
3858: /*
3859: Discard direction V(*,kl).
3860:
3861: If no random step was taken, V(*,KL) is the "non-conjugate"
3862: direction along which the greatest improvement was made.
3863: */
3864: if (lds > small_windows) {
3865: #ifdef DEBUGPRAX
3866: 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);
3867: matprint(" before shift new conjugate vectors:",v,n,n);
3868: #endif
3869: for (i=kl-1; i>=k; i--) {
3870: /* for (j=0; j < n; j++) */
3871: for (j=1; j <= n; j++)
3872: /* v[j][i+1] = v[j][i]; */ /* This is v[j][i+1]=v[j][i] i=kl-1 to k */
3873: v[j][i+1] = v[j][i]; /* This is v[j][i+1]=v[j][i] i=kl-1 to k */
3874: /* v[j][i+1] = v[j][i]; */
3875: /* d[i+1] = d[i];*/ /* last is d[k+1]= d[k] */
3876: d[i+1] = d[i]; /* last is d[k]= d[k-1] */
3877: }
3878: #ifdef DEBUGPRAX
3879: matprint(" after shift new conjugate vectors:",v,n,n);
3880: #endif /* d[k] = 0.0; */
3881: d[k] = 0.0;
3882: for (i=1; i <= n; i++)
3883: v[i][k] = y[i] / lds;
3884: /* v[i][k] = y[i] / lds; */
3885: #ifdef DEBUGPRAX
3886: 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);
3887: /* fprintf(ficlog,"Minimize along the new 'conjugate' direction V(*,k=%d), which is the normalized vector: (new x) - (old x).\n",k); */
3888: matprint(" before min new conjugate vectors:",v,n,n);
3889: #endif
3890: /* min(k-1, 4, &d[k-1], &lds, f1, 1); */
3891: minny(k, 4, &d[k], &lds, f1, 1);
3892: #ifdef DEBUGPRAX
3893: printf(" after min d(k)=%d %.7g lds=%14f\n",k,d[k],lds);
3894: matprint(" after min vectors:",v,n,n);
3895: #endif
3896: if (lds <= 0.0) {
3897: lds = -lds;
3898: #ifdef DEBUGPRAX
3899: printf(" lds changed sign lds=%.14f k=%d\n",lds,k);
3900: #endif
3901: /* for (i=0; i<n; i++) */
3902: /* v[i][k] = -v[i][k]; */
3903: for (i=1; i<=n; i++)
3904: v[i][k] = -v[i][k];
3905: }
3906: }
3907: ldt = ldfac * ldt;
3908: if (ldt < lds)
3909: ldt = lds;
3910: if (prin > 0){
3911: #ifdef DEBUGPRAX
3912: printf(" k=%d",k);
3913: /* fprintf(ficlog," k=%d",k); */
3914: #endif
3915: print2();/* n, x, prin, fx, nf, nl ); */
3916: }
3917: t2 = 0.0;
3918: /* for (i=0; i<n; i++) */
3919: for (i=1; i<=n; i++)
3920: t2 += x[i]*x[i];
3921: t2 = m2 * sqrt(t2) + t;
3922: /*
3923: See whether the length of the step taken since starting the
3924: inner loop exceeds half the tolerance.
3925: */
3926: #ifdef DEBUGPRAX
3927: printf("See if step length exceeds half the tolerance.\n"); /* ZZZZZ */
3928: /* fprintf(ficlog,"See if step length exceeds half the tolerance.\n"); */
3929: #endif
3930: if (ldt > (0.5 * t2))
3931: kt = 0;
3932: else
3933: kt++;
3934: #ifdef DEBUGPRAX
3935: printf("if kt=%d >? ktm=%d gotoL2 loop\n",kt,ktm);
3936: #endif
3937: if (kt > ktm){
3938: if ( 0 < prin ){
3939: /* printf("\nr8vec_print\n X:\n"); */
3940: /* fprintf(ficlog,"\nr8vec_print\n X:\n"); */
3941: vecprint ("END X:", x, n );
3942: }
3943: goto fret;
3944: }
3945: #ifdef DEBUGPRAX
3946: matprint(" end of L2 loop vectors:",v,n,n);
3947: #endif
3948:
3949: }
3950: /* printf("The inner loop ends here.\n"); */
3951: /* fprintf(ficlog,"The inner loop ends here.\n"); */
3952: /*
3953: The inner loop ends here.
3954:
3955: Try quadratic extrapolation in case we are in a curved valley.
3956: */
3957: #ifdef DEBUGPRAX
3958: printf("Try QUAD ratic extrapolation in case we are in a curved valley.\n");
3959: #endif
3960: /* try quadratic extrapolation in case */
3961: /* we are stuck in a curved valley */
3962: quad();
3963: dn = 0.0;
3964: /* for (i=0; i<n; i++) { */
3965: for (i=1; i<=n; i++) {
3966: d[i] = 1.0 / sqrt(d[i]);
3967: if (dn < d[i])
3968: dn = d[i];
3969: }
3970: if (prin > 2)
3971: matprint(" NEW DIRECTIONS vectors:",v,n,n);
3972: /* for (j=0; j<n; j++) { */
3973: for (j=1; j<=n; j++) {
3974: s = d[j] / dn;
3975: /* for (i=0; i < n; i++) */
3976: for (i=1; i <= n; i++)
3977: v[i][j] *= s;
3978: }
3979:
3980: if (scbd > 1.0) { /* scale axis to reduce condition number */
3981: #ifdef DEBUGPRAX
3982: printf("Scale the axes to try to reduce the condition number.\n");
3983: #endif
3984: /* fprintf(ficlog,"Scale the axes to try to reduce the condition number.\n"); */
3985: s = vlarge;
3986: /* for (i=0; i<n; i++) { */
3987: for (i=1; i<=n; i++) {
3988: sl = 0.0;
3989: /* for (j=0; j < n; j++) */
3990: for (j=1; j <= n; j++)
3991: sl += v[i][j]*v[i][j];
3992: z[i] = sqrt(sl);
3993: if (z[i] < m4)
3994: z[i] = m4;
3995: if (s > z[i])
3996: s = z[i];
3997: }
3998: /* for (i=0; i<n; i++) { */
3999: for (i=1; i<=n; i++) {
4000: sl = s / z[i];
4001: z[i] = 1.0 / sl;
4002: if (z[i] > scbd) {
4003: sl = 1.0 / scbd;
4004: z[i] = scbd;
4005: }
4006: }
4007: }
4008: for (i=1; i<=n; i++)
4009: /* for (j=0; j<=i-1; j++) { */
4010: /* for (j=1; j<=i; j++) { */
4011: for (j=1; j<=i-1; j++) {
4012: s = v[i][j];
4013: v[i][j] = v[j][i];
4014: v[j][i] = s;
4015: }
4016: #ifdef DEBUGPRAX
4017: printf(" Calculate a new set of orthogonal directions before repeating the main loop.\n Transpose V for MINFIT:...\n");
4018: #endif
4019: /*
4020: MINFIT finds the singular value decomposition of V.
4021:
4022: This gives the principal values and principal directions of the
4023: approximating quadratic form without squaring the condition number.
4024: */
4025: #ifdef DEBUGPRAX
4026: 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");
4027: #endif
4028:
4029: minfit(n, macheps, vsmall, v, d);
4030: /* for(i=0; i<n;i++)printf(" %14.7g",d[i]); */
4031: /* v is overwritten with R. */
4032: /*
4033: Unscale the axes.
4034: */
4035: if (scbd > 1.0) {
4036: #ifdef DEBUGPRAX
4037: printf(" Unscale the axes.\n");
4038: #endif
4039: /* for (i=0; i<n; i++) { */
4040: for (i=1; i<=n; i++) {
4041: s = z[i];
4042: /* for (j=0; j<n; j++) */
4043: for (j=1; j<=n; j++)
4044: v[i][j] *= s;
4045: }
4046: /* for (i=0; i<n; i++) { */
4047: for (i=1; i<=n; i++) {
4048: s = 0.0;
4049: /* for (j=0; j<n; j++) */
4050: for (j=1; j<=n; j++)
4051: s += v[j][i]*v[j][i];
4052: s = sqrt(s);
4053: d[i] *= s;
4054: s = 1.0 / s;
4055: /* for (j=0; j<n; j++) */
4056: for (j=1; j<=n; j++)
4057: v[j][i] *= s;
4058: }
4059: }
4060: /* for (i=0; i<n; i++) { */
4061: double dni; /* added for compatibility with buckhardt but not brent */
4062: for (i=1; i<=n; i++) {
4063: dni=dn*d[i]; /* added for compatibility with buckhardt but not brent */
4064: if ((dn * d[i]) > large)
4065: d[i] = vsmall;
4066: else if ((dn * d[i]) < small_windows)
4067: d[i] = vlarge;
4068: else
4069: d[i] = 1.0 / dni / dni; /* added for compatibility with buckhardt but not brent */
4070: /* d[i] = pow(dn * d[i],-2.0); */
4071: }
4072: #ifdef DEBUGPRAX
4073: vecprint ("\n Before sort Eigenvalues of a:",d,n );
4074: #endif
4075:
4076: sort(); /* the new eigenvalues and eigenvectors */
4077: #ifdef DEBUGPRAX
4078: vecprint( " After sort the eigenvalues ....\n", d, n);
4079: matprint( " After sort the eigenvectors....\n", v, n,n);
4080: #endif
4081: #ifdef DEBUGPRAX
4082: printf(" Determine the smallest eigenvalue.\n");
4083: #endif
4084: /* dmin = d[n-1]; */
4085: dmin = d[n];
4086: if (dmin < small_windows)
4087: dmin = small_windows;
4088: /*
4089: The ratio of the smallest to largest eigenvalue determines whether
4090: the system is ill conditioned.
4091: */
4092:
4093: /* illc = (m2 * d[0]) > dmin; */
4094: illc = (m2 * d[1]) > dmin;
4095: #ifdef DEBUGPRAX
4096: 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]);
4097: #endif
4098:
4099: if ((prin > 2) && (scbd > 1.0))
4100: vecprint("\n The scale factors:",z,n);
4101: if (prin > 2)
4102: vecprint(" Principal values (EIGEN VALUES OF A) of the quadratic form:",d,n);
4103: if (prin > 2)
4104: matprint(" The principal axes (EIGEN VECTORS OF A:",v,n, n);
4105:
4106: if ((maxfun > 0) && (nf > maxfun)) {
4107: if (prin)
4108: printf("\n... maximum number of function calls reached ...\n");
4109: goto fret;
4110: }
4111: #ifdef DEBUGPRAX
4112: printf("Goto main loop\n");
4113: #endif
4114: goto mloop; /* back to main loop */
4115:
4116: fret:
4117: if (prin > 0) {
4118: vecprint("\n X:", x, n);
4119: /* printf("\n... ChiSq reduced to %20.10e ...\n", fx); */
4120: /* printf("... after %20u function calls.\n", nf); */
4121: }
4122: free_vector(d, 1, n);
4123: free_vector(y, 1, n);
4124: free_vector(z, 1, n);
4125: free_vector(q0, 1, n);
4126: free_vector(q1, 1, n);
4127: free_matrix(v, 1, n, 1, n);
4128: /* double *d, *y, *z, */
4129: /* *q0, *q1, **v; */
4130: free_vector(tflin, 1, n);
4131: /* double *tflin; /\* used in flin: return (*fun)(tflin, n); *\/ */
4132: free_vector(e, 1, n);
4133: /* double *e; /\* used in minfit, don't konw how to free memory and thus made global *\/ */
4134:
4135: return(fx);
4136: }
4137:
4138: /* end praxis gegen */
1.126 brouard 4139:
4140: /*************** powell ************************/
1.162 brouard 4141: /*
1.317 brouard 4142: Minimization of a function func of n variables. Input consists in an initial starting point
4143: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
4144: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
4145: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 4146: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
4147: function value at p , and iter is the number of iterations taken. The routine linmin is used.
4148: */
1.224 brouard 4149: #ifdef LINMINORIGINAL
4150: #else
4151: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 4152: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 4153: #endif
1.126 brouard 4154: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
4155: double (*func)(double []))
4156: {
1.224 brouard 4157: #ifdef LINMINORIGINAL
4158: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 4159: double (*func)(double []));
1.224 brouard 4160: #else
1.241 brouard 4161: void linmin(double p[], double xi[], int n, double *fret,
4162: double (*func)(double []),int *flat);
1.224 brouard 4163: #endif
1.239 brouard 4164: int i,ibig,j,jk,k;
1.126 brouard 4165: double del,t,*pt,*ptt,*xit;
1.181 brouard 4166: double directest;
1.126 brouard 4167: double fp,fptt;
4168: double *xits;
4169: int niterf, itmp;
1.349 brouard 4170: int Bigter=0, nBigterf=1;
4171:
1.126 brouard 4172: pt=vector(1,n);
4173: ptt=vector(1,n);
4174: xit=vector(1,n);
4175: xits=vector(1,n);
4176: *fret=(*func)(p);
4177: for (j=1;j<=n;j++) pt[j]=p[j];
1.338 brouard 4178: rcurr_time = time(NULL);
4179: fp=(*fret); /* Initialisation */
1.126 brouard 4180: for (*iter=1;;++(*iter)) {
4181: ibig=0;
4182: del=0.0;
1.157 brouard 4183: rlast_time=rcurr_time;
1.349 brouard 4184: rlast_btime=rcurr_time;
1.157 brouard 4185: /* (void) gettimeofday(&curr_time,&tzp); */
4186: rcurr_time = time(NULL);
4187: curr_time = *localtime(&rcurr_time);
1.337 brouard 4188: /* 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); */
4189: /* 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 4190: /* Bigter=(*iter - *iter % ncovmodel)/ncovmodel +1; /\* Big iteration, i.e on ncovmodel cycle *\/ */
4191: Bigter=(*iter - (*iter-1) % n)/n +1; /* Big iteration, i.e on ncovmodel cycle */
1.349 brouard 4192: 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);
4193: 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);
4194: fprintf(ficrespow,"%d %d %.12f %d",*iter,Bigter, *fret,curr_time.tm_sec-start_time.tm_sec);
1.324 brouard 4195: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 4196: for (i=1;i<=n;i++) {
1.126 brouard 4197: fprintf(ficrespow," %.12lf", p[i]);
4198: }
1.239 brouard 4199: fprintf(ficrespow,"\n");fflush(ficrespow);
4200: printf("\n#model= 1 + age ");
4201: fprintf(ficlog,"\n#model= 1 + age ");
4202: if(nagesqr==1){
1.241 brouard 4203: printf(" + age*age ");
4204: fprintf(ficlog," + age*age ");
1.239 brouard 4205: }
4206: for(j=1;j <=ncovmodel-2;j++){
4207: if(Typevar[j]==0) {
4208: printf(" + V%d ",Tvar[j]);
4209: fprintf(ficlog," + V%d ",Tvar[j]);
4210: }else if(Typevar[j]==1) {
4211: printf(" + V%d*age ",Tvar[j]);
4212: fprintf(ficlog," + V%d*age ",Tvar[j]);
4213: }else if(Typevar[j]==2) {
4214: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
4215: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.349 brouard 4216: }else if(Typevar[j]==3) {
4217: printf(" + V%d*V%d*age ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
4218: fprintf(ficlog," + V%d*V%d*age ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.239 brouard 4219: }
4220: }
1.126 brouard 4221: printf("\n");
1.239 brouard 4222: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
4223: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 4224: fprintf(ficlog,"\n");
1.239 brouard 4225: for(i=1,jk=1; i <=nlstate; i++){
4226: for(k=1; k <=(nlstate+ndeath); k++){
4227: if (k != i) {
4228: printf("%d%d ",i,k);
4229: fprintf(ficlog,"%d%d ",i,k);
4230: for(j=1; j <=ncovmodel; j++){
4231: printf("%12.7f ",p[jk]);
4232: fprintf(ficlog,"%12.7f ",p[jk]);
4233: jk++;
4234: }
4235: printf("\n");
4236: fprintf(ficlog,"\n");
4237: }
4238: }
4239: }
1.241 brouard 4240: if(*iter <=3 && *iter >1){
1.157 brouard 4241: tml = *localtime(&rcurr_time);
4242: strcpy(strcurr,asctime(&tml));
4243: rforecast_time=rcurr_time;
1.126 brouard 4244: itmp = strlen(strcurr);
4245: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 4246: strcurr[itmp-1]='\0';
1.162 brouard 4247: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 4248: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.349 brouard 4249: for(nBigterf=1;nBigterf<=31;nBigterf+=10){
4250: niterf=nBigterf*ncovmodel;
4251: /* rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time); */
1.241 brouard 4252: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
4253: forecast_time = *localtime(&rforecast_time);
4254: strcpy(strfor,asctime(&forecast_time));
4255: itmp = strlen(strfor);
4256: if(strfor[itmp-1]=='\n')
4257: strfor[itmp-1]='\0';
1.349 brouard 4258: 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);
4259: 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 4260: }
4261: }
1.359 brouard 4262: for (i=1;i<=n;i++) { /* For each direction i, maximisation after loading directions */
4263: 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 */
4264:
4265: fptt=(*fret); /* Computes likelihood for parameters xit */
1.126 brouard 4266: #ifdef DEBUG
1.203 brouard 4267: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
4268: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 4269: #endif
1.203 brouard 4270: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 4271: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 4272: #ifdef LINMINORIGINAL
1.359 brouard 4273: linmin(p,xit,n,fret,func); /* New point i minimizing in direction xit, i has coordinates p[j].*/
1.357 brouard 4274: /* xit[j] gives the n coordinates of direction i as input.*/
4275: /* *fret gives the maximum value on direction xit */
1.224 brouard 4276: #else
4277: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.359 brouard 4278: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.224 brouard 4279: #endif
1.359 brouard 4280: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 4281: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.359 brouard 4282: /* because that direction will be replaced unless the gain del is small */
4283: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
4284: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
4285: /* with the new direction. */
4286: del=fabs(fptt-(*fret));
4287: ibig=i;
1.126 brouard 4288: }
4289: #ifdef DEBUG
4290: printf("%d %.12e",i,(*fret));
4291: fprintf(ficlog,"%d %.12e",i,(*fret));
4292: for (j=1;j<=n;j++) {
1.359 brouard 4293: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
4294: printf(" x(%d)=%.12e",j,xit[j]);
4295: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 4296: }
4297: for(j=1;j<=n;j++) {
1.359 brouard 4298: printf(" p(%d)=%.12e",j,p[j]);
4299: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 4300: }
4301: printf("\n");
4302: fprintf(ficlog,"\n");
4303: #endif
1.187 brouard 4304: } /* end loop on each direction i */
1.357 brouard 4305: /* Convergence test will use last linmin estimation (fret) and compare to former iteration (fp) */
1.188 brouard 4306: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.359 brouard 4307: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 4308: for(j=1;j<=n;j++) {
4309: if(flatdir[j] >0){
4310: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
4311: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 4312: }
1.319 brouard 4313: /* printf("\n"); */
4314: /* fprintf(ficlog,"\n"); */
4315: }
1.243 brouard 4316: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
4317: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 4318: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
4319: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
4320: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
4321: /* decreased of more than 3.84 */
4322: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
4323: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
4324: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 4325:
1.188 brouard 4326: /* Starting the program with initial values given by a former maximization will simply change */
4327: /* the scales of the directions and the directions, because the are reset to canonical directions */
4328: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
4329: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 4330: #ifdef DEBUG
4331: int k[2],l;
4332: k[0]=1;
4333: k[1]=-1;
4334: printf("Max: %.12e",(*func)(p));
4335: fprintf(ficlog,"Max: %.12e",(*func)(p));
4336: for (j=1;j<=n;j++) {
4337: printf(" %.12e",p[j]);
4338: fprintf(ficlog," %.12e",p[j]);
4339: }
4340: printf("\n");
4341: fprintf(ficlog,"\n");
4342: for(l=0;l<=1;l++) {
4343: for (j=1;j<=n;j++) {
4344: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
4345: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
4346: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
4347: }
4348: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
4349: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
4350: }
4351: #endif
4352:
4353: free_vector(xit,1,n);
4354: free_vector(xits,1,n);
4355: free_vector(ptt,1,n);
4356: free_vector(pt,1,n);
4357: return;
1.192 brouard 4358: } /* enough precision */
1.240 brouard 4359: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.359 brouard 4360: 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 4361: ptt[j]=2.0*p[j]-pt[j];
1.359 brouard 4362: xit[j]=p[j]-pt[j]; /* Coordinate j of last direction xi_n=P_n-P_0 */
4363: #ifdef DEBUG
4364: printf("\n %d xit=%12.7g p=%12.7g pt=%12.7g ",j,xit[j],p[j],pt[j]);
4365: #endif
4366: pt[j]=p[j]; /* New P0 is Pn */
4367: }
4368: #ifdef DEBUG
4369: printf("\n");
4370: #endif
1.181 brouard 4371: fptt=(*func)(ptt); /* f_3 */
1.359 brouard 4372: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in directions until some iterations are done */
1.224 brouard 4373: if (*iter <=4) {
1.225 brouard 4374: #else
4375: #endif
1.224 brouard 4376: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 4377: #else
1.161 brouard 4378: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 4379: #endif
1.162 brouard 4380: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 4381: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 4382: /* Let f"(x2) be the 2nd derivative equal everywhere. */
4383: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
4384: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 4385: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
4386: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
4387: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 4388: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 4389: /* Even if f3 <f1, directest can be negative and t >0 */
4390: /* mu² and del² are equal when f3=f1 */
1.359 brouard 4391: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
4392: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
4393: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
4394: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 4395: #ifdef NRCORIGINAL
4396: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
4397: #else
4398: 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 4399: t= t- del*SQR(fp-fptt);
1.183 brouard 4400: #endif
1.202 brouard 4401: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 4402: #ifdef DEBUG
1.181 brouard 4403: 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);
4404: 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 4405: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
4406: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
4407: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
4408: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
4409: 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);
4410: 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);
4411: #endif
1.183 brouard 4412: #ifdef POWELLORIGINAL
4413: if (t < 0.0) { /* Then we use it for new direction */
4414: #else
1.182 brouard 4415: if (directest*t < 0.0) { /* Contradiction between both tests */
1.359 brouard 4416: 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 4417: 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 4418: 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 4419: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
4420: }
1.181 brouard 4421: if (directest < 0.0) { /* Then we use it for new direction */
4422: #endif
1.191 brouard 4423: #ifdef DEBUGLINMIN
1.234 brouard 4424: printf("Before linmin in direction P%d-P0\n",n);
4425: for (j=1;j<=n;j++) {
4426: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
4427: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
4428: if(j % ncovmodel == 0){
4429: printf("\n");
4430: fprintf(ficlog,"\n");
4431: }
4432: }
1.224 brouard 4433: #endif
4434: #ifdef LINMINORIGINAL
1.234 brouard 4435: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 4436: #else
1.234 brouard 4437: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
4438: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 4439: #endif
1.234 brouard 4440:
1.191 brouard 4441: #ifdef DEBUGLINMIN
1.234 brouard 4442: for (j=1;j<=n;j++) {
4443: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
4444: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
4445: if(j % ncovmodel == 0){
4446: printf("\n");
4447: fprintf(ficlog,"\n");
4448: }
4449: }
1.224 brouard 4450: #endif
1.234 brouard 4451: for (j=1;j<=n;j++) {
4452: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
4453: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
4454: }
1.224 brouard 4455: #ifdef LINMINORIGINAL
4456: #else
1.234 brouard 4457: for (j=1, flatd=0;j<=n;j++) {
4458: if(flatdir[j]>0)
4459: flatd++;
4460: }
4461: if(flatd >0){
1.255 brouard 4462: printf("%d flat directions: ",flatd);
4463: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 4464: for (j=1;j<=n;j++) {
4465: if(flatdir[j]>0){
4466: printf("%d ",j);
4467: fprintf(ficlog,"%d ",j);
4468: }
4469: }
4470: printf("\n");
4471: fprintf(ficlog,"\n");
1.319 brouard 4472: #ifdef FLATSUP
4473: free_vector(xit,1,n);
4474: free_vector(xits,1,n);
4475: free_vector(ptt,1,n);
4476: free_vector(pt,1,n);
4477: return;
4478: #endif
1.234 brouard 4479: }
1.191 brouard 4480: #endif
1.234 brouard 4481: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
4482: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
4483:
1.126 brouard 4484: #ifdef DEBUG
1.234 brouard 4485: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
4486: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
4487: for(j=1;j<=n;j++){
4488: printf(" %lf",xit[j]);
4489: fprintf(ficlog," %lf",xit[j]);
4490: }
4491: printf("\n");
4492: fprintf(ficlog,"\n");
1.126 brouard 4493: #endif
1.192 brouard 4494: } /* end of t or directest negative */
1.359 brouard 4495: printf(" Directest is positive, P_n-P_0 does not increase the conjugacy. n=%d\n",n);
4496: fprintf(ficlog," Directest is positive, P_n-P_0 does not increase the conjugacy. n=%d\n",n);
1.224 brouard 4497: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 4498: #else
1.234 brouard 4499: } /* end if (fptt < fp) */
1.192 brouard 4500: #endif
1.225 brouard 4501: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 4502: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 4503: #else
1.224 brouard 4504: #endif
1.234 brouard 4505: } /* loop iteration */
1.126 brouard 4506: }
1.234 brouard 4507:
1.126 brouard 4508: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 4509:
1.235 brouard 4510: 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 4511: {
1.338 brouard 4512: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij . Nicely done
1.279 brouard 4513: * (and selected quantitative values in nres)
4514: * by left multiplying the unit
4515: * matrix by transitions matrix until convergence is reached with precision ftolpl
4516: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
4517: * Wx is row vector: population in state 1, population in state 2, population dead
4518: * or prevalence in state 1, prevalence in state 2, 0
4519: * newm is the matrix after multiplications, its rows are identical at a factor.
4520: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
4521: * Output is prlim.
4522: * Initial matrix pimij
4523: */
1.206 brouard 4524: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
4525: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
4526: /* 0, 0 , 1} */
4527: /*
4528: * and after some iteration: */
4529: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
4530: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
4531: /* 0, 0 , 1} */
4532: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
4533: /* {0.51571254859325999, 0.4842874514067399, */
4534: /* 0.51326036147820708, 0.48673963852179264} */
4535: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 4536:
1.332 brouard 4537: int i, ii,j,k, k1;
1.209 brouard 4538: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 4539: /* double **matprod2(); */ /* test */
1.218 brouard 4540: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 4541: double **newm;
1.209 brouard 4542: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 4543: int ncvloop=0;
1.288 brouard 4544: int first=0;
1.169 brouard 4545:
1.209 brouard 4546: min=vector(1,nlstate);
4547: max=vector(1,nlstate);
4548: meandiff=vector(1,nlstate);
4549:
1.218 brouard 4550: /* Starting with matrix unity */
1.126 brouard 4551: for (ii=1;ii<=nlstate+ndeath;ii++)
4552: for (j=1;j<=nlstate+ndeath;j++){
4553: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4554: }
1.169 brouard 4555:
4556: cov[1]=1.;
4557:
4558: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 4559: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 4560: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 4561: ncvloop++;
1.126 brouard 4562: newm=savm;
4563: /* Covariates have to be included here again */
1.138 brouard 4564: cov[2]=agefin;
1.319 brouard 4565: if(nagesqr==1){
4566: cov[3]= agefin*agefin;
4567: }
1.332 brouard 4568: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
4569: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
4570: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.349 brouard 4571: if(Typevar[k1]==1 || Typevar[k1]==3){ /* A product with age */
1.332 brouard 4572: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
4573: }else{
4574: cov[2+nagesqr+k1]=precov[nres][k1];
4575: }
4576: }/* End of loop on model equation */
4577:
4578: /* Start of old code (replaced by a loop on position in the model equation */
4579: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only of the model *\/ */
4580: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
4581: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; *\/ */
4582: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])]; */
4583: /* /\* model = 1 +age + V1*V3 + age*V1 + V2 + V1 + age*V2 + V3 + V3*age + V1*V2 */
4584: /* * k 1 2 3 4 5 6 7 8 */
4585: /* *cov[] 1 2 3 4 5 6 7 8 9 10 */
4586: /* *TypeVar[k] 2 1 0 0 1 0 1 2 */
4587: /* *Dummy[k] 0 2 0 0 2 0 2 0 */
4588: /* *Tvar[k] 4 1 2 1 2 3 3 5 */
4589: /* *nsd=3 (1) (2) (3) */
4590: /* *TvarsD[nsd] [1]=2 1 3 */
4591: /* *TnsdVar [2]=2 [1]=1 [3]=3 */
4592: /* *TvarsDind[nsd](=k) [1]=3 [2]=4 [3]=6 */
4593: /* *Tage[] [1]=1 [2]=2 [3]=3 */
4594: /* *Tvard[] [1][1]=1 [2][1]=1 */
4595: /* * [1][2]=3 [2][2]=2 */
4596: /* *Tprod[](=k) [1]=1 [2]=8 */
4597: /* *TvarsDp(=Tvar) [1]=1 [2]=2 [3]=3 [4]=5 */
4598: /* *TvarD (=k) [1]=1 [2]=3 [3]=4 [3]=6 [4]=6 */
4599: /* *TvarsDpType */
4600: /* *si model= 1 + age + V3 + V2*age + V2 + V3*age */
4601: /* * nsd=1 (1) (2) */
4602: /* *TvarsD[nsd] 3 2 */
4603: /* *TnsdVar (3)=1 (2)=2 */
4604: /* *TvarsDind[nsd](=k) [1]=1 [2]=3 */
4605: /* *Tage[] [1]=2 [2]= 3 */
4606: /* *\/ */
4607: /* /\* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; *\/ */
4608: /* /\* 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)); *\/ */
4609: /* } */
4610: /* for (k=1; k<=nsq;k++) { /\* For single quantitative varying covariates only of the model *\/ */
4611: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
4612: /* /\* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline *\/ */
4613: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
4614: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][resultmodel[nres][k1]] */
4615: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
4616: /* /\* 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]); *\/ */
4617: /* } */
4618: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
4619: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
4620: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
4621: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
4622: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
4623: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
4624: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
4625: /* } */
4626: /* /\* 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]); *\/ */
4627: /* } */
4628: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
4629: /* /\* 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]); *\/ */
4630: /* if(Dummy[Tvard[k][1]]==0){ */
4631: /* if(Dummy[Tvard[k][2]]==0){ */
4632: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
4633: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
4634: /* }else{ */
4635: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
4636: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
4637: /* } */
4638: /* }else{ */
4639: /* if(Dummy[Tvard[k][2]]==0){ */
4640: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
4641: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
4642: /* }else{ */
4643: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
4644: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
4645: /* } */
4646: /* } */
4647: /* } /\* End product without age *\/ */
4648: /* ENd of old code */
1.138 brouard 4649: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
4650: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
4651: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 4652: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4653: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 4654: /* age and covariate values of ij are in 'cov' */
1.142 brouard 4655: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 4656:
1.126 brouard 4657: savm=oldm;
4658: oldm=newm;
1.209 brouard 4659:
4660: for(j=1; j<=nlstate; j++){
4661: max[j]=0.;
4662: min[j]=1.;
4663: }
4664: for(i=1;i<=nlstate;i++){
4665: sumnew=0;
4666: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
4667: for(j=1; j<=nlstate; j++){
4668: prlim[i][j]= newm[i][j]/(1-sumnew);
4669: max[j]=FMAX(max[j],prlim[i][j]);
4670: min[j]=FMIN(min[j],prlim[i][j]);
4671: }
4672: }
4673:
1.126 brouard 4674: maxmax=0.;
1.209 brouard 4675: for(j=1; j<=nlstate; j++){
4676: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
4677: maxmax=FMAX(maxmax,meandiff[j]);
4678: /* 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 4679: } /* j loop */
1.203 brouard 4680: *ncvyear= (int)age- (int)agefin;
1.208 brouard 4681: /* 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 4682: if(maxmax < ftolpl){
1.209 brouard 4683: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
4684: free_vector(min,1,nlstate);
4685: free_vector(max,1,nlstate);
4686: free_vector(meandiff,1,nlstate);
1.126 brouard 4687: return prlim;
4688: }
1.288 brouard 4689: } /* agefin loop */
1.208 brouard 4690: /* After some age loop it doesn't converge */
1.288 brouard 4691: if(!first){
4692: first=1;
4693: 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 4694: 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);
4695: }else if (first >=1 && first <10){
4696: 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);
4697: first++;
4698: }else if (first ==10){
4699: 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);
4700: 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");
4701: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
4702: first++;
1.288 brouard 4703: }
4704:
1.359 brouard 4705: /* Try to lower 'ftol', for example from 1.e-8 to 6.e-9.\n", ftolpl,
4706: * (int)age, (int)delaymax, (int)agefin, ncvloop,
4707: * (int)age-(int)agefin); */
1.209 brouard 4708: free_vector(min,1,nlstate);
4709: free_vector(max,1,nlstate);
4710: free_vector(meandiff,1,nlstate);
1.208 brouard 4711:
1.169 brouard 4712: return prlim; /* should not reach here */
1.126 brouard 4713: }
4714:
1.217 brouard 4715:
4716: /**** Back Prevalence limit (stable or period prevalence) ****************/
4717:
1.218 brouard 4718: /* 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) */
4719: /* 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 4720: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 4721: {
1.264 brouard 4722: /* 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 4723: matrix by transitions matrix until convergence is reached with precision ftolpl */
4724: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
4725: /* Wx is row vector: population in state 1, population in state 2, population dead */
4726: /* or prevalence in state 1, prevalence in state 2, 0 */
4727: /* newm is the matrix after multiplications, its rows are identical at a factor */
4728: /* Initial matrix pimij */
4729: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
4730: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
4731: /* 0, 0 , 1} */
4732: /*
4733: * and after some iteration: */
4734: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
4735: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
4736: /* 0, 0 , 1} */
4737: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
4738: /* {0.51571254859325999, 0.4842874514067399, */
4739: /* 0.51326036147820708, 0.48673963852179264} */
4740: /* If we start from prlim again, prlim tends to a constant matrix */
4741:
1.359 brouard 4742: int i, ii,j, k1;
1.247 brouard 4743: int first=0;
1.217 brouard 4744: double *min, *max, *meandiff, maxmax,sumnew=0.;
4745: /* double **matprod2(); */ /* test */
4746: double **out, cov[NCOVMAX+1], **bmij();
4747: double **newm;
1.218 brouard 4748: double **dnewm, **doldm, **dsavm; /* for use */
4749: double **oldm, **savm; /* for use */
4750:
1.217 brouard 4751: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
4752: int ncvloop=0;
4753:
4754: min=vector(1,nlstate);
4755: max=vector(1,nlstate);
4756: meandiff=vector(1,nlstate);
4757:
1.266 brouard 4758: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
4759: oldm=oldms; savm=savms;
4760:
4761: /* Starting with matrix unity */
4762: for (ii=1;ii<=nlstate+ndeath;ii++)
4763: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 4764: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4765: }
4766:
4767: cov[1]=1.;
4768:
4769: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
4770: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 4771: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 4772: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
4773: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 4774: ncvloop++;
1.218 brouard 4775: newm=savm; /* oldm should be kept from previous iteration or unity at start */
4776: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 4777: /* Covariates have to be included here again */
4778: cov[2]=agefin;
1.319 brouard 4779: if(nagesqr==1){
1.217 brouard 4780: cov[3]= agefin*agefin;;
1.319 brouard 4781: }
1.332 brouard 4782: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.349 brouard 4783: if(Typevar[k1]==1 || Typevar[k1]==3){ /* A product with age */
1.332 brouard 4784: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.242 brouard 4785: }else{
1.332 brouard 4786: cov[2+nagesqr+k1]=precov[nres][k1];
1.242 brouard 4787: }
1.332 brouard 4788: }/* End of loop on model equation */
4789:
4790: /* Old code */
4791:
4792: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
4793: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
4794: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; */
4795: /* /\* 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)); *\/ */
4796: /* } */
4797: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
4798: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
4799: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
4800: /* /\* /\\* 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])]); *\\/ *\/ */
4801: /* /\* } *\/ */
4802: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
4803: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
4804: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
4805: /* /\* 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]); *\/ */
4806: /* } */
4807: /* /\* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; *\/ */
4808: /* /\* for (k=1; k<=cptcovprod;k++) /\\* Useless *\\/ *\/ */
4809: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\\/ *\/ */
4810: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
4811: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
4812: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ ERROR ???*\/ */
4813: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
4814: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
4815: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
4816: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
4817: /* } */
4818: /* /\* 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]); *\/ */
4819: /* } */
4820: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
4821: /* /\* 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]); *\/ */
4822: /* if(Dummy[Tvard[k][1]]==0){ */
4823: /* if(Dummy[Tvard[k][2]]==0){ */
4824: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
4825: /* }else{ */
4826: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
4827: /* } */
4828: /* }else{ */
4829: /* if(Dummy[Tvard[k][2]]==0){ */
4830: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
4831: /* }else{ */
4832: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
4833: /* } */
4834: /* } */
4835: /* } */
1.217 brouard 4836:
4837: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
4838: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
4839: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
4840: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4841: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 4842: /* ij should be linked to the correct index of cov */
4843: /* age and covariate values ij are in 'cov', but we need to pass
4844: * ij for the observed prevalence at age and status and covariate
4845: * number: prevacurrent[(int)agefin][ii][ij]
4846: */
4847: /* 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 *\/ */
4848: /* 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 *\/ */
4849: 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 4850: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 4851: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
4852: /* for(i=1; i<=nlstate+ndeath; i++) { */
4853: /* printf("%d newm= ",i); */
4854: /* for(j=1;j<=nlstate+ndeath;j++) { */
4855: /* printf("%f ",newm[i][j]); */
4856: /* } */
4857: /* printf("oldm * "); */
4858: /* for(j=1;j<=nlstate+ndeath;j++) { */
4859: /* printf("%f ",oldm[i][j]); */
4860: /* } */
1.268 brouard 4861: /* printf(" bmmij "); */
1.266 brouard 4862: /* for(j=1;j<=nlstate+ndeath;j++) { */
4863: /* printf("%f ",pmmij[i][j]); */
4864: /* } */
4865: /* printf("\n"); */
4866: /* } */
4867: /* } */
1.217 brouard 4868: savm=oldm;
4869: oldm=newm;
1.266 brouard 4870:
1.217 brouard 4871: for(j=1; j<=nlstate; j++){
4872: max[j]=0.;
4873: min[j]=1.;
4874: }
4875: for(j=1; j<=nlstate; j++){
4876: for(i=1;i<=nlstate;i++){
1.234 brouard 4877: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
4878: bprlim[i][j]= newm[i][j];
4879: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
4880: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 4881: }
4882: }
1.218 brouard 4883:
1.217 brouard 4884: maxmax=0.;
4885: for(i=1; i<=nlstate; i++){
1.318 brouard 4886: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 4887: maxmax=FMAX(maxmax,meandiff[i]);
4888: /* 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 4889: } /* i loop */
1.217 brouard 4890: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 4891: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 4892: if(maxmax < ftolpl){
1.220 brouard 4893: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 4894: free_vector(min,1,nlstate);
4895: free_vector(max,1,nlstate);
4896: free_vector(meandiff,1,nlstate);
4897: return bprlim;
4898: }
1.288 brouard 4899: } /* agefin loop */
1.217 brouard 4900: /* After some age loop it doesn't converge */
1.288 brouard 4901: if(!first){
1.247 brouard 4902: first=1;
4903: 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\
4904: 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);
4905: }
4906: 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 4907: 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);
4908: /* 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); */
4909: free_vector(min,1,nlstate);
4910: free_vector(max,1,nlstate);
4911: free_vector(meandiff,1,nlstate);
4912:
4913: return bprlim; /* should not reach here */
4914: }
4915:
1.126 brouard 4916: /*************** transition probabilities ***************/
4917:
4918: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
4919: {
1.138 brouard 4920: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 4921: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 4922: model to the ncovmodel covariates (including constant and age).
4923: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
4924: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
4925: ncth covariate in the global vector x is given by the formula:
4926: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
4927: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
4928: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
4929: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 4930: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 4931: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 4932: Sum on j ps[i][j] should equal to 1.
1.138 brouard 4933: */
4934: double s1, lnpijopii;
1.126 brouard 4935: /*double t34;*/
1.164 brouard 4936: int i,j, nc, ii, jj;
1.126 brouard 4937:
1.223 brouard 4938: for(i=1; i<= nlstate; i++){
4939: for(j=1; j<i;j++){
4940: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
4941: /*lnpijopii += param[i][j][nc]*cov[nc];*/
4942: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
4943: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
4944: }
4945: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 4946: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 4947: }
4948: for(j=i+1; j<=nlstate+ndeath;j++){
4949: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
4950: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
4951: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
4952: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
4953: }
4954: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 4955: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 4956: }
4957: }
1.218 brouard 4958:
1.223 brouard 4959: for(i=1; i<= nlstate; i++){
4960: s1=0;
4961: for(j=1; j<i; j++){
1.339 brouard 4962: /* printf("debug1 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 4963: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
4964: }
4965: for(j=i+1; j<=nlstate+ndeath; j++){
1.339 brouard 4966: /* printf("debug2 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 4967: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
4968: }
4969: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
4970: ps[i][i]=1./(s1+1.);
4971: /* Computing other pijs */
4972: for(j=1; j<i; j++)
1.325 brouard 4973: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 4974: for(j=i+1; j<=nlstate+ndeath; j++)
4975: ps[i][j]= exp(ps[i][j])*ps[i][i];
4976: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
4977: } /* end i */
1.218 brouard 4978:
1.223 brouard 4979: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
4980: for(jj=1; jj<= nlstate+ndeath; jj++){
4981: ps[ii][jj]=0;
4982: ps[ii][ii]=1;
4983: }
4984: }
1.294 brouard 4985:
4986:
1.223 brouard 4987: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
4988: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
4989: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
4990: /* } */
4991: /* printf("\n "); */
4992: /* } */
4993: /* printf("\n ");printf("%lf ",cov[2]);*/
4994: /*
4995: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 4996: goto end;*/
1.266 brouard 4997: return ps; /* Pointer is unchanged since its call */
1.126 brouard 4998: }
4999:
1.218 brouard 5000: /*************** backward transition probabilities ***************/
5001:
5002: /* 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 ) */
5003: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
5004: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
5005: {
1.302 brouard 5006: /* 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 5007: * 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 5008: */
1.359 brouard 5009: int ii, j;
1.222 brouard 5010:
1.359 brouard 5011: double **pmij();
1.222 brouard 5012: double sumnew=0.;
1.218 brouard 5013: double agefin;
1.292 brouard 5014: 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 5015: double **dnewm, **dsavm, **doldm;
5016: double **bbmij;
5017:
1.218 brouard 5018: doldm=ddoldms; /* global pointers */
1.222 brouard 5019: dnewm=ddnewms;
5020: dsavm=ddsavms;
1.318 brouard 5021:
5022: /* Debug */
5023: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 5024: agefin=cov[2];
1.268 brouard 5025: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 5026: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 5027: the observed prevalence (with this covariate ij) at beginning of transition */
5028: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 5029:
5030: /* P_x */
1.325 brouard 5031: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 5032: /* outputs pmmij which is a stochastic matrix in row */
5033:
5034: /* Diag(w_x) */
1.292 brouard 5035: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 5036: sumnew=0.;
1.269 brouard 5037: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 5038: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 5039: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 5040: sumnew+=prevacurrent[(int)agefin][ii][ij];
5041: }
5042: if(sumnew >0.01){ /* At least some value in the prevalence */
5043: for (ii=1;ii<=nlstate+ndeath;ii++){
5044: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 5045: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 5046: }
5047: }else{
5048: for (ii=1;ii<=nlstate+ndeath;ii++){
5049: for (j=1;j<=nlstate+ndeath;j++)
5050: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
5051: }
5052: /* if(sumnew <0.9){ */
5053: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
5054: /* } */
5055: }
5056: k3=0.0; /* We put the last diagonal to 0 */
5057: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
5058: doldm[ii][ii]= k3;
5059: }
5060: /* End doldm, At the end doldm is diag[(w_i)] */
5061:
1.292 brouard 5062: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
5063: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 5064:
1.292 brouard 5065: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 5066: /* 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 5067: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 5068: sumnew=0.;
1.222 brouard 5069: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 5070: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 5071: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 5072: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 5073: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 5074: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 5075: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 5076: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 5077: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 5078: /* }else */
1.268 brouard 5079: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
5080: } /*End ii */
5081: } /* 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 */
5082:
1.292 brouard 5083: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 5084: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 5085: /* end bmij */
1.266 brouard 5086: return ps; /*pointer is unchanged */
1.218 brouard 5087: }
1.217 brouard 5088: /*************** transition probabilities ***************/
5089:
1.218 brouard 5090: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 5091: {
5092: /* According to parameters values stored in x and the covariate's values stored in cov,
5093: computes the probability to be observed in state j being in state i by appying the
5094: model to the ncovmodel covariates (including constant and age).
5095: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
5096: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
5097: ncth covariate in the global vector x is given by the formula:
5098: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
5099: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
5100: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
5101: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
5102: Outputs ps[i][j] the probability to be observed in j being in j according to
5103: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
5104: */
5105: double s1, lnpijopii;
5106: /*double t34;*/
5107: int i,j, nc, ii, jj;
5108:
1.234 brouard 5109: for(i=1; i<= nlstate; i++){
5110: for(j=1; j<i;j++){
5111: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
5112: /*lnpijopii += param[i][j][nc]*cov[nc];*/
5113: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
5114: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
5115: }
5116: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
5117: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
5118: }
5119: for(j=i+1; j<=nlstate+ndeath;j++){
5120: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
5121: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
5122: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
5123: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
5124: }
5125: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
5126: }
5127: }
5128:
5129: for(i=1; i<= nlstate; i++){
5130: s1=0;
5131: for(j=1; j<i; j++){
5132: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
5133: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
5134: }
5135: for(j=i+1; j<=nlstate+ndeath; j++){
5136: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
5137: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
5138: }
5139: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
5140: ps[i][i]=1./(s1+1.);
5141: /* Computing other pijs */
5142: for(j=1; j<i; j++)
5143: ps[i][j]= exp(ps[i][j])*ps[i][i];
5144: for(j=i+1; j<=nlstate+ndeath; j++)
5145: ps[i][j]= exp(ps[i][j])*ps[i][i];
5146: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
5147: } /* end i */
5148:
5149: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
5150: for(jj=1; jj<= nlstate+ndeath; jj++){
5151: ps[ii][jj]=0;
5152: ps[ii][ii]=1;
5153: }
5154: }
1.296 brouard 5155: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 5156: for(jj=1; jj<= nlstate+ndeath; jj++){
5157: s1=0.;
5158: for(ii=1; ii<= nlstate+ndeath; ii++){
5159: s1+=ps[ii][jj];
5160: }
5161: for(ii=1; ii<= nlstate; ii++){
5162: ps[ii][jj]=ps[ii][jj]/s1;
5163: }
5164: }
5165: /* Transposition */
5166: for(jj=1; jj<= nlstate+ndeath; jj++){
5167: for(ii=jj; ii<= nlstate+ndeath; ii++){
5168: s1=ps[ii][jj];
5169: ps[ii][jj]=ps[jj][ii];
5170: ps[jj][ii]=s1;
5171: }
5172: }
5173: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
5174: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
5175: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
5176: /* } */
5177: /* printf("\n "); */
5178: /* } */
5179: /* printf("\n ");printf("%lf ",cov[2]);*/
5180: /*
5181: for(i=1; i<= npar; i++) printf("%f ",x[i]);
5182: goto end;*/
5183: return ps;
1.217 brouard 5184: }
5185:
5186:
1.126 brouard 5187: /**************** Product of 2 matrices ******************/
5188:
1.145 brouard 5189: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 5190: {
5191: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
5192: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
5193: /* in, b, out are matrice of pointers which should have been initialized
5194: before: only the contents of out is modified. The function returns
5195: a pointer to pointers identical to out */
1.145 brouard 5196: int i, j, k;
1.126 brouard 5197: for(i=nrl; i<= nrh; i++)
1.145 brouard 5198: for(k=ncolol; k<=ncoloh; k++){
5199: out[i][k]=0.;
5200: for(j=ncl; j<=nch; j++)
5201: out[i][k] +=in[i][j]*b[j][k];
5202: }
1.126 brouard 5203: return out;
5204: }
5205:
5206:
5207: /************* Higher Matrix Product ***************/
5208:
1.235 brouard 5209: 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 5210: {
1.336 brouard 5211: /* Already optimized with precov.
5212: 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 5213: 'nhstepm*hstepm*stepm' months (i.e. until
5214: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
5215: nhstepm*hstepm matrices.
5216: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
5217: (typically every 2 years instead of every month which is too big
5218: for the memory).
5219: Model is determined by parameters x and covariates have to be
5220: included manually here.
5221:
5222: */
5223:
1.359 brouard 5224: int i, j, d, h, k1;
1.131 brouard 5225: double **out, cov[NCOVMAX+1];
1.126 brouard 5226: double **newm;
1.187 brouard 5227: double agexact;
1.359 brouard 5228: /*double agebegin, ageend;*/
1.126 brouard 5229:
5230: /* Hstepm could be zero and should return the unit matrix */
5231: for (i=1;i<=nlstate+ndeath;i++)
5232: for (j=1;j<=nlstate+ndeath;j++){
5233: oldm[i][j]=(i==j ? 1.0 : 0.0);
5234: po[i][j][0]=(i==j ? 1.0 : 0.0);
5235: }
5236: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
5237: for(h=1; h <=nhstepm; h++){
5238: for(d=1; d <=hstepm; d++){
5239: newm=savm;
5240: /* Covariates have to be included here again */
5241: cov[1]=1.;
1.214 brouard 5242: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 5243: cov[2]=agexact;
1.319 brouard 5244: if(nagesqr==1){
1.227 brouard 5245: cov[3]= agexact*agexact;
1.319 brouard 5246: }
1.330 brouard 5247: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
5248: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
5249: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.349 brouard 5250: if(Typevar[k1]==1 || Typevar[k1]==3){ /* A product with age */
1.332 brouard 5251: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
5252: }else{
5253: cov[2+nagesqr+k1]=precov[nres][k1];
5254: }
5255: }/* End of loop on model equation */
5256: /* Old code */
5257: /* if( Dummy[k1]==0 && Typevar[k1]==0 ){ /\* Single dummy *\/ */
5258: /* /\* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) *\/ */
5259: /* /\* for (k=1; k<=nsd;k++) { /\\* For single dummy covariates only *\\/ *\/ */
5260: /* /\* /\\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\\/ *\/ */
5261: /* /\* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 *\/ */
5262: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
5263: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
5264: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
5265: /* /\* nsd 1 2 3 *\/ /\* Counting single dummies covar fixed or tv *\/ */
5266: /* /\*TvarsD[nsd] 4 3 1 *\/ /\* ID of single dummy cova fixed or timevary*\/ */
5267: /* /\*TvarsDind[k] 2 3 9 *\/ /\* position K of single dummy cova *\/ */
5268: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] *\/ */
5269: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
5270: /* /\* 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]])); *\/ */
5271: /* 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); */
5272: /* printf("hpxij new Dummy precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
5273: /* }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /\* Single quantitative variables *\/ */
5274: /* /\* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline *\/ */
5275: /* cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]]; */
5276: /* /\* for (k=1; k<=nsq;k++) { /\\* For single varying covariates only *\\/ *\/ */
5277: /* /\* /\\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\\/ *\/ */
5278: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
5279: /* 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]]); */
5280: /* printf("hpxij new Quanti precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
5281: /* }else if( Dummy[k1]==2 ){ /\* For dummy with age product *\/ */
5282: /* /\* Tvar[k1] Variable in the age product age*V1 is 1 *\/ */
5283: /* /\* [Tinvresult[nres][V1] is its value in the resultline nres *\/ */
5284: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvar[k1]]*cov[2]; */
5285: /* 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]); */
5286: /* printf("hpxij new Dummy with age product precov[nres=%d][k1=%d]=%.4f * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
5287:
5288: /* /\* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; *\/ */
5289: /* /\* for (k=1; k<=cptcovage;k++){ /\\* For product with age V1+V1*age +V4 +age*V3 *\\/ *\/ */
5290: /* /\* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*\/ */
5291: /* /\* *\/ */
1.330 brouard 5292: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
5293: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
5294: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
1.332 brouard 5295: /* /\*cptcovage=2 1 2 *\/ */
5296: /* /\*Tage[k]= 5 8 *\/ */
5297: /* }else if( Dummy[k1]==3 ){ /\* For quant with age product *\/ */
5298: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
5299: /* 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]]); */
5300: /* printf("hpxij new Quanti with age product precov[nres=%d][k1=%d] * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
5301: /* /\* if(Dummy[Tage[k]]== 2){ /\\* dummy with age *\\/ *\/ */
5302: /* /\* /\\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\\* dummy with age *\\\/ *\\/ *\/ */
5303: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\\/ *\/ */
5304: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\\/ *\/ */
5305: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
5306: /* /\* 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); *\/ */
5307: /* /\* } else if(Dummy[Tage[k]]== 3){ /\\* quantitative with age *\\/ *\/ */
5308: /* /\* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; *\/ */
5309: /* /\* } *\/ */
5310: /* /\* 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]); *\/ */
5311: /* }else if(Typevar[k1]==2 ){ /\* For product (not with age) *\/ */
5312: /* /\* for (k=1; k<=cptcovprod;k++){ /\\* For product without age *\\/ *\/ */
5313: /* /\* /\\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\\/ *\/ */
5314: /* /\* /\\* k 1 2 3 4 5 6 7 8 9 *\\/ *\/ */
5315: /* /\* /\\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\\/ *\/ */
5316: /* /\* /\\*cptcovprod=1 1 2 *\\/ *\/ */
5317: /* /\* /\\*Tprod[]= 4 7 *\\/ *\/ */
5318: /* /\* /\\*Tvard[][1] 4 1 *\\/ *\/ */
5319: /* /\* /\\*Tvard[][2] 3 2 *\\/ *\/ */
1.330 brouard 5320:
1.332 brouard 5321: /* /\* 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])]); *\/ */
5322: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
5323: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]]; */
5324: /* 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]]); */
5325: /* printf("hpxij new Product no age product precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
5326:
5327: /* /\* if(Dummy[Tvardk[k1][1]]==0){ *\/ */
5328: /* /\* if(Dummy[Tvardk[k1][2]]==0){ /\\* Product of dummies *\\/ *\/ */
5329: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
5330: /* /\* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; *\/ */
5331: /* /\* 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]])]; *\/ */
5332: /* /\* }else{ /\\* Product of dummy by quantitative *\\/ *\/ */
5333: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; *\/ */
5334: /* /\* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; *\/ */
5335: /* /\* } *\/ */
5336: /* /\* }else{ /\\* Product of quantitative by...*\\/ *\/ */
5337: /* /\* if(Dummy[Tvard[k][2]]==0){ /\\* quant by dummy *\\/ *\/ */
5338: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\\/ *\/ */
5339: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
5340: /* /\* }else{ /\\* Product of two quant *\\/ *\/ */
5341: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\\/ *\/ */
5342: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
5343: /* /\* } *\/ */
5344: /* /\* }/\\*end of products quantitative *\\/ *\/ */
5345: /* }/\*end of products *\/ */
5346: /* } /\* End of loop on model equation *\/ */
1.235 brouard 5347: /* for (k=1; k<=cptcovn;k++) */
5348: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
5349: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
5350: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
5351: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
5352: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 5353:
5354:
1.126 brouard 5355: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
5356: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 5357: /* right multiplication of oldm by the current matrix */
1.126 brouard 5358: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
5359: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 5360: /* if((int)age == 70){ */
5361: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
5362: /* for(i=1; i<=nlstate+ndeath; i++) { */
5363: /* printf("%d pmmij ",i); */
5364: /* for(j=1;j<=nlstate+ndeath;j++) { */
5365: /* printf("%f ",pmmij[i][j]); */
5366: /* } */
5367: /* printf(" oldm "); */
5368: /* for(j=1;j<=nlstate+ndeath;j++) { */
5369: /* printf("%f ",oldm[i][j]); */
5370: /* } */
5371: /* printf("\n"); */
5372: /* } */
5373: /* } */
1.126 brouard 5374: savm=oldm;
5375: oldm=newm;
5376: }
5377: for(i=1; i<=nlstate+ndeath; i++)
5378: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 5379: po[i][j][h]=newm[i][j];
5380: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 5381: }
1.128 brouard 5382: /*printf("h=%d ",h);*/
1.126 brouard 5383: } /* end h */
1.267 brouard 5384: /* printf("\n H=%d \n",h); */
1.126 brouard 5385: return po;
5386: }
5387:
1.217 brouard 5388: /************* Higher Back Matrix Product ***************/
1.218 brouard 5389: /* 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 5390: 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 5391: {
1.332 brouard 5392: /* For dummy covariates given in each resultline (for historical, computes the corresponding combination ij),
5393: computes the transition matrix starting at age 'age' over
1.217 brouard 5394: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 5395: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
5396: nhstepm*hstepm matrices.
5397: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
5398: (typically every 2 years instead of every month which is too big
1.217 brouard 5399: for the memory).
1.218 brouard 5400: Model is determined by parameters x and covariates have to be
1.266 brouard 5401: included manually here. Then we use a call to bmij(x and cov)
5402: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 5403: */
1.217 brouard 5404:
1.359 brouard 5405: int i, j, d, h, k1;
1.266 brouard 5406: double **out, cov[NCOVMAX+1], **bmij();
5407: double **newm, ***newmm;
1.217 brouard 5408: double agexact;
1.359 brouard 5409: /*double agebegin, ageend;*/
1.222 brouard 5410: double **oldm, **savm;
1.217 brouard 5411:
1.266 brouard 5412: newmm=po; /* To be saved */
5413: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 5414: /* Hstepm could be zero and should return the unit matrix */
5415: for (i=1;i<=nlstate+ndeath;i++)
5416: for (j=1;j<=nlstate+ndeath;j++){
5417: oldm[i][j]=(i==j ? 1.0 : 0.0);
5418: po[i][j][0]=(i==j ? 1.0 : 0.0);
5419: }
5420: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
5421: for(h=1; h <=nhstepm; h++){
5422: for(d=1; d <=hstepm; d++){
5423: newm=savm;
5424: /* Covariates have to be included here again */
5425: cov[1]=1.;
1.271 brouard 5426: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 5427: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 5428: /* Debug */
5429: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 5430: cov[2]=agexact;
1.332 brouard 5431: if(nagesqr==1){
1.222 brouard 5432: cov[3]= agexact*agexact;
1.332 brouard 5433: }
5434: /** New code */
5435: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.349 brouard 5436: if(Typevar[k1]==1 || Typevar[k1]==3){ /* A product with age */
1.332 brouard 5437: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.325 brouard 5438: }else{
1.332 brouard 5439: cov[2+nagesqr+k1]=precov[nres][k1];
1.325 brouard 5440: }
1.332 brouard 5441: }/* End of loop on model equation */
5442: /** End of new code */
5443: /** This was old code */
5444: /* for (k=1; k<=nsd;k++){ /\* For single dummy covariates only *\//\* cptcovn error *\/ */
5445: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
5446: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
5447: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/\* Bug valgrind *\/ */
5448: /* /\* 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)); *\/ */
5449: /* } */
5450: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
5451: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
5452: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
5453: /* /\* 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]); *\/ */
5454: /* } */
5455: /* for (k=1; k<=cptcovage;k++){ /\* Should start at cptcovn+1 *\//\* For product with age *\/ */
5456: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age error!!!*\\/ *\/ */
5457: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
5458: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
5459: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
5460: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
5461: /* } */
5462: /* /\* 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]); *\/ */
5463: /* } */
5464: /* for (k=1; k<=cptcovprod;k++){ /\* Useless because included in cptcovn *\/ */
5465: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
5466: /* if(Dummy[Tvard[k][1]]==0){ */
5467: /* if(Dummy[Tvard[k][2]]==0){ */
5468: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]; */
5469: /* }else{ */
5470: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
5471: /* } */
5472: /* }else{ */
5473: /* if(Dummy[Tvard[k][2]]==0){ */
5474: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
5475: /* }else{ */
5476: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
5477: /* } */
5478: /* } */
5479: /* } */
5480: /* /\*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*\/ */
5481: /* /\*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*\/ */
5482: /** End of old code */
5483:
1.218 brouard 5484: /* Careful transposed matrix */
1.266 brouard 5485: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 5486: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 5487: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 5488: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 5489: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 5490: /* if((int)age == 70){ */
5491: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
5492: /* for(i=1; i<=nlstate+ndeath; i++) { */
5493: /* printf("%d pmmij ",i); */
5494: /* for(j=1;j<=nlstate+ndeath;j++) { */
5495: /* printf("%f ",pmmij[i][j]); */
5496: /* } */
5497: /* printf(" oldm "); */
5498: /* for(j=1;j<=nlstate+ndeath;j++) { */
5499: /* printf("%f ",oldm[i][j]); */
5500: /* } */
5501: /* printf("\n"); */
5502: /* } */
5503: /* } */
5504: savm=oldm;
5505: oldm=newm;
5506: }
5507: for(i=1; i<=nlstate+ndeath; i++)
5508: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 5509: po[i][j][h]=newm[i][j];
1.268 brouard 5510: /* if(h==nhstepm) */
5511: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 5512: }
1.268 brouard 5513: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 5514: } /* end h */
1.268 brouard 5515: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 5516: return po;
5517: }
5518:
5519:
1.162 brouard 5520: #ifdef NLOPT
5521: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
5522: double fret;
5523: double *xt;
5524: int j;
5525: myfunc_data *d2 = (myfunc_data *) pd;
5526: /* xt = (p1-1); */
5527: xt=vector(1,n);
5528: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
5529:
5530: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
5531: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
5532: printf("Function = %.12lf ",fret);
5533: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
5534: printf("\n");
5535: free_vector(xt,1,n);
5536: return fret;
5537: }
5538: #endif
1.126 brouard 5539:
5540: /*************** log-likelihood *************/
5541: double func( double *x)
5542: {
1.336 brouard 5543: int i, ii, j, k, mi, d, kk, kf=0;
1.226 brouard 5544: int ioffset=0;
1.339 brouard 5545: int ipos=0,iposold=0,ncovv=0;
5546:
1.340 brouard 5547: double cotvarv, cotvarvold;
1.226 brouard 5548: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
5549: double **out;
5550: double lli; /* Individual log likelihood */
5551: int s1, s2;
1.228 brouard 5552: 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 5553:
1.226 brouard 5554: double bbh, survp;
5555: double agexact;
1.336 brouard 5556: double agebegin, ageend;
1.226 brouard 5557: /*extern weight */
5558: /* We are differentiating ll according to initial status */
5559: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
5560: /*for(i=1;i<imx;i++)
5561: printf(" %d\n",s[4][i]);
5562: */
1.162 brouard 5563:
1.226 brouard 5564: ++countcallfunc;
1.162 brouard 5565:
1.226 brouard 5566: cov[1]=1.;
1.126 brouard 5567:
1.226 brouard 5568: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 5569: ioffset=0;
1.226 brouard 5570: if(mle==1){
5571: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
5572: /* Computes the values of the ncovmodel covariates of the model
5573: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
5574: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
5575: to be observed in j being in i according to the model.
5576: */
1.243 brouard 5577: ioffset=2+nagesqr ;
1.233 brouard 5578: /* Fixed */
1.345 brouard 5579: for (kf=1; kf<=ncovf;kf++){ /* For each fixed covariate dummy or quant or prod */
1.319 brouard 5580: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
5581: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
5582: /* 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 5583: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.336 brouard 5584: 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 5585: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 5586: }
1.226 brouard 5587: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 5588: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 5589: has been calculated etc */
5590: /* For an individual i, wav[i] gives the number of effective waves */
5591: /* We compute the contribution to Likelihood of each effective transition
5592: mw[mi][i] is real wave of the mi th effectve wave */
5593: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
5594: s2=s[mw[mi+1][i]][i];
1.341 brouard 5595: 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 5596: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
5597: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
5598: */
1.336 brouard 5599: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
5600: /* Wave varying (but not age varying) */
1.339 brouard 5601: /* 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*\/ */
5602: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? *\/ */
5603: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
5604: /* } */
1.340 brouard 5605: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age )*/
5606: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
5607: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
1.345 brouard 5608: if(FixedV[itv]!=0){ /* Not a fixed covariate */
1.341 brouard 5609: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* cotvar[wav][ncovcol+nqv+iv][i] */
1.340 brouard 5610: }else{ /* fixed covariate */
1.345 brouard 5611: 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 5612: }
1.339 brouard 5613: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 5614: cotvarvold=cotvarv;
5615: }else{ /* A second product */
5616: cotvarv=cotvarv*cotvarvold;
1.339 brouard 5617: }
5618: iposold=ipos;
1.340 brouard 5619: cov[ioffset+ipos]=cotvarv;
1.234 brouard 5620: }
1.339 brouard 5621: /* for products of time varying to be done */
1.234 brouard 5622: for (ii=1;ii<=nlstate+ndeath;ii++)
5623: for (j=1;j<=nlstate+ndeath;j++){
5624: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
5625: savm[ii][j]=(ii==j ? 1.0 : 0.0);
5626: }
1.336 brouard 5627:
5628: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
5629: 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 5630: for(d=0; d<dh[mi][i]; d++){
5631: newm=savm;
5632: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
5633: cov[2]=agexact;
5634: if(nagesqr==1)
5635: cov[3]= agexact*agexact; /* Should be changed here */
1.349 brouard 5636: /* for (kk=1; kk<=cptcovage;kk++) { */
5637: /* if(!FixedV[Tvar[Tage[kk]]]) */
5638: /* cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /\* Tage[kk] gives the data-covariate associated with age *\/ */
5639: /* else */
5640: /* 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) *\/ */
5641: /* } */
5642: for(ncovva=1, iposold=0; ncovva <= ncovta ; ncovva++){ /* Time varying covariates with age including individual from products, product is computed dynamically */
5643: itv=TvarAVVA[ncovva]; /* TvarVV={3, 1, 3} gives the name of each varying covariate, exploding product Vn*Vm into Vn and then Vm */
5644: ipos=TvarAVVAind[ncovva]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
5645: if(FixedV[itv]!=0){ /* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv */
5646: cotvarv=cotvar[mw[mi][i]][TvarAVVA[ncovva]][i]; /* because cotvar starts now at first ncovcol+nqv+ntv+nqtv (1 to nqtv) */
5647: }else{ /* fixed covariate */
5648: cotvarv=covar[itv][i]; /* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model */
5649: }
5650: if(ipos!=iposold){ /* Not a product or first of a product */
5651: cotvarvold=cotvarv;
5652: }else{ /* A second product */
5653: cotvarv=cotvarv*cotvarvold;
5654: }
5655: iposold=ipos;
5656: cov[ioffset+ipos]=cotvarv*agexact;
5657: /* For products */
1.234 brouard 5658: }
1.349 brouard 5659:
1.234 brouard 5660: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
5661: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
5662: savm=oldm;
5663: oldm=newm;
5664: } /* end mult */
5665:
5666: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
5667: /* But now since version 0.9 we anticipate for bias at large stepm.
5668: * If stepm is larger than one month (smallest stepm) and if the exact delay
5669: * (in months) between two waves is not a multiple of stepm, we rounded to
5670: * the nearest (and in case of equal distance, to the lowest) interval but now
5671: * we keep into memory the bias bh[mi][i] and also the previous matrix product
5672: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
5673: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 5674: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
5675: * -stepm/2 to stepm/2 .
5676: * For stepm=1 the results are the same as for previous versions of Imach.
5677: * For stepm > 1 the results are less biased than in previous versions.
5678: */
1.234 brouard 5679: s1=s[mw[mi][i]][i];
5680: s2=s[mw[mi+1][i]][i];
5681: bbh=(double)bh[mi][i]/(double)stepm;
5682: /* bias bh is positive if real duration
5683: * is higher than the multiple of stepm and negative otherwise.
5684: */
5685: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
5686: if( s2 > nlstate){
5687: /* i.e. if s2 is a death state and if the date of death is known
5688: then the contribution to the likelihood is the probability to
5689: die between last step unit time and current step unit time,
5690: which is also equal to probability to die before dh
5691: minus probability to die before dh-stepm .
5692: In version up to 0.92 likelihood was computed
5693: as if date of death was unknown. Death was treated as any other
5694: health state: the date of the interview describes the actual state
5695: and not the date of a change in health state. The former idea was
5696: to consider that at each interview the state was recorded
5697: (healthy, disable or death) and IMaCh was corrected; but when we
5698: introduced the exact date of death then we should have modified
5699: the contribution of an exact death to the likelihood. This new
5700: contribution is smaller and very dependent of the step unit
5701: stepm. It is no more the probability to die between last interview
5702: and month of death but the probability to survive from last
5703: interview up to one month before death multiplied by the
5704: probability to die within a month. Thanks to Chris
5705: Jackson for correcting this bug. Former versions increased
5706: mortality artificially. The bad side is that we add another loop
5707: which slows down the processing. The difference can be up to 10%
5708: lower mortality.
5709: */
5710: /* If, at the beginning of the maximization mostly, the
5711: cumulative probability or probability to be dead is
5712: constant (ie = 1) over time d, the difference is equal to
5713: 0. out[s1][3] = savm[s1][3]: probability, being at state
5714: s1 at precedent wave, to be dead a month before current
5715: wave is equal to probability, being at state s1 at
5716: precedent wave, to be dead at mont of the current
5717: wave. Then the observed probability (that this person died)
5718: is null according to current estimated parameter. In fact,
5719: it should be very low but not zero otherwise the log go to
5720: infinity.
5721: */
1.183 brouard 5722: /* #ifdef INFINITYORIGINAL */
5723: /* lli=log(out[s1][s2] - savm[s1][s2]); */
5724: /* #else */
5725: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
5726: /* lli=log(mytinydouble); */
5727: /* else */
5728: /* lli=log(out[s1][s2] - savm[s1][s2]); */
5729: /* #endif */
1.226 brouard 5730: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 5731:
1.226 brouard 5732: } else if ( s2==-1 ) { /* alive */
5733: for (j=1,survp=0. ; j<=nlstate; j++)
5734: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
5735: /*survp += out[s1][j]; */
5736: lli= log(survp);
5737: }
1.336 brouard 5738: /* else if (s2==-4) { */
5739: /* for (j=3,survp=0. ; j<=nlstate; j++) */
5740: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
5741: /* lli= log(survp); */
5742: /* } */
5743: /* else if (s2==-5) { */
5744: /* for (j=1,survp=0. ; j<=2; j++) */
5745: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
5746: /* lli= log(survp); */
5747: /* } */
1.226 brouard 5748: else{
5749: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
5750: /* 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 */
5751: }
5752: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
5753: /*if(lli ==000.0)*/
1.340 brouard 5754: /* 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 5755: ipmx +=1;
5756: sw += weight[i];
5757: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
5758: /* if (lli < log(mytinydouble)){ */
5759: /* 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); */
5760: /* 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]); */
5761: /* } */
5762: } /* end of wave */
5763: } /* end of individual */
5764: } else if(mle==2){
5765: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 5766: ioffset=2+nagesqr ;
5767: for (k=1; k<=ncovf;k++)
5768: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 5769: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 5770: for(k=1; k <= ncovv ; k++){
1.341 brouard 5771: 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 5772: }
1.226 brouard 5773: for (ii=1;ii<=nlstate+ndeath;ii++)
5774: for (j=1;j<=nlstate+ndeath;j++){
5775: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
5776: savm[ii][j]=(ii==j ? 1.0 : 0.0);
5777: }
5778: for(d=0; d<=dh[mi][i]; d++){
5779: newm=savm;
5780: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
5781: cov[2]=agexact;
5782: if(nagesqr==1)
5783: cov[3]= agexact*agexact;
5784: for (kk=1; kk<=cptcovage;kk++) {
5785: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
5786: }
5787: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
5788: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
5789: savm=oldm;
5790: oldm=newm;
5791: } /* end mult */
5792:
5793: s1=s[mw[mi][i]][i];
5794: s2=s[mw[mi+1][i]][i];
5795: bbh=(double)bh[mi][i]/(double)stepm;
5796: 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 */
5797: ipmx +=1;
5798: sw += weight[i];
5799: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
5800: } /* end of wave */
5801: } /* end of individual */
5802: } else if(mle==3){ /* exponential inter-extrapolation */
5803: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
5804: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
5805: for(mi=1; mi<= wav[i]-1; mi++){
5806: for (ii=1;ii<=nlstate+ndeath;ii++)
5807: for (j=1;j<=nlstate+ndeath;j++){
5808: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
5809: savm[ii][j]=(ii==j ? 1.0 : 0.0);
5810: }
5811: for(d=0; d<dh[mi][i]; d++){
5812: newm=savm;
5813: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
5814: cov[2]=agexact;
5815: if(nagesqr==1)
5816: cov[3]= agexact*agexact;
5817: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 5818: if(!FixedV[Tvar[Tage[kk]]])
5819: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
5820: else
1.341 brouard 5821: 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 5822: }
5823: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
5824: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
5825: savm=oldm;
5826: oldm=newm;
5827: } /* end mult */
5828:
5829: s1=s[mw[mi][i]][i];
5830: s2=s[mw[mi+1][i]][i];
5831: bbh=(double)bh[mi][i]/(double)stepm;
5832: 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 */
5833: ipmx +=1;
5834: sw += weight[i];
5835: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
5836: } /* end of wave */
5837: } /* end of individual */
5838: }else if (mle==4){ /* ml=4 no inter-extrapolation */
5839: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
5840: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
5841: for(mi=1; mi<= wav[i]-1; mi++){
5842: for (ii=1;ii<=nlstate+ndeath;ii++)
5843: for (j=1;j<=nlstate+ndeath;j++){
5844: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
5845: savm[ii][j]=(ii==j ? 1.0 : 0.0);
5846: }
5847: for(d=0; d<dh[mi][i]; d++){
5848: newm=savm;
5849: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
5850: cov[2]=agexact;
5851: if(nagesqr==1)
5852: cov[3]= agexact*agexact;
5853: for (kk=1; kk<=cptcovage;kk++) {
5854: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
5855: }
1.126 brouard 5856:
1.226 brouard 5857: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
5858: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
5859: savm=oldm;
5860: oldm=newm;
5861: } /* end mult */
5862:
5863: s1=s[mw[mi][i]][i];
5864: s2=s[mw[mi+1][i]][i];
5865: if( s2 > nlstate){
5866: lli=log(out[s1][s2] - savm[s1][s2]);
5867: } else if ( s2==-1 ) { /* alive */
5868: for (j=1,survp=0. ; j<=nlstate; j++)
5869: survp += out[s1][j];
5870: lli= log(survp);
5871: }else{
5872: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
5873: }
5874: ipmx +=1;
5875: sw += weight[i];
5876: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.343 brouard 5877: /* 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 5878: } /* end of wave */
5879: } /* end of individual */
5880: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
5881: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
5882: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
5883: for(mi=1; mi<= wav[i]-1; mi++){
5884: for (ii=1;ii<=nlstate+ndeath;ii++)
5885: for (j=1;j<=nlstate+ndeath;j++){
5886: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
5887: savm[ii][j]=(ii==j ? 1.0 : 0.0);
5888: }
5889: for(d=0; d<dh[mi][i]; d++){
5890: newm=savm;
5891: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
5892: cov[2]=agexact;
5893: if(nagesqr==1)
5894: cov[3]= agexact*agexact;
5895: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 5896: if(!FixedV[Tvar[Tage[kk]]])
5897: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
5898: else
1.341 brouard 5899: 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 5900: }
1.126 brouard 5901:
1.226 brouard 5902: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
5903: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
5904: savm=oldm;
5905: oldm=newm;
5906: } /* end mult */
5907:
5908: s1=s[mw[mi][i]][i];
5909: s2=s[mw[mi+1][i]][i];
5910: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
5911: ipmx +=1;
5912: sw += weight[i];
5913: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
5914: /*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]);*/
5915: } /* end of wave */
5916: } /* end of individual */
5917: } /* End of if */
5918: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
5919: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
5920: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
5921: return -l;
1.126 brouard 5922: }
5923:
5924: /*************** log-likelihood *************/
5925: double funcone( double *x)
5926: {
1.228 brouard 5927: /* Same as func but slower because of a lot of printf and if */
1.359 brouard 5928: int i, ii, j, k, mi, d, kv=0, kf=0;
1.228 brouard 5929: int ioffset=0;
1.339 brouard 5930: int ipos=0,iposold=0,ncovv=0;
5931:
1.340 brouard 5932: double cotvarv, cotvarvold;
1.131 brouard 5933: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 5934: double **out;
5935: double lli; /* Individual log likelihood */
5936: double llt;
5937: int s1, s2;
1.228 brouard 5938: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
5939:
1.126 brouard 5940: double bbh, survp;
1.187 brouard 5941: double agexact;
1.214 brouard 5942: double agebegin, ageend;
1.126 brouard 5943: /*extern weight */
5944: /* We are differentiating ll according to initial status */
5945: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
5946: /*for(i=1;i<imx;i++)
5947: printf(" %d\n",s[4][i]);
5948: */
5949: cov[1]=1.;
5950:
5951: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 5952: ioffset=0;
5953: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.336 brouard 5954: /* Computes the values of the ncovmodel covariates of the model
5955: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
5956: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
5957: to be observed in j being in i according to the model.
5958: */
1.243 brouard 5959: /* ioffset=2+nagesqr+cptcovage; */
5960: ioffset=2+nagesqr;
1.232 brouard 5961: /* Fixed */
1.224 brouard 5962: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 5963: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.349 brouard 5964: 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 5965: /* printf("Debug3 TvarFind[%d]=%d",kf, TvarFind[kf]); */
5966: /* printf(" Tvar[TvarFind[kf]]=%d", Tvar[TvarFind[kf]]); */
5967: /* printf(" i=%d covar[Tvar[TvarFind[kf]]][i]=%f\n",i,covar[Tvar[TvarFind[kf]]][i]); */
1.335 brouard 5968: 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 5969: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
5970: /* cov[2+6]=covar[Tvar[6]][i]; */
5971: /* cov[2+6]=covar[2][i]; V2 */
5972: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
5973: /* cov[2+7]=covar[Tvar[7]][i]; */
5974: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
5975: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
5976: /* cov[2+9]=covar[Tvar[9]][i]; */
5977: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 5978: }
1.336 brouard 5979: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
5980: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
5981: has been calculated etc */
5982: /* For an individual i, wav[i] gives the number of effective waves */
5983: /* We compute the contribution to Likelihood of each effective transition
5984: mw[mi][i] is real wave of the mi th effectve wave */
5985: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
5986: s2=s[mw[mi+1][i]][i];
1.341 brouard 5987: And the iv th varying covariate in the DATA is the cotvar[mw[mi+1][i]][ncovcol+nqv+iv][i]
1.336 brouard 5988: */
5989: /* This part may be useless now because everythin should be in covar */
1.232 brouard 5990: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
5991: /* 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?)*\/ */
5992: /* } */
1.231 brouard 5993: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
5994: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
5995: /* } */
1.225 brouard 5996:
1.233 brouard 5997:
5998: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.339 brouard 5999: /* 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 */
6000: /* for(k=1; k <= ncovv ; k++){ /\* Varying covariates (single and product but no age )*\/ */
6001: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; *\/ */
6002: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
6003: /* } */
6004:
6005: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
6006: /* model V1+V3+age*V1+age*V3+V1*V3 */
6007: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
6008: /* TvarVV[1]=V3 (first time varying in the model equation, TvarVV[2]=V1 (in V1*V3) TvarVV[3]=3(V3) */
6009: /* We need the position of the time varying or product in the model */
6010: /* 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 */
6011: /* TvarVV gives the variable name */
1.340 brouard 6012: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
6013: * k= 1 2 3 4 5 6 7 8 9
6014: * varying 1 2 3 4 5
6015: * ncovv 1 2 3 4 5 6 7 8
1.343 brouard 6016: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
1.340 brouard 6017: * TvarVVind 2 3 7 7 8 8 9 9
6018: * TvarFind[k] 1 0 0 0 0 0 0 0 0
6019: */
1.345 brouard 6020: /* Other model ncovcol=5 nqv=0 ntv=3 nqtv=0 nlstate=3
1.349 brouard 6021: * 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 6022: * FixedV[ncovcol+qv+ntv+nqtv] V5
1.349 brouard 6023: * 3 V1 V2 V3 V4 V5 V6 V7 V8 V3*V2 V7*V2 V6*V3 V7*V3 V6*V4 V7*V4
6024: * 0 0 0 0 0 1 1 1 0, 0, 1,1, 1, 0, 1, 0, 1, 0, 1, 0}
6025: * 3 0 0 0 0 0 1 1 1 0, 1 1 1 1 1}
6026: * model= V2 + V3 + V4 + V6 + V7 + V6*V2 + V7*V2 + V6*V3 + V7*V3 + V6*V4 + V7*V4
6027: * +age*V2 +age*V3 +age*V4 +age*V6 + age*V7
6028: * +age*V6*V2 + age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4
6029: * model2= V2 + V3 + V4 + V6 + V7 + V3*V2 + V7*V2 + V6*V3 + V7*V3 + V6*V4 + V7*V4
6030: * +age*V2 +age*V3 +age*V4 +age*V6 + age*V7
6031: * +age*V3*V2 + age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4
6032: * model3= V2 + V3 + V4 + V6 + V7 + age*V3*V2 + V7*V2 + V6*V3 + V7*V3 + V6*V4 + V7*V4
6033: * +age*V2 +age*V3 +age*V4 +age*V6 + age*V7
6034: * +V3*V2 + age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4
6035: * kmodel 1 2 3 4 5 6 7 8 9 10 11
6036: * 12 13 14 15 16
6037: * 17 18 19 20 21
6038: * Tvar[kmodel] 2 3 4 6 7 9 10 11 12 13 14
6039: * 2 3 4 6 7
6040: * 9 11 12 13 14
6041: * cptcovage=5+5 total of covariates with age
6042: * Tage[cptcovage] age*V2=12 13 14 15 16
6043: *1 17 18 19 20 21 gives the position in model of covariates associated with age
6044: *3 Tage[cptcovage] age*V3*V2=6
6045: *3 age*V2=12 13 14 15 16
6046: *3 age*V6*V3=18 19 20 21
6047: * Tvar[Tage[cptcovage]] Tvar[12]=2 3 4 6 Tvar[16]=7(age*V7)
6048: * 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
6049: * 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
6050: * 3 Tvar[Tage[cptcovage]] Tvar[6]=9 Tvar[12]=2 3 4 6 Tvar[16]=7(age*V7)
6051: * 3 Tvar[18]age*V6*V3=11 age*V7*V3=12 age*V6*V4=13 Tvar[21]age*V7*V4=14
6052: * 3 Tage[cptcovage] age*V3*V2=6 age*V2=12 age*V3 13 14 15 16
6053: * age*V6*V3=18 19 20 21 gives the position in model of covariates associated with age
6054: * 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
6055: * Tvar= {2, 3, 4, 6, 7,
6056: * 9, 10, 11, 12, 13, 14,
6057: * Tvar[12]=2, 3, 4, 6, 7,
6058: * Tvar[17]=9, 11, 12, 13, 14}
6059: * Typevar[1]@21 = {0, 0, 0, 0, 0,
6060: * 2, 2, 2, 2, 2, 2,
6061: * 3 3, 2, 2, 2, 2, 2,
6062: * 1, 1, 1, 1, 1,
6063: * 3, 3, 3, 3, 3}
6064: * 3 2, 3, 3, 3, 3}
6065: * 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
6066: * p Tprod[1]@21 {6, 7, 8, 9, 10, 11, 0 <repeats 15 times>}
6067: * 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}
6068: * 3 Tprod[1]@21 {17, 7, 8, 9, 10, 11, 0 <repeats 15 times>}
6069: * cptcovprod=11 (6+5)
6070: * FixedV[Tvar[Tage[cptcovage]]]] FixedV[2]=0 FixedV[3]=0 0 1 (age*V7)Tvar[16]=1 FixedV[absolute] not [kmodel]
6071: * FixedV[Tvar[17]=FixedV[age*V6*V2]=FixedV[9]=1 1 1 1 1
6072: * 3 FixedV[Tvar[17]=FixedV[age*V3*V2]=FixedV[9]=0 [11]=1 1 1 1
6073: * FixedV[] V1=0 V2=0 V3=0 v4=0 V5=0 V6=1 V7=1 v8=1 OK then model dependent
6074: * 9=1 [V7*V2]=[10]=1 11=1 12=1 13=1 14=1
6075: * 3 9=0 [V7*V2]=[10]=1 11=1 12=1 13=1 14=1
6076: * cptcovdageprod=5 for gnuplot printing
6077: * cptcovprodvage=6
6078: * ncova=15 1 2 3 4 5
6079: * 6 7 8 9 10 11 12 13 14 15
6080: * TvarA 2 3 4 6 7
6081: * 6 2 6 7 7 3 6 4 7 4
6082: * TvaAind 12 12 13 13 14 14 15 15 16 16
1.345 brouard 6083: * ncovf 1 2 3
1.349 brouard 6084: * V6 V7 V6*V2 V7*V2 V6*V3 V7*V3 V6*V4 V7*V4
6085: * ncovvt=14 1 2 3 4 5 6 7 8 9 10 11 12 13 14
6086: * TvarVV[1]@14 = itv {V6=6, 7, V6*V2=6, 2, 7, 2, 6, 3, 7, 3, 6, 4, 7, 4}
6087: * TvarVVind[1]@14= {4, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11}
6088: * 3 ncovvt=12 V6 V7 V7*V2 V6*V3 V7*V3 V6*V4 V7*V4
6089: * 3 TvarVV[1]@12 = itv {6, 7, V7*V2=7, 2, 6, 3, 7, 3, 6, 4, 7, 4}
6090: * 3 1 2 3 4 5 6 7 8 9 10 11 12
6091: * TvarVVind[1]@12= {V6 is in k=4, 5, 7,(4isV2)=7, 8, 8, 9, 9, 10,10, 11,11}TvarVVind[12]=k=11
6092: * TvarV 6, 7, 9, 10, 11, 12, 13, 14
6093: * 3 cptcovprodvage=6
6094: * 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
6095: * 3 TvarAVVA[1]@15= itva 3 2 2 3 4 6 7 6 3 7 3 6 4 7 4
6096: * 3 ncovta 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1.354 brouard 6097: *?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 6098: * TvarAVVAind[1]@15= V3 is in k=6 6 12 13 14 15 16 18 18 19,19, 20,20 21,21}TvarVVAind[]
6099: * 3 ncovvta=10 +age*V6 + age*V7 + age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4
6100: * 3 we want to compute =cotvar[mw[mi][i]][TvarVVA[ncovva]][i] at position TvarVVAind[ncovva]
6101: * 3 TvarVVA[1]@10= itva 6 7 6 3 7 3 6 4 7 4
6102: * 3 ncovva 1 2 3 4 5 6 7 8 9 10
6103: * TvarVVAind[1]@10= V6 is in k=4 5 8,8 9, 9, 10,10 11 11}TvarVVAind[]
6104: * TvarVVAind[1]@10= 15 16 18,18 19,19, 20,20 21 21}TvarVVAind[]
6105: * TvarVA V3*V2=6 6 , 1, 2, 11, 12, 13, 14
1.345 brouard 6106: * TvarFind[1]@14= {1, 2, 3, 0 <repeats 12 times>}
1.349 brouard 6107: * Tvar[1]@21= {2, 3, 4, 6, 7, 9, 10, 11, 12, 13, 14,
6108: * 2, 3, 4, 6, 7,
6109: * 6, 8, 9, 10, 11}
1.345 brouard 6110: * TvarFind[itv] 0 0 0
6111: * FixedV[itv] 1 1 1 0 1 0 1 0 1 0 0
1.354 brouard 6112: *? FixedV[itv] 1 1 1 0 1 0 1 0 1 0 1 0 1 0
1.345 brouard 6113: * Tvar[TvarFind[ncovf]]=[1]=2 [2]=3 [4]=4
6114: * Tvar[TvarFind[itv]] [0]=? ?ncovv 1 à ncovvt]
6115: * Not a fixed cotvar[mw][itv][i] 6 7 6 2 7, 2, 6, 3, 7, 3, 6, 4, 7, 4}
1.349 brouard 6116: * fixed covar[itv] [6] [7] [6][2]
1.345 brouard 6117: */
6118:
1.349 brouard 6119: 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 */
6120: 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 6121: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
1.345 brouard 6122: /* if(TvarFind[itv]==0){ /\* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv *\/ */
6123: 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 6124: /* printf("DEBUG ncovv=%d, Varying TvarVV[ncovv]=%d\n",ncovv, TvarVV[ncovv]); */
1.345 brouard 6125: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* because cotvar starts now at first ncovcol+nqv+ntv+nqtv (1 to nqtv) */
1.354 brouard 6126: /* printf("DEBUG Varying cov[ioffset+ipos=%d]=%g \n",ioffset+ipos,cotvarv); */
1.340 brouard 6127: }else{ /* fixed covariate */
1.345 brouard 6128: /* 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 6129: /* printf("DEBUG ncovv=%d, Fixed TvarVV[ncovv]=%d\n",ncovv, TvarVV[ncovv]); */
1.349 brouard 6130: cotvarv=covar[itv][i]; /* Good: In V6*V3, 3 is fixed at position of the data */
1.354 brouard 6131: /* printf("DEBUG Fixed cov[ioffset+ipos=%d]=%g \n",ioffset+ipos,cotvarv); */
1.340 brouard 6132: }
1.339 brouard 6133: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 6134: cotvarvold=cotvarv;
6135: }else{ /* A second product */
6136: cotvarv=cotvarv*cotvarvold;
1.339 brouard 6137: }
6138: iposold=ipos;
1.340 brouard 6139: cov[ioffset+ipos]=cotvarv;
1.354 brouard 6140: /* printf("DEBUG Product cov[ioffset+ipos=%d] \n",ioffset+ipos); */
1.339 brouard 6141: /* For products */
6142: }
6143: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates single *\/ */
6144: /* iv=TvarVDind[itv]; /\* iv, position in the model equation of time varying covariate itv *\/ */
6145: /* /\* "V1+V3+age*V1+age*V3+V1*V3" with V3 time varying *\/ */
6146: /* /\* 1 2 3 4 5 *\/ */
6147: /* /\*itv 1 *\/ */
6148: /* /\* TvarVInd[1]= 2 *\/ */
6149: /* /\* iv= Tvar[Tmodelind[itv]]-ncovcol-nqv; /\\* Counting the # varying covariate from 1 to ntveff *\\/ *\/ */
6150: /* /\* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; *\/ */
6151: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; *\/ */
6152: /* /\* k=ioffset-2-nagesqr-cptcovage+itv; /\\* position in simple model *\\/ *\/ */
6153: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; *\/ */
6154: /* cov[ioffset+iv]=cotvar[mw[mi][i]][itv][i]; */
6155: /* /\* 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]); *\/ */
6156: /* } */
1.232 brouard 6157: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 6158: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
6159: /* /\* 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]); *\/ */
6160: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 6161: /* } */
1.126 brouard 6162: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 6163: for (j=1;j<=nlstate+ndeath;j++){
6164: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
6165: savm[ii][j]=(ii==j ? 1.0 : 0.0);
6166: }
1.214 brouard 6167:
6168: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
6169: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
6170: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 6171: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 6172: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
6173: and mw[mi+1][i]. dh depends on stepm.*/
6174: newm=savm;
1.247 brouard 6175: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 6176: cov[2]=agexact;
6177: if(nagesqr==1)
6178: cov[3]= agexact*agexact;
1.349 brouard 6179: for(ncovva=1, iposold=0; ncovva <= ncovta ; ncovva++){ /* Time varying covariates with age including individual from products, product is computed dynamically */
6180: itv=TvarAVVA[ncovva]; /* TvarVV={3, 1, 3} gives the name of each varying covariate, exploding product Vn*Vm into Vn and then Vm */
6181: ipos=TvarAVVAind[ncovva]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
6182: /* if(TvarFind[itv]==0){ /\* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv *\/ */
6183: if(FixedV[itv]!=0){ /* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv */
6184: /* printf("DEBUG ncovva=%d, Varying TvarAVVA[ncovva]=%d\n", ncovva, TvarAVVA[ncovva]); */
6185: cotvarv=cotvar[mw[mi][i]][TvarAVVA[ncovva]][i]; /* because cotvar starts now at first ncovcol+nqv+ntv+nqtv (1 to nqtv) */
6186: }else{ /* fixed covariate */
6187: /* cotvarv=covar[Tvar[TvarFind[itv]]][i]; /\* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model *\/ */
6188: /* printf("DEBUG ncovva=%d, Fixed TvarAVVA[ncovva]=%d\n", ncovva, TvarAVVA[ncovva]); */
6189: cotvarv=covar[itv][i]; /* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model */
6190: }
6191: if(ipos!=iposold){ /* Not a product or first of a product */
6192: cotvarvold=cotvarv;
6193: }else{ /* A second product */
6194: /* printf("DEBUG * \n"); */
6195: cotvarv=cotvarv*cotvarvold;
6196: }
6197: iposold=ipos;
6198: /* printf("DEBUG Product cov[ioffset+ipos=%d] \n",ioffset+ipos); */
6199: cov[ioffset+ipos]=cotvarv*agexact;
6200: /* For products */
1.242 brouard 6201: }
1.349 brouard 6202:
1.242 brouard 6203: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
6204: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
6205: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
6206: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
6207: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
6208: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
6209: savm=oldm;
6210: oldm=newm;
1.126 brouard 6211: } /* end mult */
1.336 brouard 6212: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
6213: /* But now since version 0.9 we anticipate for bias at large stepm.
6214: * If stepm is larger than one month (smallest stepm) and if the exact delay
6215: * (in months) between two waves is not a multiple of stepm, we rounded to
6216: * the nearest (and in case of equal distance, to the lowest) interval but now
6217: * we keep into memory the bias bh[mi][i] and also the previous matrix product
6218: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
6219: * probability in order to take into account the bias as a fraction of the way
6220: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
6221: * -stepm/2 to stepm/2 .
6222: * For stepm=1 the results are the same as for previous versions of Imach.
6223: * For stepm > 1 the results are less biased than in previous versions.
6224: */
1.126 brouard 6225: s1=s[mw[mi][i]][i];
6226: s2=s[mw[mi+1][i]][i];
1.217 brouard 6227: /* if(s2==-1){ */
1.268 brouard 6228: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 6229: /* /\* exit(1); *\/ */
6230: /* } */
1.126 brouard 6231: bbh=(double)bh[mi][i]/(double)stepm;
6232: /* bias is positive if real duration
6233: * is higher than the multiple of stepm and negative otherwise.
6234: */
6235: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 6236: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 6237: } else if ( s2==-1 ) { /* alive */
1.242 brouard 6238: for (j=1,survp=0. ; j<=nlstate; j++)
6239: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
6240: lli= log(survp);
1.126 brouard 6241: }else if (mle==1){
1.242 brouard 6242: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 6243: } else if(mle==2){
1.242 brouard 6244: 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 6245: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 6246: 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 6247: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 6248: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 6249: } else{ /* mle=0 back to 1 */
1.242 brouard 6250: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
6251: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 6252: } /* End of if */
6253: ipmx +=1;
6254: sw += weight[i];
6255: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.342 brouard 6256: /* Printing covariates values for each contribution for checking */
1.343 brouard 6257: /* 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 6258: if(globpr){
1.246 brouard 6259: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 6260: %11.6f %11.6f %11.6f ", \
1.242 brouard 6261: 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 6262: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.343 brouard 6263: /* printf("%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\ */
6264: /* %11.6f %11.6f %11.6f ", \ */
6265: /* num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw, */
6266: /* 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.242 brouard 6267: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
6268: llt +=ll[k]*gipmx/gsw;
6269: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
1.335 brouard 6270: /* printf(" %10.6f",-ll[k]*gipmx/gsw); */
1.242 brouard 6271: }
1.343 brouard 6272: fprintf(ficresilk," %10.6f ", -llt);
1.335 brouard 6273: /* printf(" %10.6f\n", -llt); */
1.342 brouard 6274: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
1.343 brouard 6275: /* fprintf(ficresilk,"%09ld ", num[i]); */ /* not necessary */
6276: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
6277: fprintf(ficresilk," %g",covar[Tvar[TvarFind[kf]]][i]);
6278: }
6279: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
6280: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
6281: if(ipos!=iposold){ /* Not a product or first of a product */
6282: fprintf(ficresilk," %g",cov[ioffset+ipos]);
6283: /* printf(" %g",cov[ioffset+ipos]); */
6284: }else{
6285: fprintf(ficresilk,"*");
6286: /* printf("*"); */
1.342 brouard 6287: }
1.343 brouard 6288: iposold=ipos;
6289: }
1.349 brouard 6290: /* for (kk=1; kk<=cptcovage;kk++) { */
6291: /* if(!FixedV[Tvar[Tage[kk]]]){ */
6292: /* fprintf(ficresilk," %g*age",covar[Tvar[Tage[kk]]][i]); */
6293: /* /\* printf(" %g*age",covar[Tvar[Tage[kk]]][i]); *\/ */
6294: /* }else{ */
6295: /* fprintf(ficresilk," %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
6296: /* /\* printf(" %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/\\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\\/ *\/ */
6297: /* } */
6298: /* } */
6299: for(ncovva=1, iposold=0; ncovva <= ncovta ; ncovva++){ /* Time varying covariates with age including individual from products, product is computed dynamically */
6300: itv=TvarAVVA[ncovva]; /* TvarVV={3, 1, 3} gives the name of each varying covariate, exploding product Vn*Vm into Vn and then Vm */
6301: ipos=TvarAVVAind[ncovva]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
6302: /* if(TvarFind[itv]==0){ /\* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv *\/ */
6303: if(FixedV[itv]!=0){ /* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv */
6304: /* printf("DEBUG ncovva=%d, Varying TvarAVVA[ncovva]=%d\n", ncovva, TvarAVVA[ncovva]); */
6305: cotvarv=cotvar[mw[mi][i]][TvarAVVA[ncovva]][i]; /* because cotvar starts now at first ncovcol+nqv+ntv+nqtv (1 to nqtv) */
6306: }else{ /* fixed covariate */
6307: /* cotvarv=covar[Tvar[TvarFind[itv]]][i]; /\* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model *\/ */
6308: /* printf("DEBUG ncovva=%d, Fixed TvarAVVA[ncovva]=%d\n", ncovva, TvarAVVA[ncovva]); */
6309: cotvarv=covar[itv][i]; /* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model */
6310: }
6311: if(ipos!=iposold){ /* Not a product or first of a product */
6312: cotvarvold=cotvarv;
6313: }else{ /* A second product */
6314: /* printf("DEBUG * \n"); */
6315: cotvarv=cotvarv*cotvarvold;
1.342 brouard 6316: }
1.349 brouard 6317: cotvarv=cotvarv*agexact;
6318: fprintf(ficresilk," %g*age",cotvarv);
6319: iposold=ipos;
6320: /* printf("DEBUG Product cov[ioffset+ipos=%d] \n",ioffset+ipos); */
6321: cov[ioffset+ipos]=cotvarv;
6322: /* For products */
1.343 brouard 6323: }
6324: /* printf("\n"); */
1.342 brouard 6325: /* } /\* End debugILK *\/ */
6326: fprintf(ficresilk,"\n");
6327: } /* End if globpr */
1.335 brouard 6328: } /* end of wave */
6329: } /* end of individual */
6330: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
1.232 brouard 6331: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
1.335 brouard 6332: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
6333: if(globpr==0){ /* First time we count the contributions and weights */
6334: gipmx=ipmx;
6335: gsw=sw;
6336: }
1.343 brouard 6337: return -l;
1.126 brouard 6338: }
6339:
6340:
6341: /*************** function likelione ***********/
1.292 brouard 6342: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 6343: {
6344: /* This routine should help understanding what is done with
6345: the selection of individuals/waves and
6346: to check the exact contribution to the likelihood.
6347: Plotting could be done.
1.342 brouard 6348: */
6349: void pstamp(FILE *ficres);
1.343 brouard 6350: int k, kf, kk, kvar, ncovv, iposold, ipos;
1.126 brouard 6351:
6352: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 6353: strcpy(fileresilk,"ILK_");
1.202 brouard 6354: strcat(fileresilk,fileresu);
1.126 brouard 6355: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
6356: printf("Problem with resultfile: %s\n", fileresilk);
6357: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
6358: }
1.342 brouard 6359: pstamp(ficresilk);fprintf(ficresilk,"# model=1+age+%s\n",model);
1.214 brouard 6360: 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");
6361: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 6362: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
6363: for(k=1; k<=nlstate; k++)
6364: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
1.342 brouard 6365: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total) ");
6366:
6367: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
6368: for(kf=1;kf <= ncovf; kf++){
6369: fprintf(ficresilk,"V%d",Tvar[TvarFind[kf]]);
6370: /* printf("V%d",Tvar[TvarFind[kf]]); */
6371: }
6372: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){
1.343 brouard 6373: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
1.342 brouard 6374: if(ipos!=iposold){ /* Not a product or first of a product */
6375: /* printf(" %d",ipos); */
6376: fprintf(ficresilk," V%d",TvarVV[ncovv]);
6377: }else{
6378: /* printf("*"); */
6379: fprintf(ficresilk,"*");
1.343 brouard 6380: }
1.342 brouard 6381: iposold=ipos;
6382: }
6383: for (kk=1; kk<=cptcovage;kk++) {
6384: if(!FixedV[Tvar[Tage[kk]]]){
6385: /* printf(" %d*age(Fixed)",Tvar[Tage[kk]]); */
6386: fprintf(ficresilk," %d*age(Fixed)",Tvar[Tage[kk]]);
6387: }else{
6388: fprintf(ficresilk," %d*age(Varying)",Tvar[Tage[kk]]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
6389: /* printf(" %d*age(Varying)",Tvar[Tage[kk]]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
6390: }
6391: }
6392: /* } /\* End if debugILK *\/ */
6393: /* printf("\n"); */
6394: fprintf(ficresilk,"\n");
6395: } /* End glogpri */
1.126 brouard 6396:
1.292 brouard 6397: *fretone=(*func)(p);
1.126 brouard 6398: if(*globpri !=0){
6399: fclose(ficresilk);
1.205 brouard 6400: if (mle ==0)
6401: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
6402: else if(mle >=1)
6403: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
6404: 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 6405: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 6406:
1.207 brouard 6407: 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 6408: <img src=\"%s-ori.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 6409: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.343 brouard 6410: <img src=\"%s-dest.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
6411:
6412: for (k=1; k<= nlstate ; k++) {
6413: 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 \
6414: <img src=\"%s-p%dj.png\">\n",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
6415: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
1.350 brouard 6416: kvar=Tvar[TvarFind[kf]]; /* variable */
6417: 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]]);
6418: 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);
6419: fprintf(fichtm,"<img src=\"%s-p%dj-%d.png\">",subdirf2(optionfilefiname,"ILK_"),k,Tvar[TvarFind[kf]]);
1.343 brouard 6420: }
6421: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Loop on the time varying extended covariates (with extension of Vn*Vm */
6422: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
6423: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
6424: /* 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]); */
6425: if(ipos!=iposold){ /* Not a product or first of a product */
6426: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
6427: /* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); */
6428: 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) */
6429: 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> \
6430: <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);
6431: } /* End only for dummies time varying (single?) */
6432: }else{ /* Useless product */
6433: /* printf("*"); */
6434: /* fprintf(ficresilk,"*"); */
6435: }
6436: iposold=ipos;
6437: } /* For each time varying covariate */
6438: } /* End loop on states */
6439:
6440: /* if(debugILK){ */
6441: /* for(kf=1; kf <= ncovf; kf++){ /\* For each simple dummy covariate of the model *\/ */
6442: /* /\* kvar=Tvar[TvarFind[kf]]; *\/ /\* variable *\/ */
6443: /* for (k=1; k<= nlstate ; k++) { */
6444: /* 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> \ */
6445: /* <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]]); */
6446: /* } */
6447: /* } */
6448: /* for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /\* Loop on the time varying extended covariates (with extension of Vn*Vm *\/ */
6449: /* ipos=TvarVVind[ncovv]; /\* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate *\/ */
6450: /* kvar=TvarVV[ncovv]; /\* TvarVV={3, 1, 3} gives the name of each varying covariate *\/ */
6451: /* /\* 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]); *\/ */
6452: /* if(ipos!=iposold){ /\* Not a product or first of a product *\/ */
6453: /* /\* fprintf(ficresilk," V%d",TvarVV[ncovv]); *\/ */
6454: /* /\* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); *\/ */
6455: /* 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) *\/ */
6456: /* for (k=1; k<= nlstate ; k++) { */
6457: /* 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> \ */
6458: /* <img src=\"%s-p%dj-%d.png\">",k,k,kvar,kvar,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,kvar); */
6459: /* } /\* End state *\/ */
6460: /* } /\* End only for dummies time varying (single?) *\/ */
6461: /* }else{ /\* Useless product *\/ */
6462: /* /\* printf("*"); *\/ */
6463: /* /\* fprintf(ficresilk,"*"); *\/ */
6464: /* } */
6465: /* iposold=ipos; */
6466: /* } /\* For each time varying covariate *\/ */
6467: /* }/\* End debugILK *\/ */
1.207 brouard 6468: fflush(fichtm);
1.343 brouard 6469: }/* End globpri */
1.126 brouard 6470: return;
6471: }
6472:
6473:
6474: /*********** Maximum Likelihood Estimation ***************/
6475:
6476: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
6477: {
1.359 brouard 6478: int i,j, jkk=0, iter=0;
1.126 brouard 6479: double **xi;
1.359 brouard 6480: /*double fret;*/
6481: /*double fretone;*/ /* Only one call to likelihood */
1.126 brouard 6482: /* char filerespow[FILENAMELENGTH];*/
1.354 brouard 6483:
1.359 brouard 6484: /*double * p1;*/ /* Shifted parameters from 0 instead of 1 */
1.162 brouard 6485: #ifdef NLOPT
6486: int creturn;
6487: nlopt_opt opt;
6488: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
6489: double *lb;
6490: double minf; /* the minimum objective value, upon return */
1.354 brouard 6491:
1.162 brouard 6492: myfunc_data dinst, *d = &dinst;
6493: #endif
6494:
6495:
1.126 brouard 6496: xi=matrix(1,npar,1,npar);
1.357 brouard 6497: for (i=1;i<=npar;i++) /* Starting with canonical directions j=1,n xi[i=1,n][j] */
1.126 brouard 6498: for (j=1;j<=npar;j++)
6499: xi[i][j]=(i==j ? 1.0 : 0.0);
1.359 brouard 6500: printf("Powell-prax\n"); fprintf(ficlog,"Powell-prax\n");
1.201 brouard 6501: strcpy(filerespow,"POW_");
1.126 brouard 6502: strcat(filerespow,fileres);
6503: if((ficrespow=fopen(filerespow,"w"))==NULL) {
6504: printf("Problem with resultfile: %s\n", filerespow);
6505: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
6506: }
6507: fprintf(ficrespow,"# Powell\n# iter -2*LL");
6508: for (i=1;i<=nlstate;i++)
6509: for(j=1;j<=nlstate+ndeath;j++)
6510: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
6511: fprintf(ficrespow,"\n");
1.162 brouard 6512: #ifdef POWELL
1.319 brouard 6513: #ifdef LINMINORIGINAL
6514: #else /* LINMINORIGINAL */
6515:
6516: flatdir=ivector(1,npar);
6517: for (j=1;j<=npar;j++) flatdir[j]=0;
6518: #endif /*LINMINORIGINAL */
6519:
6520: #ifdef FLATSUP
6521: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
6522: /* reorganizing p by suppressing flat directions */
6523: for(i=1, jk=1; i <=nlstate; i++){
6524: for(k=1; k <=(nlstate+ndeath); k++){
6525: if (k != i) {
6526: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
6527: if(flatdir[jk]==1){
6528: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
6529: }
6530: for(j=1; j <=ncovmodel; j++){
6531: printf("%12.7f ",p[jk]);
6532: jk++;
6533: }
6534: printf("\n");
6535: }
6536: }
6537: }
6538: /* skipping */
6539: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
6540: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
6541: for(k=1; k <=(nlstate+ndeath); k++){
6542: if (k != i) {
6543: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
6544: if(flatdir[jk]==1){
6545: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
6546: for(j=1; j <=ncovmodel; jk++,j++){
6547: printf(" p[%d]=%12.7f",jk, p[jk]);
6548: /*q[jjk]=p[jk];*/
6549: }
6550: }else{
6551: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
6552: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
6553: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
6554: /*q[jjk]=p[jk];*/
6555: }
6556: }
6557: printf("\n");
6558: }
6559: fflush(stdout);
6560: }
6561: }
6562: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
6563: #else /* FLATSUP */
1.359 brouard 6564: /* powell(p,xi,npar,ftol,&iter,&fret,func);*/
6565: /* praxis ( t0, h0, n, prin, x, beale_f ); */
6566: int prin=1;
6567: double h0=0.25;
6568: double macheps;
6569: double fmin;
6570: macheps=pow(16.0,-13.0);
6571: /* #include "praxis.h" */
6572: /* Be careful that praxis start at x[0] and powell start at p[1] */
6573: /* praxis ( ftol, h0, npar, prin, p, func ); */
6574: /* p1= (p+1); */ /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
6575: printf("Praxis Gegenfurtner \n");
6576: fprintf(ficlog, "Praxis Gegenfurtner\n");fflush(ficlog);
6577: /* praxis ( ftol, h0, npar, prin, p1, func ); */
6578: /* fmin = praxis(1.e-5,macheps, h, n, prin, x, func); */
6579: fmin = praxis(ftol,macheps, h0, npar, prin, p, func);
6580: printf("End Praxis\n");
1.319 brouard 6581: #endif /* FLATSUP */
6582:
6583: #ifdef LINMINORIGINAL
6584: #else
6585: free_ivector(flatdir,1,npar);
6586: #endif /* LINMINORIGINAL*/
6587: #endif /* POWELL */
1.126 brouard 6588:
1.162 brouard 6589: #ifdef NLOPT
6590: #ifdef NEWUOA
6591: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
6592: #else
6593: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
6594: #endif
6595: lb=vector(0,npar-1);
6596: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
6597: nlopt_set_lower_bounds(opt, lb);
6598: nlopt_set_initial_step1(opt, 0.1);
6599:
6600: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
6601: d->function = func;
6602: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
6603: nlopt_set_min_objective(opt, myfunc, d);
6604: nlopt_set_xtol_rel(opt, ftol);
6605: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
6606: printf("nlopt failed! %d\n",creturn);
6607: }
6608: else {
6609: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
6610: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
6611: iter=1; /* not equal */
6612: }
6613: nlopt_destroy(opt);
6614: #endif
1.319 brouard 6615: #ifdef FLATSUP
6616: /* npared = npar -flatd/ncovmodel; */
6617: /* xired= matrix(1,npared,1,npared); */
6618: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
6619: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
6620: /* free_matrix(xire,1,npared,1,npared); */
6621: #else /* FLATSUP */
6622: #endif /* FLATSUP */
1.126 brouard 6623: free_matrix(xi,1,npar,1,npar);
6624: fclose(ficrespow);
1.203 brouard 6625: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
6626: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 6627: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 6628:
6629: }
6630:
6631: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 6632: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 6633: {
6634: double **a,**y,*x,pd;
1.203 brouard 6635: /* double **hess; */
1.164 brouard 6636: int i, j;
1.126 brouard 6637: int *indx;
6638:
6639: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 6640: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 6641: void lubksb(double **a, int npar, int *indx, double b[]) ;
6642: void ludcmp(double **a, int npar, int *indx, double *d) ;
6643: double gompertz(double p[]);
1.203 brouard 6644: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 6645:
6646: printf("\nCalculation of the hessian matrix. Wait...\n");
6647: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
6648: for (i=1;i<=npar;i++){
1.203 brouard 6649: printf("%d-",i);fflush(stdout);
6650: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 6651:
6652: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
6653:
6654: /* printf(" %f ",p[i]);
6655: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
6656: }
6657:
6658: for (i=1;i<=npar;i++) {
6659: for (j=1;j<=npar;j++) {
6660: if (j>i) {
1.203 brouard 6661: printf(".%d-%d",i,j);fflush(stdout);
6662: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
6663: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 6664:
6665: hess[j][i]=hess[i][j];
6666: /*printf(" %lf ",hess[i][j]);*/
6667: }
6668: }
6669: }
6670: printf("\n");
6671: fprintf(ficlog,"\n");
6672:
6673: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
6674: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
6675:
6676: a=matrix(1,npar,1,npar);
6677: y=matrix(1,npar,1,npar);
6678: x=vector(1,npar);
6679: indx=ivector(1,npar);
6680: for (i=1;i<=npar;i++)
6681: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
6682: ludcmp(a,npar,indx,&pd);
6683:
6684: for (j=1;j<=npar;j++) {
6685: for (i=1;i<=npar;i++) x[i]=0;
6686: x[j]=1;
6687: lubksb(a,npar,indx,x);
6688: for (i=1;i<=npar;i++){
6689: matcov[i][j]=x[i];
6690: }
6691: }
6692:
6693: printf("\n#Hessian matrix#\n");
6694: fprintf(ficlog,"\n#Hessian matrix#\n");
6695: for (i=1;i<=npar;i++) {
6696: for (j=1;j<=npar;j++) {
1.203 brouard 6697: printf("%.6e ",hess[i][j]);
6698: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 6699: }
6700: printf("\n");
6701: fprintf(ficlog,"\n");
6702: }
6703:
1.203 brouard 6704: /* printf("\n#Covariance matrix#\n"); */
6705: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
6706: /* for (i=1;i<=npar;i++) { */
6707: /* for (j=1;j<=npar;j++) { */
6708: /* printf("%.6e ",matcov[i][j]); */
6709: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
6710: /* } */
6711: /* printf("\n"); */
6712: /* fprintf(ficlog,"\n"); */
6713: /* } */
6714:
1.126 brouard 6715: /* Recompute Inverse */
1.203 brouard 6716: /* for (i=1;i<=npar;i++) */
6717: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
6718: /* ludcmp(a,npar,indx,&pd); */
6719:
6720: /* printf("\n#Hessian matrix recomputed#\n"); */
6721:
6722: /* for (j=1;j<=npar;j++) { */
6723: /* for (i=1;i<=npar;i++) x[i]=0; */
6724: /* x[j]=1; */
6725: /* lubksb(a,npar,indx,x); */
6726: /* for (i=1;i<=npar;i++){ */
6727: /* y[i][j]=x[i]; */
6728: /* printf("%.3e ",y[i][j]); */
6729: /* fprintf(ficlog,"%.3e ",y[i][j]); */
6730: /* } */
6731: /* printf("\n"); */
6732: /* fprintf(ficlog,"\n"); */
6733: /* } */
6734:
6735: /* Verifying the inverse matrix */
6736: #ifdef DEBUGHESS
6737: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 6738:
1.203 brouard 6739: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
6740: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 6741:
6742: for (j=1;j<=npar;j++) {
6743: for (i=1;i<=npar;i++){
1.203 brouard 6744: printf("%.2f ",y[i][j]);
6745: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 6746: }
6747: printf("\n");
6748: fprintf(ficlog,"\n");
6749: }
1.203 brouard 6750: #endif
1.126 brouard 6751:
6752: free_matrix(a,1,npar,1,npar);
6753: free_matrix(y,1,npar,1,npar);
6754: free_vector(x,1,npar);
6755: free_ivector(indx,1,npar);
1.203 brouard 6756: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 6757:
6758:
6759: }
6760:
6761: /*************** hessian matrix ****************/
6762: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 6763: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 6764: int i;
6765: int l=1, lmax=20;
1.203 brouard 6766: double k1,k2, res, fx;
1.132 brouard 6767: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 6768: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
6769: int k=0,kmax=10;
6770: double l1;
6771:
6772: fx=func(x);
6773: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 6774: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 6775: l1=pow(10,l);
6776: delts=delt;
6777: for(k=1 ; k <kmax; k=k+1){
6778: delt = delta*(l1*k);
6779: p2[theta]=x[theta] +delt;
1.145 brouard 6780: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 6781: p2[theta]=x[theta]-delt;
6782: k2=func(p2)-fx;
6783: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 6784: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 6785:
1.203 brouard 6786: #ifdef DEBUGHESSII
1.126 brouard 6787: 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);
6788: 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);
6789: #endif
6790: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
6791: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
6792: k=kmax;
6793: }
6794: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 6795: k=kmax; l=lmax*10;
1.126 brouard 6796: }
6797: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
6798: delts=delt;
6799: }
1.203 brouard 6800: } /* End loop k */
1.126 brouard 6801: }
6802: delti[theta]=delts;
6803: return res;
6804:
6805: }
6806:
1.203 brouard 6807: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 6808: {
6809: int i;
1.164 brouard 6810: int l=1, lmax=20;
1.126 brouard 6811: double k1,k2,k3,k4,res,fx;
1.132 brouard 6812: double p2[MAXPARM+1];
1.203 brouard 6813: int k, kmax=1;
6814: double v1, v2, cv12, lc1, lc2;
1.208 brouard 6815:
6816: int firstime=0;
1.203 brouard 6817:
1.126 brouard 6818: fx=func(x);
1.203 brouard 6819: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 6820: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 6821: p2[thetai]=x[thetai]+delti[thetai]*k;
6822: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 6823: k1=func(p2)-fx;
6824:
1.203 brouard 6825: p2[thetai]=x[thetai]+delti[thetai]*k;
6826: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 6827: k2=func(p2)-fx;
6828:
1.203 brouard 6829: p2[thetai]=x[thetai]-delti[thetai]*k;
6830: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 6831: k3=func(p2)-fx;
6832:
1.203 brouard 6833: p2[thetai]=x[thetai]-delti[thetai]*k;
6834: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 6835: k4=func(p2)-fx;
1.203 brouard 6836: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
6837: if(k1*k2*k3*k4 <0.){
1.208 brouard 6838: firstime=1;
1.203 brouard 6839: kmax=kmax+10;
1.208 brouard 6840: }
6841: if(kmax >=10 || firstime ==1){
1.354 brouard 6842: /* What are the thetai and thetaj? thetai/ncovmodel thetai=(thetai-thetai%ncovmodel)/ncovmodel +thetai%ncovmodel=(line,pos) */
1.246 brouard 6843: 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);
6844: 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 6845: 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);
6846: 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);
6847: }
6848: #ifdef DEBUGHESSIJ
6849: v1=hess[thetai][thetai];
6850: v2=hess[thetaj][thetaj];
6851: cv12=res;
6852: /* Computing eigen value of Hessian matrix */
6853: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6854: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6855: if ((lc2 <0) || (lc1 <0) ){
6856: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
6857: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
6858: 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);
6859: 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);
6860: }
1.126 brouard 6861: #endif
6862: }
6863: return res;
6864: }
6865:
1.203 brouard 6866: /* Not done yet: Was supposed to fix if not exactly at the maximum */
6867: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
6868: /* { */
6869: /* int i; */
6870: /* int l=1, lmax=20; */
6871: /* double k1,k2,k3,k4,res,fx; */
6872: /* double p2[MAXPARM+1]; */
6873: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
6874: /* int k=0,kmax=10; */
6875: /* double l1; */
6876:
6877: /* fx=func(x); */
6878: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
6879: /* l1=pow(10,l); */
6880: /* delts=delt; */
6881: /* for(k=1 ; k <kmax; k=k+1){ */
6882: /* delt = delti*(l1*k); */
6883: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
6884: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
6885: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
6886: /* k1=func(p2)-fx; */
6887:
6888: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
6889: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
6890: /* k2=func(p2)-fx; */
6891:
6892: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
6893: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
6894: /* k3=func(p2)-fx; */
6895:
6896: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
6897: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
6898: /* k4=func(p2)-fx; */
6899: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
6900: /* #ifdef DEBUGHESSIJ */
6901: /* 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); */
6902: /* 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); */
6903: /* #endif */
6904: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
6905: /* k=kmax; */
6906: /* } */
6907: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
6908: /* k=kmax; l=lmax*10; */
6909: /* } */
6910: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
6911: /* delts=delt; */
6912: /* } */
6913: /* } /\* End loop k *\/ */
6914: /* } */
6915: /* delti[theta]=delts; */
6916: /* return res; */
6917: /* } */
6918:
6919:
1.126 brouard 6920: /************** Inverse of matrix **************/
6921: void ludcmp(double **a, int n, int *indx, double *d)
6922: {
6923: int i,imax,j,k;
6924: double big,dum,sum,temp;
6925: double *vv;
6926:
6927: vv=vector(1,n);
6928: *d=1.0;
6929: for (i=1;i<=n;i++) {
6930: big=0.0;
6931: for (j=1;j<=n;j++)
6932: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 6933: if (big == 0.0){
6934: printf(" Singular Hessian matrix at row %d:\n",i);
6935: for (j=1;j<=n;j++) {
6936: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
6937: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
6938: }
6939: fflush(ficlog);
6940: fclose(ficlog);
6941: nrerror("Singular matrix in routine ludcmp");
6942: }
1.126 brouard 6943: vv[i]=1.0/big;
6944: }
6945: for (j=1;j<=n;j++) {
6946: for (i=1;i<j;i++) {
6947: sum=a[i][j];
6948: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
6949: a[i][j]=sum;
6950: }
6951: big=0.0;
6952: for (i=j;i<=n;i++) {
6953: sum=a[i][j];
6954: for (k=1;k<j;k++)
6955: sum -= a[i][k]*a[k][j];
6956: a[i][j]=sum;
6957: if ( (dum=vv[i]*fabs(sum)) >= big) {
6958: big=dum;
6959: imax=i;
6960: }
6961: }
6962: if (j != imax) {
6963: for (k=1;k<=n;k++) {
6964: dum=a[imax][k];
6965: a[imax][k]=a[j][k];
6966: a[j][k]=dum;
6967: }
6968: *d = -(*d);
6969: vv[imax]=vv[j];
6970: }
6971: indx[j]=imax;
6972: if (a[j][j] == 0.0) a[j][j]=TINY;
6973: if (j != n) {
6974: dum=1.0/(a[j][j]);
6975: for (i=j+1;i<=n;i++) a[i][j] *= dum;
6976: }
6977: }
6978: free_vector(vv,1,n); /* Doesn't work */
6979: ;
6980: }
6981:
6982: void lubksb(double **a, int n, int *indx, double b[])
6983: {
6984: int i,ii=0,ip,j;
6985: double sum;
6986:
6987: for (i=1;i<=n;i++) {
6988: ip=indx[i];
6989: sum=b[ip];
6990: b[ip]=b[i];
6991: if (ii)
6992: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
6993: else if (sum) ii=i;
6994: b[i]=sum;
6995: }
6996: for (i=n;i>=1;i--) {
6997: sum=b[i];
6998: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
6999: b[i]=sum/a[i][i];
7000: }
7001: }
7002:
7003: void pstamp(FILE *fichier)
7004: {
1.196 brouard 7005: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 7006: }
7007:
1.297 brouard 7008: void date2dmy(double date,double *day, double *month, double *year){
7009: double yp=0., yp1=0., yp2=0.;
7010:
7011: yp1=modf(date,&yp);/* extracts integral of date in yp and
7012: fractional in yp1 */
7013: *year=yp;
7014: yp2=modf((yp1*12),&yp);
7015: *month=yp;
7016: yp1=modf((yp2*30.5),&yp);
7017: *day=yp;
7018: if(*day==0) *day=1;
7019: if(*month==0) *month=1;
7020: }
7021:
1.253 brouard 7022:
7023:
1.126 brouard 7024: /************ Frequencies ********************/
1.251 brouard 7025: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 7026: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
7027: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 7028: { /* Some frequencies as well as proposing some starting values */
1.332 brouard 7029: /* Frequencies of any combination of dummy covariate used in the model equation */
1.265 brouard 7030: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 7031: int iind=0, iage=0;
7032: int mi; /* Effective wave */
7033: int first;
7034: double ***freq; /* Frequencies */
1.268 brouard 7035: 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 */
7036: 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 7037: double *meanq, *stdq, *idq;
1.226 brouard 7038: double **meanqt;
7039: double *pp, **prop, *posprop, *pospropt;
7040: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
7041: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
7042: double agebegin, ageend;
7043:
7044: pp=vector(1,nlstate);
1.251 brouard 7045: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 7046: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
7047: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
7048: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
7049: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 7050: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 7051: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 7052: meanqt=matrix(1,lastpass,1,nqtveff);
7053: strcpy(fileresp,"P_");
7054: strcat(fileresp,fileresu);
7055: /*strcat(fileresphtm,fileresu);*/
7056: if((ficresp=fopen(fileresp,"w"))==NULL) {
7057: printf("Problem with prevalence resultfile: %s\n", fileresp);
7058: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
7059: exit(0);
7060: }
1.240 brouard 7061:
1.226 brouard 7062: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
7063: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
7064: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
7065: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
7066: fflush(ficlog);
7067: exit(70);
7068: }
7069: else{
7070: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 7071: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 7072: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 7073: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
7074: }
1.319 brouard 7075: 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 7076:
1.226 brouard 7077: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
7078: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
7079: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
7080: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
7081: fflush(ficlog);
7082: exit(70);
1.240 brouard 7083: } else{
1.226 brouard 7084: 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 7085: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 7086: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 7087: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
7088: }
1.319 brouard 7089: 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 7090:
1.253 brouard 7091: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
7092: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 7093: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 7094: j1=0;
1.126 brouard 7095:
1.227 brouard 7096: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
1.335 brouard 7097: j=cptcoveff; /* Only simple dummy covariates used in the model */
1.330 brouard 7098: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 7099: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 7100:
7101:
1.226 brouard 7102: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
7103: reference=low_education V1=0,V2=0
7104: med_educ V1=1 V2=0,
7105: high_educ V1=0 V2=1
1.330 brouard 7106: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 7107: */
1.249 brouard 7108: dateintsum=0;
7109: k2cpt=0;
7110:
1.253 brouard 7111: if(cptcoveff == 0 )
1.265 brouard 7112: nl=1; /* Constant and age model only */
1.253 brouard 7113: else
7114: nl=2;
1.265 brouard 7115:
7116: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
7117: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.335 brouard 7118: * Loop on j1(1 to 2**cptcoveff) covariate combination
1.265 brouard 7119: * freq[s1][s2][iage] =0.
7120: * Loop on iind
7121: * ++freq[s1][s2][iage] weighted
7122: * end iind
7123: * if covariate and j!0
7124: * headers Variable on one line
7125: * endif cov j!=0
7126: * header of frequency table by age
7127: * Loop on age
7128: * pp[s1]+=freq[s1][s2][iage] weighted
7129: * pos+=freq[s1][s2][iage] weighted
7130: * Loop on s1 initial state
7131: * fprintf(ficresp
7132: * end s1
7133: * end age
7134: * if j!=0 computes starting values
7135: * end compute starting values
7136: * end j1
7137: * end nl
7138: */
1.253 brouard 7139: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
7140: if(nj==1)
7141: j=0; /* First pass for the constant */
1.265 brouard 7142: else{
1.335 brouard 7143: 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 7144: }
1.251 brouard 7145: first=1;
1.332 brouard 7146: 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 7147: posproptt=0.;
1.330 brouard 7148: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 7149: scanf("%d", i);*/
7150: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 7151: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 7152: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 7153: freq[i][s2][m]=0;
1.251 brouard 7154:
7155: for (i=1; i<=nlstate; i++) {
1.240 brouard 7156: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 7157: prop[i][m]=0;
7158: posprop[i]=0;
7159: pospropt[i]=0;
7160: }
1.283 brouard 7161: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 7162: idq[z1]=0.;
7163: meanq[z1]=0.;
7164: stdq[z1]=0.;
1.283 brouard 7165: }
7166: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 7167: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 7168: /* meanqt[m][z1]=0.; */
7169: /* } */
7170: /* } */
1.251 brouard 7171: /* dateintsum=0; */
7172: /* k2cpt=0; */
7173:
1.265 brouard 7174: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 7175: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
7176: bool=1;
7177: if(j !=0){
7178: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.335 brouard 7179: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
7180: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
1.251 brouard 7181: /* if(Tvaraff[z1] ==-20){ */
7182: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
7183: /* }else if(Tvaraff[z1] ==-10){ */
7184: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 brouard 7185: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
1.335 brouard 7186: /* 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); */
7187: if(Tvaraff[z1]<1 || Tvaraff[z1]>=NCOVMAX)
1.338 brouard 7188: printf("Error Tvaraff[z1]=%d<1 or >=%d, cptcoveff=%d model=1+age+%s\n",Tvaraff[z1],NCOVMAX, cptcoveff, model);
1.332 brouard 7189: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]){ /* for combination j1 of covariates */
1.265 brouard 7190: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 7191: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.332 brouard 7192: /* 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", */
7193: /* bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),*/
7194: /* j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
1.251 brouard 7195: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
7196: } /* Onlyf fixed */
7197: } /* end z1 */
1.335 brouard 7198: } /* cptcoveff > 0 */
1.251 brouard 7199: } /* end any */
7200: }/* end j==0 */
1.265 brouard 7201: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 7202: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 7203: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 7204: m=mw[mi][iind];
7205: if(j!=0){
7206: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.335 brouard 7207: for (z1=1; z1<=cptcoveff; z1++) {
1.251 brouard 7208: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 7209: /* iv= Tvar[Tmodelind[z1]]-ncovcol-nqv; /\* Good *\/ */
7210: iv= Tvar[Tmodelind[z1]]; /* Good *//* because cotvar starts now at first at ncovcol+nqv+ntv */
1.332 brouard 7211: 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 7212: value is -1, we don't select. It differs from the
7213: constant and age model which counts them. */
7214: bool=0; /* not selected */
7215: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.334 brouard 7216: /* i1=Tvaraff[z1]; */
7217: /* i2=TnsdVar[i1]; */
7218: /* i3=nbcode[i1][i2]; */
7219: /* i4=covar[i1][iind]; */
7220: /* if(i4 != i3){ */
7221: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) { /* Bug valgrind */
1.251 brouard 7222: bool=0;
7223: }
7224: }
7225: }
7226: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
7227: } /* end j==0 */
7228: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 7229: if(bool==1){ /*Selected */
1.251 brouard 7230: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
7231: and mw[mi+1][iind]. dh depends on stepm. */
7232: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
7233: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
7234: if(m >=firstpass && m <=lastpass){
7235: k2=anint[m][iind]+(mint[m][iind]/12.);
7236: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
7237: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
7238: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
7239: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
7240: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
7241: if (m<lastpass) {
7242: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
7243: /* 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]); */
7244: if(s[m][iind]==-1)
7245: 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.));
7246: 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 7247: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
7248: if(!isnan(covar[ncovcol+z1][iind])){
1.332 brouard 7249: idq[z1]=idq[z1]+weight[iind];
7250: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
7251: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
7252: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
1.311 brouard 7253: }
1.284 brouard 7254: }
1.251 brouard 7255: /* if((int)agev[m][iind] == 55) */
7256: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
7257: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
7258: 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 7259: }
1.251 brouard 7260: } /* end if between passes */
7261: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
7262: dateintsum=dateintsum+k2; /* on all covariates ?*/
7263: k2cpt++;
7264: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 7265: }
1.251 brouard 7266: }else{
7267: bool=1;
7268: }/* end bool 2 */
7269: } /* end m */
1.284 brouard 7270: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
7271: /* idq[z1]=idq[z1]+weight[iind]; */
7272: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
7273: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
7274: /* } */
1.251 brouard 7275: } /* end bool */
7276: } /* end iind = 1 to imx */
1.319 brouard 7277: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 7278: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
7279:
7280:
7281: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.335 brouard 7282: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
1.265 brouard 7283: pstamp(ficresp);
1.335 brouard 7284: if (cptcoveff>0 && j!=0){
1.265 brouard 7285: pstamp(ficresp);
1.251 brouard 7286: printf( "\n#********** Variable ");
7287: fprintf(ficresp, "\n#********** Variable ");
7288: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
7289: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
7290: fprintf(ficlog, "\n#********** Variable ");
1.340 brouard 7291: for (z1=1; z1<=cptcoveff; z1++){
1.251 brouard 7292: if(!FixedV[Tvaraff[z1]]){
1.330 brouard 7293: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7294: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7295: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7296: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7297: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 7298: }else{
1.330 brouard 7299: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7300: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7301: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7302: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7303: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 7304: }
7305: }
7306: printf( "**********\n#");
7307: fprintf(ficresp, "**********\n#");
7308: fprintf(ficresphtm, "**********</h3>\n");
7309: fprintf(ficresphtmfr, "**********</h3>\n");
7310: fprintf(ficlog, "**********\n");
7311: }
1.284 brouard 7312: /*
7313: Printing means of quantitative variables if any
7314: */
7315: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 7316: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 7317: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 7318: if(weightopt==1){
7319: printf(" Weighted mean and standard deviation of");
7320: fprintf(ficlog," Weighted mean and standard deviation of");
7321: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
7322: }
1.311 brouard 7323: /* mu = \frac{w x}{\sum w}
7324: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
7325: */
7326: 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]));
7327: 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]));
7328: 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 7329: }
7330: /* for (z1=1; z1<= nqtveff; z1++) { */
7331: /* for(m=1;m<=lastpass;m++){ */
7332: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
7333: /* } */
7334: /* } */
1.283 brouard 7335:
1.251 brouard 7336: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.335 brouard 7337: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 7338: fprintf(ficresp, " Age");
1.335 brouard 7339: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) {
7340: 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]]);
7341: fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
7342: }
1.251 brouard 7343: for(i=1; i<=nlstate;i++) {
1.335 brouard 7344: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 7345: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
7346: }
1.335 brouard 7347: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 7348: fprintf(ficresphtm, "\n");
7349:
7350: /* Header of frequency table by age */
7351: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
7352: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 7353: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 7354: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 7355: if(s2!=0 && m!=0)
7356: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 7357: }
1.226 brouard 7358: }
1.251 brouard 7359: fprintf(ficresphtmfr, "\n");
7360:
7361: /* For each age */
7362: for(iage=iagemin; iage <= iagemax+3; iage++){
7363: fprintf(ficresphtm,"<tr>");
7364: if(iage==iagemax+1){
7365: fprintf(ficlog,"1");
7366: fprintf(ficresphtmfr,"<tr><th>0</th> ");
7367: }else if(iage==iagemax+2){
7368: fprintf(ficlog,"0");
7369: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
7370: }else if(iage==iagemax+3){
7371: fprintf(ficlog,"Total");
7372: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
7373: }else{
1.240 brouard 7374: if(first==1){
1.251 brouard 7375: first=0;
7376: printf("See log file for details...\n");
7377: }
7378: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
7379: fprintf(ficlog,"Age %d", iage);
7380: }
1.265 brouard 7381: for(s1=1; s1 <=nlstate ; s1++){
7382: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
7383: pp[s1] += freq[s1][m][iage];
1.251 brouard 7384: }
1.265 brouard 7385: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 7386: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 7387: pos += freq[s1][m][iage];
7388: if(pp[s1]>=1.e-10){
1.251 brouard 7389: if(first==1){
1.265 brouard 7390: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 7391: }
1.265 brouard 7392: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 7393: }else{
7394: if(first==1)
1.265 brouard 7395: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
7396: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 7397: }
7398: }
7399:
1.265 brouard 7400: for(s1=1; s1 <=nlstate ; s1++){
7401: /* posprop[s1]=0; */
7402: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
7403: pp[s1] += freq[s1][m][iage];
7404: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
7405:
7406: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
7407: pos += pp[s1]; /* pos is the total number of transitions until this age */
7408: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
7409: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
7410: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
7411: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
7412: }
7413:
7414: /* Writing ficresp */
1.335 brouard 7415: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 7416: if( iage <= iagemax){
7417: fprintf(ficresp," %d",iage);
7418: }
7419: }else if( nj==2){
7420: if( iage <= iagemax){
7421: fprintf(ficresp," %d",iage);
1.335 brouard 7422: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.265 brouard 7423: }
1.240 brouard 7424: }
1.265 brouard 7425: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 7426: if(pos>=1.e-5){
1.251 brouard 7427: if(first==1)
1.265 brouard 7428: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
7429: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 7430: }else{
7431: if(first==1)
1.265 brouard 7432: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
7433: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 7434: }
7435: if( iage <= iagemax){
7436: if(pos>=1.e-5){
1.335 brouard 7437: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 7438: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
7439: }else if( nj==2){
7440: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
7441: }
7442: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
7443: /*probs[iage][s1][j1]= pp[s1]/pos;*/
7444: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
7445: } else{
1.335 brouard 7446: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 7447: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 7448: }
1.240 brouard 7449: }
1.265 brouard 7450: pospropt[s1] +=posprop[s1];
7451: } /* end loop s1 */
1.251 brouard 7452: /* pospropt=0.; */
1.265 brouard 7453: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 7454: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 7455: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 7456: if(first==1){
1.265 brouard 7457: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 7458: }
1.265 brouard 7459: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
7460: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 7461: }
1.265 brouard 7462: if(s1!=0 && m!=0)
7463: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 7464: }
1.265 brouard 7465: } /* end loop s1 */
1.251 brouard 7466: posproptt=0.;
1.265 brouard 7467: for(s1=1; s1 <=nlstate; s1++){
7468: posproptt += pospropt[s1];
1.251 brouard 7469: }
7470: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 7471: fprintf(ficresphtm,"</tr>\n");
1.335 brouard 7472: if((cptcoveff==0 && nj==1)|| nj==2 ) {
1.265 brouard 7473: if(iage <= iagemax)
7474: fprintf(ficresp,"\n");
1.240 brouard 7475: }
1.251 brouard 7476: if(first==1)
7477: printf("Others in log...\n");
7478: fprintf(ficlog,"\n");
7479: } /* end loop age iage */
1.265 brouard 7480:
1.251 brouard 7481: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 7482: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 7483: if(posproptt < 1.e-5){
1.265 brouard 7484: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 7485: }else{
1.265 brouard 7486: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 7487: }
1.226 brouard 7488: }
1.251 brouard 7489: fprintf(ficresphtm,"</tr>\n");
7490: fprintf(ficresphtm,"</table>\n");
7491: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 7492: if(posproptt < 1.e-5){
1.251 brouard 7493: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
7494: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 7495: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
7496: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 7497: invalidvarcomb[j1]=1;
1.226 brouard 7498: }else{
1.338 brouard 7499: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced (or no resultline).</p>",j1);
1.251 brouard 7500: invalidvarcomb[j1]=0;
1.226 brouard 7501: }
1.251 brouard 7502: fprintf(ficresphtmfr,"</table>\n");
7503: fprintf(ficlog,"\n");
7504: if(j!=0){
7505: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 7506: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 7507: for(k=1; k <=(nlstate+ndeath); k++){
7508: if (k != i) {
1.265 brouard 7509: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 7510: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 7511: if(j1==1){ /* All dummy covariates to zero */
7512: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
7513: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 7514: printf("%d%d ",i,k);
7515: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 7516: 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]));
7517: 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]));
7518: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 7519: }
1.253 brouard 7520: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
7521: for(iage=iagemin; iage <= iagemax+3; iage++){
7522: x[iage]= (double)iage;
7523: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 7524: /* 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 7525: }
1.268 brouard 7526: /* Some are not finite, but linreg will ignore these ages */
7527: no=0;
1.253 brouard 7528: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 7529: pstart[s1]=b;
7530: pstart[s1-1]=a;
1.252 brouard 7531: }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 */
7532: 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]);
7533: 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 7534: 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 7535: printf("%d%d ",i,k);
7536: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 7537: 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 7538: }else{ /* Other cases, like quantitative fixed or varying covariates */
7539: ;
7540: }
7541: /* printf("%12.7f )", param[i][jj][k]); */
7542: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 7543: s1++;
1.251 brouard 7544: } /* end jj */
7545: } /* end k!= i */
7546: } /* end k */
1.265 brouard 7547: } /* end i, s1 */
1.251 brouard 7548: } /* end j !=0 */
7549: } /* end selected combination of covariate j1 */
7550: if(j==0){ /* We can estimate starting values from the occurences in each case */
7551: printf("#Freqsummary: Starting values for the constants:\n");
7552: fprintf(ficlog,"\n");
1.265 brouard 7553: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 7554: for(k=1; k <=(nlstate+ndeath); k++){
7555: if (k != i) {
7556: printf("%d%d ",i,k);
7557: fprintf(ficlog,"%d%d ",i,k);
7558: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 7559: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 7560: if(jj==1){ /* Age has to be done */
1.265 brouard 7561: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
7562: 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]));
7563: 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 7564: }
7565: /* printf("%12.7f )", param[i][jj][k]); */
7566: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 7567: s1++;
1.250 brouard 7568: }
1.251 brouard 7569: printf("\n");
7570: fprintf(ficlog,"\n");
1.250 brouard 7571: }
7572: }
1.284 brouard 7573: } /* end of state i */
1.251 brouard 7574: printf("#Freqsummary\n");
7575: fprintf(ficlog,"\n");
1.265 brouard 7576: for(s1=-1; s1 <=nlstate+ndeath; s1++){
7577: for(s2=-1; s2 <=nlstate+ndeath; s2++){
7578: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
7579: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
7580: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
7581: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
7582: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
7583: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 7584: /* } */
7585: }
1.265 brouard 7586: } /* end loop s1 */
1.251 brouard 7587:
7588: printf("\n");
7589: fprintf(ficlog,"\n");
7590: } /* end j=0 */
1.249 brouard 7591: } /* end j */
1.252 brouard 7592:
1.253 brouard 7593: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 7594: for(i=1, jk=1; i <=nlstate; i++){
7595: for(j=1; j <=nlstate+ndeath; j++){
7596: if(j!=i){
7597: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7598: printf("%1d%1d",i,j);
7599: fprintf(ficparo,"%1d%1d",i,j);
7600: for(k=1; k<=ncovmodel;k++){
7601: /* printf(" %lf",param[i][j][k]); */
7602: /* fprintf(ficparo," %lf",param[i][j][k]); */
7603: p[jk]=pstart[jk];
7604: printf(" %f ",pstart[jk]);
7605: fprintf(ficparo," %f ",pstart[jk]);
7606: jk++;
7607: }
7608: printf("\n");
7609: fprintf(ficparo,"\n");
7610: }
7611: }
7612: }
7613: } /* end mle=-2 */
1.226 brouard 7614: dateintmean=dateintsum/k2cpt;
1.296 brouard 7615: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 7616:
1.226 brouard 7617: fclose(ficresp);
7618: fclose(ficresphtm);
7619: fclose(ficresphtmfr);
1.283 brouard 7620: free_vector(idq,1,nqfveff);
1.226 brouard 7621: free_vector(meanq,1,nqfveff);
1.284 brouard 7622: free_vector(stdq,1,nqfveff);
1.226 brouard 7623: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 7624: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
7625: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 7626: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 7627: free_vector(pospropt,1,nlstate);
7628: free_vector(posprop,1,nlstate);
1.251 brouard 7629: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 7630: free_vector(pp,1,nlstate);
7631: /* End of freqsummary */
7632: }
1.126 brouard 7633:
1.268 brouard 7634: /* Simple linear regression */
7635: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
7636:
7637: /* y=a+bx regression */
7638: double sumx = 0.0; /* sum of x */
7639: double sumx2 = 0.0; /* sum of x**2 */
7640: double sumxy = 0.0; /* sum of x * y */
7641: double sumy = 0.0; /* sum of y */
7642: double sumy2 = 0.0; /* sum of y**2 */
7643: double sume2 = 0.0; /* sum of square or residuals */
7644: double yhat;
7645:
7646: double denom=0;
7647: int i;
7648: int ne=*no;
7649:
7650: for ( i=ifi, ne=0;i<=ila;i++) {
7651: if(!isfinite(x[i]) || !isfinite(y[i])){
7652: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
7653: continue;
7654: }
7655: ne=ne+1;
7656: sumx += x[i];
7657: sumx2 += x[i]*x[i];
7658: sumxy += x[i] * y[i];
7659: sumy += y[i];
7660: sumy2 += y[i]*y[i];
7661: denom = (ne * sumx2 - sumx*sumx);
7662: /* 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); */
7663: }
7664:
7665: denom = (ne * sumx2 - sumx*sumx);
7666: if (denom == 0) {
7667: // vertical, slope m is infinity
7668: *b = INFINITY;
7669: *a = 0;
7670: if (r) *r = 0;
7671: return 1;
7672: }
7673:
7674: *b = (ne * sumxy - sumx * sumy) / denom;
7675: *a = (sumy * sumx2 - sumx * sumxy) / denom;
7676: if (r!=NULL) {
7677: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
7678: sqrt((sumx2 - sumx*sumx/ne) *
7679: (sumy2 - sumy*sumy/ne));
7680: }
7681: *no=ne;
7682: for ( i=ifi, ne=0;i<=ila;i++) {
7683: if(!isfinite(x[i]) || !isfinite(y[i])){
7684: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
7685: continue;
7686: }
7687: ne=ne+1;
7688: yhat = y[i] - *a -*b* x[i];
7689: sume2 += yhat * yhat ;
7690:
7691: denom = (ne * sumx2 - sumx*sumx);
7692: /* 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); */
7693: }
7694: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
7695: *sa= *sb * sqrt(sumx2/ne);
7696:
7697: return 0;
7698: }
7699:
1.126 brouard 7700: /************ Prevalence ********************/
1.227 brouard 7701: 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)
7702: {
7703: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7704: in each health status at the date of interview (if between dateprev1 and dateprev2).
7705: We still use firstpass and lastpass as another selection.
7706: */
1.126 brouard 7707:
1.227 brouard 7708: int i, m, jk, j1, bool, z1,j, iv;
7709: int mi; /* Effective wave */
7710: int iage;
1.359 brouard 7711: double agebegin; /*, ageend;*/
1.227 brouard 7712:
7713: double **prop;
7714: double posprop;
7715: double y2; /* in fractional years */
7716: int iagemin, iagemax;
7717: int first; /** to stop verbosity which is redirected to log file */
7718:
7719: iagemin= (int) agemin;
7720: iagemax= (int) agemax;
7721: /*pp=vector(1,nlstate);*/
1.251 brouard 7722: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 7723: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
7724: j1=0;
1.222 brouard 7725:
1.227 brouard 7726: /*j=cptcoveff;*/
7727: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 7728:
1.288 brouard 7729: first=0;
1.335 brouard 7730: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of simple dummy covariates */
1.227 brouard 7731: for (i=1; i<=nlstate; i++)
1.251 brouard 7732: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 7733: prop[i][iage]=0.0;
7734: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
7735: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
7736: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
7737:
7738: for (i=1; i<=imx; i++) { /* Each individual */
7739: bool=1;
7740: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
7741: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
7742: m=mw[mi][i];
7743: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
7744: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
7745: for (z1=1; z1<=cptcoveff; z1++){
7746: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 7747: iv= Tvar[Tmodelind[z1]];/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.332 brouard 7748: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality */
1.227 brouard 7749: bool=0;
7750: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.332 brouard 7751: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.227 brouard 7752: bool=0;
7753: }
7754: }
7755: if(bool==1){ /* Otherwise we skip that wave/person */
7756: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
7757: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
7758: if(m >=firstpass && m <=lastpass){
7759: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
7760: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
7761: if(agev[m][i]==0) agev[m][i]=iagemax+1;
7762: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 7763: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 7764: 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);
7765: exit(1);
7766: }
7767: if (s[m][i]>0 && s[m][i]<=nlstate) {
7768: /*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]]);*/
7769: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
7770: prop[s[m][i]][iagemax+3] += weight[i];
7771: } /* end valid statuses */
7772: } /* end selection of dates */
7773: } /* end selection of waves */
7774: } /* end bool */
7775: } /* end wave */
7776: } /* end individual */
7777: for(i=iagemin; i <= iagemax+3; i++){
7778: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
7779: posprop += prop[jk][i];
7780: }
7781:
7782: for(jk=1; jk <=nlstate ; jk++){
7783: if( i <= iagemax){
7784: if(posprop>=1.e-5){
7785: probs[i][jk][j1]= prop[jk][i]/posprop;
7786: } else{
1.288 brouard 7787: if(!first){
7788: first=1;
1.266 brouard 7789: 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]);
7790: }else{
1.288 brouard 7791: 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 7792: }
7793: }
7794: }
7795: }/* end jk */
7796: }/* end i */
1.222 brouard 7797: /*} *//* end i1 */
1.227 brouard 7798: } /* end j1 */
1.222 brouard 7799:
1.227 brouard 7800: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
7801: /*free_vector(pp,1,nlstate);*/
1.251 brouard 7802: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 7803: } /* End of prevalence */
1.126 brouard 7804:
7805: /************* Waves Concatenation ***************/
7806:
7807: 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)
7808: {
1.298 brouard 7809: /* 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 7810: Death is a valid wave (if date is known).
7811: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
7812: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 7813: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 7814: */
1.126 brouard 7815:
1.224 brouard 7816: int i=0, mi=0, m=0, mli=0;
1.126 brouard 7817: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
7818: double sum=0., jmean=0.;*/
1.224 brouard 7819: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 7820: int j, k=0,jk, ju, jl;
7821: double sum=0.;
7822: first=0;
1.214 brouard 7823: firstwo=0;
1.217 brouard 7824: firsthree=0;
1.218 brouard 7825: firstfour=0;
1.164 brouard 7826: jmin=100000;
1.126 brouard 7827: jmax=-1;
7828: jmean=0.;
1.224 brouard 7829:
7830: /* Treating live states */
1.214 brouard 7831: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 7832: mi=0; /* First valid wave */
1.227 brouard 7833: mli=0; /* Last valid wave */
1.309 brouard 7834: m=firstpass; /* Loop on waves */
7835: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 7836: 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 */
7837: mli=m-1;/* mw[++mi][i]=m-1; */
7838: }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 7839: 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 7840: mli=m;
1.224 brouard 7841: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
7842: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 7843: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 7844: }
1.309 brouard 7845: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 7846: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 7847: break;
1.224 brouard 7848: #else
1.317 brouard 7849: 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 7850: if(firsthree == 0){
1.302 brouard 7851: 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 7852: firsthree=1;
1.317 brouard 7853: }else if(firsthree >=1 && firsthree < 10){
7854: 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);
7855: firsthree++;
7856: }else if(firsthree == 10){
7857: printf("Information, too many Information flags: no more reported to log either\n");
7858: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
7859: firsthree++;
7860: }else{
7861: firsthree++;
1.227 brouard 7862: }
1.309 brouard 7863: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 7864: mli=m;
7865: }
7866: if(s[m][i]==-2){ /* Vital status is really unknown */
7867: nbwarn++;
1.309 brouard 7868: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 7869: 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);
7870: 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);
7871: }
7872: break;
7873: }
7874: break;
1.224 brouard 7875: #endif
1.227 brouard 7876: }/* End m >= lastpass */
1.126 brouard 7877: }/* end while */
1.224 brouard 7878:
1.227 brouard 7879: /* 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 7880: /* After last pass */
1.224 brouard 7881: /* Treating death states */
1.214 brouard 7882: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 7883: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
7884: /* } */
1.126 brouard 7885: mi++; /* Death is another wave */
7886: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 7887: /* Only death is a correct wave */
1.126 brouard 7888: mw[mi][i]=m;
1.257 brouard 7889: } /* else not in a death state */
1.224 brouard 7890: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 7891: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 7892: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 7893: 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 7894: nbwarn++;
7895: if(firstfiv==0){
1.309 brouard 7896: 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 7897: firstfiv=1;
7898: }else{
1.309 brouard 7899: 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 7900: }
1.309 brouard 7901: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
7902: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 7903: nberr++;
7904: if(firstwo==0){
1.309 brouard 7905: 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 7906: firstwo=1;
7907: }
1.309 brouard 7908: 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 7909: }
1.257 brouard 7910: }else{ /* if date of interview is unknown */
1.227 brouard 7911: /* death is known but not confirmed by death status at any wave */
7912: if(firstfour==0){
1.309 brouard 7913: 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 7914: firstfour=1;
7915: }
1.309 brouard 7916: 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 7917: }
1.224 brouard 7918: } /* end if date of death is known */
7919: #endif
1.309 brouard 7920: wav[i]=mi; /* mi should be the last effective wave (or mli), */
7921: /* wav[i]=mw[mi][i]; */
1.126 brouard 7922: if(mi==0){
7923: nbwarn++;
7924: if(first==0){
1.227 brouard 7925: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
7926: first=1;
1.126 brouard 7927: }
7928: if(first==1){
1.227 brouard 7929: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 7930: }
7931: } /* end mi==0 */
7932: } /* End individuals */
1.214 brouard 7933: /* wav and mw are no more changed */
1.223 brouard 7934:
1.317 brouard 7935: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
7936: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
7937:
7938:
1.126 brouard 7939: for(i=1; i<=imx; i++){
7940: for(mi=1; mi<wav[i];mi++){
7941: if (stepm <=0)
1.227 brouard 7942: dh[mi][i]=1;
1.126 brouard 7943: else{
1.260 brouard 7944: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 7945: if (agedc[i] < 2*AGESUP) {
7946: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
7947: if(j==0) j=1; /* Survives at least one month after exam */
7948: else if(j<0){
7949: nberr++;
1.359 brouard 7950: 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 7951: j=1; /* Temporary Dangerous patch */
7952: 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 7953: 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 7954: 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);
7955: }
7956: k=k+1;
7957: if (j >= jmax){
7958: jmax=j;
7959: ijmax=i;
7960: }
7961: if (j <= jmin){
7962: jmin=j;
7963: ijmin=i;
7964: }
7965: sum=sum+j;
7966: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
7967: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
7968: }
7969: }
7970: else{
7971: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 7972: /* 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 7973:
1.227 brouard 7974: k=k+1;
7975: if (j >= jmax) {
7976: jmax=j;
7977: ijmax=i;
7978: }
7979: else if (j <= jmin){
7980: jmin=j;
7981: ijmin=i;
7982: }
7983: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
7984: /*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]);*/
7985: if(j<0){
7986: nberr++;
1.359 brouard 7987: 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]);
7988: 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 7989: }
7990: sum=sum+j;
7991: }
7992: jk= j/stepm;
7993: jl= j -jk*stepm;
7994: ju= j -(jk+1)*stepm;
7995: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
7996: if(jl==0){
7997: dh[mi][i]=jk;
7998: bh[mi][i]=0;
7999: }else{ /* We want a negative bias in order to only have interpolation ie
8000: * to avoid the price of an extra matrix product in likelihood */
8001: dh[mi][i]=jk+1;
8002: bh[mi][i]=ju;
8003: }
8004: }else{
8005: if(jl <= -ju){
8006: dh[mi][i]=jk;
8007: bh[mi][i]=jl; /* bias is positive if real duration
8008: * is higher than the multiple of stepm and negative otherwise.
8009: */
8010: }
8011: else{
8012: dh[mi][i]=jk+1;
8013: bh[mi][i]=ju;
8014: }
8015: if(dh[mi][i]==0){
8016: dh[mi][i]=1; /* At least one step */
8017: bh[mi][i]=ju; /* At least one step */
8018: /* 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);*/
8019: }
8020: } /* end if mle */
1.126 brouard 8021: }
8022: } /* end wave */
8023: }
8024: jmean=sum/k;
8025: 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 8026: 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 8027: }
1.126 brouard 8028:
8029: /*********** Tricode ****************************/
1.220 brouard 8030: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 8031: {
8032: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
8033: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
8034: * Boring subroutine which should only output nbcode[Tvar[j]][k]
8035: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
8036: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
8037: */
1.130 brouard 8038:
1.242 brouard 8039: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
8040: int modmaxcovj=0; /* Modality max of covariates j */
8041: int cptcode=0; /* Modality max of covariates j */
8042: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 8043:
8044:
1.242 brouard 8045: /* cptcoveff=0; */
8046: /* *cptcov=0; */
1.126 brouard 8047:
1.242 brouard 8048: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 8049: for (k=1; k <= maxncov; k++)
8050: for(j=1; j<=2; j++)
8051: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 8052:
1.242 brouard 8053: /* Loop on covariates without age and products and no quantitative variable */
1.335 brouard 8054: 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 8055: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
1.343 brouard 8056: /* printf("Testing k=%d, cptcovt=%d\n",k, cptcovt); */
1.349 brouard 8057: 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 8058: switch(Fixed[k]) {
8059: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 8060: modmaxcovj=0;
8061: modmincovj=0;
1.242 brouard 8062: 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 8063: /* 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 8064: ij=(int)(covar[Tvar[k]][i]);
8065: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
8066: * If product of Vn*Vm, still boolean *:
8067: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
8068: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
8069: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
8070: modality of the nth covariate of individual i. */
8071: if (ij > modmaxcovj)
8072: modmaxcovj=ij;
8073: else if (ij < modmincovj)
8074: modmincovj=ij;
1.287 brouard 8075: if (ij <0 || ij >1 ){
1.311 brouard 8076: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
8077: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
8078: fflush(ficlog);
8079: exit(1);
1.287 brouard 8080: }
8081: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 8082: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
8083: exit(1);
8084: }else
8085: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
8086: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
8087: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
8088: /* getting the maximum value of the modality of the covariate
8089: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
8090: female ies 1, then modmaxcovj=1.
8091: */
8092: } /* end for loop on individuals i */
8093: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
8094: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
8095: cptcode=modmaxcovj;
8096: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
8097: /*for (i=0; i<=cptcode; i++) {*/
8098: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
8099: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
8100: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
8101: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
8102: if( j != -1){
8103: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
8104: covariate for which somebody answered excluding
8105: undefined. Usually 2: 0 and 1. */
8106: }
8107: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
8108: covariate for which somebody answered including
8109: undefined. Usually 3: -1, 0 and 1. */
8110: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
8111: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
8112: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 8113:
1.242 brouard 8114: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
8115: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
8116: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
8117: /* modmincovj=3; modmaxcovj = 7; */
8118: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
8119: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
8120: /* defining two dummy variables: variables V1_1 and V1_2.*/
8121: /* nbcode[Tvar[j]][ij]=k; */
8122: /* nbcode[Tvar[j]][1]=0; */
8123: /* nbcode[Tvar[j]][2]=1; */
8124: /* nbcode[Tvar[j]][3]=2; */
8125: /* To be continued (not working yet). */
8126: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 8127:
8128: /* 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*/
8129: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
8130: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
8131: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
8132: /*, could be restored in the future */
8133: 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 8134: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
8135: break;
8136: }
8137: ij++;
1.287 brouard 8138: 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 8139: cptcode = ij; /* New max modality for covar j */
8140: } /* end of loop on modality i=-1 to 1 or more */
8141: break;
8142: case 1: /* Testing on varying covariate, could be simple and
8143: * should look at waves or product of fixed *
8144: * varying. No time to test -1, assuming 0 and 1 only */
8145: ij=0;
8146: for(i=0; i<=1;i++){
8147: nbcode[Tvar[k]][++ij]=i;
8148: }
8149: break;
8150: default:
8151: break;
8152: } /* end switch */
8153: } /* end dummy test */
1.349 brouard 8154: if(Dummy[k]==1 && Typevar[k] !=1 && Typevar[k] !=3 && Fixed ==0){ /* Fixed Quantitative covariate and not age product */
1.311 brouard 8155: 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 8156: if(Tvar[k]<=0 || Tvar[k]>=NCOVMAX){
8157: printf("Error k=%d \n",k);
8158: exit(1);
8159: }
1.311 brouard 8160: if(isnan(covar[Tvar[k]][i])){
8161: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
8162: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
8163: fflush(ficlog);
8164: exit(1);
8165: }
8166: }
1.335 brouard 8167: } /* end Quanti */
1.287 brouard 8168: } /* 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 8169:
8170: for (k=-1; k< maxncov; k++) Ndum[k]=0;
8171: /* Look at fixed dummy (single or product) covariates to check empty modalities */
8172: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
8173: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
8174: 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 */
8175: 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 */
8176: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
8177: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
8178:
8179: ij=0;
8180: /* 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 8181: 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 */
8182: /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
1.242 brouard 8183: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
8184: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
1.335 brouard 8185: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy simple and non empty in the model */
8186: /* Typevar[k] =0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
8187: /* 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 8188: /* If product not in single variable we don't print results */
8189: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
1.335 brouard 8190: ++ij;/* V5 + V4 + V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V1, *//* V5 quanti, V2 quanti, V4, V3, V1 dummies */
8191: /* k= 1 2 3 4 5 6 7 8 9 */
8192: /* Tvar[k]= 5 4 3 6 5 2 7 1 1 */
8193: /* ij 1 2 3 */
8194: /* Tvaraff[ij]= 4 3 1 */
8195: /* Tmodelind[ij]=2 3 9 */
8196: /* TmodelInvind[ij]=2 1 1 */
1.242 brouard 8197: 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*/
8198: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
8199: 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 */
8200: if(Fixed[k]!=0)
8201: anyvaryingduminmodel=1;
8202: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
8203: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
8204: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
8205: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
8206: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
8207: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
8208: }
8209: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
8210: /* ij--; */
8211: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.335 brouard 8212: *cptcov=ij; /* cptcov= Number of total real effective simple dummies (fixed or time arying) effective (used as cptcoveff in other functions)
1.242 brouard 8213: * because they can be excluded from the model and real
8214: * if in the model but excluded because missing values, but how to get k from ij?*/
8215: for(j=ij+1; j<= cptcovt; j++){
8216: Tvaraff[j]=0;
8217: Tmodelind[j]=0;
8218: }
8219: for(j=ntveff+1; j<= cptcovt; j++){
8220: TmodelInvind[j]=0;
8221: }
8222: /* To be sorted */
8223: ;
8224: }
1.126 brouard 8225:
1.145 brouard 8226:
1.126 brouard 8227: /*********** Health Expectancies ****************/
8228:
1.235 brouard 8229: 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 8230:
8231: {
8232: /* Health expectancies, no variances */
1.329 brouard 8233: /* cij is the combination in the list of combination of dummy covariates */
8234: /* strstart is a string of time at start of computing */
1.164 brouard 8235: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 8236: int nhstepma, nstepma; /* Decreasing with age */
8237: double age, agelim, hf;
8238: double ***p3mat;
8239: double eip;
8240:
1.238 brouard 8241: /* pstamp(ficreseij); */
1.126 brouard 8242: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
8243: fprintf(ficreseij,"# Age");
8244: for(i=1; i<=nlstate;i++){
8245: for(j=1; j<=nlstate;j++){
8246: fprintf(ficreseij," e%1d%1d ",i,j);
8247: }
8248: fprintf(ficreseij," e%1d. ",i);
8249: }
8250: fprintf(ficreseij,"\n");
8251:
8252:
8253: if(estepm < stepm){
8254: printf ("Problem %d lower than %d\n",estepm, stepm);
8255: }
8256: else hstepm=estepm;
8257: /* We compute the life expectancy from trapezoids spaced every estepm months
8258: * This is mainly to measure the difference between two models: for example
8259: * if stepm=24 months pijx are given only every 2 years and by summing them
8260: * we are calculating an estimate of the Life Expectancy assuming a linear
8261: * progression in between and thus overestimating or underestimating according
8262: * to the curvature of the survival function. If, for the same date, we
8263: * estimate the model with stepm=1 month, we can keep estepm to 24 months
8264: * to compare the new estimate of Life expectancy with the same linear
8265: * hypothesis. A more precise result, taking into account a more precise
8266: * curvature will be obtained if estepm is as small as stepm. */
8267:
8268: /* For example we decided to compute the life expectancy with the smallest unit */
8269: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
8270: nhstepm is the number of hstepm from age to agelim
8271: nstepm is the number of stepm from age to agelin.
1.270 brouard 8272: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 8273: and note for a fixed period like estepm months */
8274: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
8275: survival function given by stepm (the optimization length). Unfortunately it
8276: means that if the survival funtion is printed only each two years of age and if
8277: you sum them up and add 1 year (area under the trapezoids) you won't get the same
8278: results. So we changed our mind and took the option of the best precision.
8279: */
8280: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
8281:
8282: agelim=AGESUP;
8283: /* If stepm=6 months */
8284: /* Computed by stepm unit matrices, product of hstepm matrices, stored
8285: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
8286:
8287: /* nhstepm age range expressed in number of stepm */
8288: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
8289: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
8290: /* if (stepm >= YEARM) hstepm=1;*/
8291: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
8292: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8293:
8294: for (age=bage; age<=fage; age ++){
8295: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
8296: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
8297: /* if (stepm >= YEARM) hstepm=1;*/
8298: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
8299:
8300: /* If stepm=6 months */
8301: /* Computed by stepm unit matrices, product of hstepma matrices, stored
8302: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 brouard 8303: /* 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 8304: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 8305:
8306: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
8307:
8308: printf("%d|",(int)age);fflush(stdout);
8309: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
8310:
8311: /* Computing expectancies */
8312: for(i=1; i<=nlstate;i++)
8313: for(j=1; j<=nlstate;j++)
8314: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
8315: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
8316:
8317: /* 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]);*/
8318:
8319: }
8320:
8321: fprintf(ficreseij,"%3.0f",age );
8322: for(i=1; i<=nlstate;i++){
8323: eip=0;
8324: for(j=1; j<=nlstate;j++){
8325: eip +=eij[i][j][(int)age];
8326: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
8327: }
8328: fprintf(ficreseij,"%9.4f", eip );
8329: }
8330: fprintf(ficreseij,"\n");
8331:
8332: }
8333: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8334: printf("\n");
8335: fprintf(ficlog,"\n");
8336:
8337: }
8338:
1.235 brouard 8339: 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 8340:
8341: {
8342: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 8343: to initial status i, ei. .
1.126 brouard 8344: */
1.336 brouard 8345: /* Very time consuming function, but already optimized with precov */
1.126 brouard 8346: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
8347: int nhstepma, nstepma; /* Decreasing with age */
8348: double age, agelim, hf;
8349: double ***p3matp, ***p3matm, ***varhe;
8350: double **dnewm,**doldm;
8351: double *xp, *xm;
8352: double **gp, **gm;
8353: double ***gradg, ***trgradg;
8354: int theta;
8355:
8356: double eip, vip;
8357:
8358: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
8359: xp=vector(1,npar);
8360: xm=vector(1,npar);
8361: dnewm=matrix(1,nlstate*nlstate,1,npar);
8362: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
8363:
8364: pstamp(ficresstdeij);
8365: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
8366: fprintf(ficresstdeij,"# Age");
8367: for(i=1; i<=nlstate;i++){
8368: for(j=1; j<=nlstate;j++)
8369: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
8370: fprintf(ficresstdeij," e%1d. ",i);
8371: }
8372: fprintf(ficresstdeij,"\n");
8373:
8374: pstamp(ficrescveij);
8375: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
8376: fprintf(ficrescveij,"# Age");
8377: for(i=1; i<=nlstate;i++)
8378: for(j=1; j<=nlstate;j++){
8379: cptj= (j-1)*nlstate+i;
8380: for(i2=1; i2<=nlstate;i2++)
8381: for(j2=1; j2<=nlstate;j2++){
8382: cptj2= (j2-1)*nlstate+i2;
8383: if(cptj2 <= cptj)
8384: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
8385: }
8386: }
8387: fprintf(ficrescveij,"\n");
8388:
8389: if(estepm < stepm){
8390: printf ("Problem %d lower than %d\n",estepm, stepm);
8391: }
8392: else hstepm=estepm;
8393: /* We compute the life expectancy from trapezoids spaced every estepm months
8394: * This is mainly to measure the difference between two models: for example
8395: * if stepm=24 months pijx are given only every 2 years and by summing them
8396: * we are calculating an estimate of the Life Expectancy assuming a linear
8397: * progression in between and thus overestimating or underestimating according
8398: * to the curvature of the survival function. If, for the same date, we
8399: * estimate the model with stepm=1 month, we can keep estepm to 24 months
8400: * to compare the new estimate of Life expectancy with the same linear
8401: * hypothesis. A more precise result, taking into account a more precise
8402: * curvature will be obtained if estepm is as small as stepm. */
8403:
8404: /* For example we decided to compute the life expectancy with the smallest unit */
8405: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
8406: nhstepm is the number of hstepm from age to agelim
8407: nstepm is the number of stepm from age to agelin.
8408: Look at hpijx to understand the reason of that which relies in memory size
8409: and note for a fixed period like estepm months */
8410: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
8411: survival function given by stepm (the optimization length). Unfortunately it
8412: means that if the survival funtion is printed only each two years of age and if
8413: you sum them up and add 1 year (area under the trapezoids) you won't get the same
8414: results. So we changed our mind and took the option of the best precision.
8415: */
8416: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
8417:
8418: /* If stepm=6 months */
8419: /* nhstepm age range expressed in number of stepm */
8420: agelim=AGESUP;
8421: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
8422: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
8423: /* if (stepm >= YEARM) hstepm=1;*/
8424: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
8425:
8426: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8427: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8428: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
8429: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
8430: gp=matrix(0,nhstepm,1,nlstate*nlstate);
8431: gm=matrix(0,nhstepm,1,nlstate*nlstate);
8432:
8433: for (age=bage; age<=fage; age ++){
8434: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
8435: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
8436: /* if (stepm >= YEARM) hstepm=1;*/
8437: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 8438:
1.126 brouard 8439: /* If stepm=6 months */
8440: /* Computed by stepm unit matrices, product of hstepma matrices, stored
8441: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
8442:
8443: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 8444:
1.126 brouard 8445: /* Computing Variances of health expectancies */
8446: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
8447: decrease memory allocation */
8448: for(theta=1; theta <=npar; theta++){
8449: for(i=1; i<=npar; i++){
1.222 brouard 8450: xp[i] = x[i] + (i==theta ?delti[theta]:0);
8451: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 8452: }
1.235 brouard 8453: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
8454: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 8455:
1.126 brouard 8456: for(j=1; j<= nlstate; j++){
1.222 brouard 8457: for(i=1; i<=nlstate; i++){
8458: for(h=0; h<=nhstepm-1; h++){
8459: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
8460: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
8461: }
8462: }
1.126 brouard 8463: }
1.218 brouard 8464:
1.126 brouard 8465: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 8466: for(h=0; h<=nhstepm-1; h++){
8467: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
8468: }
1.126 brouard 8469: }/* End theta */
8470:
8471:
8472: for(h=0; h<=nhstepm-1; h++)
8473: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 8474: for(theta=1; theta <=npar; theta++)
8475: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 8476:
1.218 brouard 8477:
1.222 brouard 8478: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 8479: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 8480: varhe[ij][ji][(int)age] =0.;
1.218 brouard 8481:
1.222 brouard 8482: printf("%d|",(int)age);fflush(stdout);
8483: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
8484: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 8485: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 8486: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
8487: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
8488: for(ij=1;ij<=nlstate*nlstate;ij++)
8489: for(ji=1;ji<=nlstate*nlstate;ji++)
8490: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 8491: }
8492: }
1.320 brouard 8493: /* if((int)age ==50){ */
8494: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
8495: /* } */
1.126 brouard 8496: /* Computing expectancies */
1.235 brouard 8497: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 8498: for(i=1; i<=nlstate;i++)
8499: for(j=1; j<=nlstate;j++)
1.222 brouard 8500: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
8501: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 8502:
1.222 brouard 8503: /* 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 8504:
1.222 brouard 8505: }
1.269 brouard 8506:
8507: /* Standard deviation of expectancies ij */
1.126 brouard 8508: fprintf(ficresstdeij,"%3.0f",age );
8509: for(i=1; i<=nlstate;i++){
8510: eip=0.;
8511: vip=0.;
8512: for(j=1; j<=nlstate;j++){
1.222 brouard 8513: eip += eij[i][j][(int)age];
8514: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
8515: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
8516: 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 8517: }
8518: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
8519: }
8520: fprintf(ficresstdeij,"\n");
1.218 brouard 8521:
1.269 brouard 8522: /* Variance of expectancies ij */
1.126 brouard 8523: fprintf(ficrescveij,"%3.0f",age );
8524: for(i=1; i<=nlstate;i++)
8525: for(j=1; j<=nlstate;j++){
1.222 brouard 8526: cptj= (j-1)*nlstate+i;
8527: for(i2=1; i2<=nlstate;i2++)
8528: for(j2=1; j2<=nlstate;j2++){
8529: cptj2= (j2-1)*nlstate+i2;
8530: if(cptj2 <= cptj)
8531: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
8532: }
1.126 brouard 8533: }
8534: fprintf(ficrescveij,"\n");
1.218 brouard 8535:
1.126 brouard 8536: }
8537: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
8538: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
8539: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
8540: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
8541: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8542: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8543: printf("\n");
8544: fprintf(ficlog,"\n");
1.218 brouard 8545:
1.126 brouard 8546: free_vector(xm,1,npar);
8547: free_vector(xp,1,npar);
8548: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
8549: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
8550: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
8551: }
1.218 brouard 8552:
1.126 brouard 8553: /************ Variance ******************/
1.235 brouard 8554: 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 8555: {
1.279 brouard 8556: /** Variance of health expectancies
8557: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
8558: * double **newm;
8559: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
8560: */
1.218 brouard 8561:
8562: /* int movingaverage(); */
8563: double **dnewm,**doldm;
8564: double **dnewmp,**doldmp;
8565: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 8566: int first=0;
1.218 brouard 8567: int k;
8568: double *xp;
1.279 brouard 8569: double **gp, **gm; /**< for var eij */
8570: double ***gradg, ***trgradg; /**< for var eij */
8571: double **gradgp, **trgradgp; /**< for var p point j */
8572: double *gpp, *gmp; /**< for var p point j */
8573: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 8574: double ***p3mat;
8575: double age,agelim, hf;
8576: /* double ***mobaverage; */
8577: int theta;
8578: char digit[4];
8579: char digitp[25];
8580:
8581: char fileresprobmorprev[FILENAMELENGTH];
8582:
8583: if(popbased==1){
8584: if(mobilav!=0)
8585: strcpy(digitp,"-POPULBASED-MOBILAV_");
8586: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
8587: }
8588: else
8589: strcpy(digitp,"-STABLBASED_");
1.126 brouard 8590:
1.218 brouard 8591: /* if (mobilav!=0) { */
8592: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8593: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
8594: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
8595: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
8596: /* } */
8597: /* } */
8598:
8599: strcpy(fileresprobmorprev,"PRMORPREV-");
8600: sprintf(digit,"%-d",ij);
8601: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
8602: strcat(fileresprobmorprev,digit); /* Tvar to be done */
8603: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
8604: strcat(fileresprobmorprev,fileresu);
8605: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
8606: printf("Problem with resultfile: %s\n", fileresprobmorprev);
8607: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
8608: }
8609: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
8610: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
8611: pstamp(ficresprobmorprev);
8612: 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 8613: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
1.337 brouard 8614:
8615: /* We use TinvDoQresult[nres][resultmodel[nres][j] we sort according to the equation model and the resultline: it is a choice */
8616: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ /\* To be done*\/ */
8617: /* fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
8618: /* } */
8619: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */ /* To be done*/
1.344 brouard 8620: /* fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); */
1.337 brouard 8621: fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 8622: }
1.337 brouard 8623: /* for(j=1;j<=cptcoveff;j++) */
8624: /* fprintf(ficresprobmorprev," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,TnsdVar[Tvaraff[j]])]); */
1.238 brouard 8625: fprintf(ficresprobmorprev,"\n");
8626:
1.218 brouard 8627: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
8628: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
8629: fprintf(ficresprobmorprev," p.%-d SE",j);
8630: for(i=1; i<=nlstate;i++)
8631: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
8632: }
8633: fprintf(ficresprobmorprev,"\n");
8634:
8635: fprintf(ficgp,"\n# Routine varevsij");
8636: fprintf(ficgp,"\nunset title \n");
8637: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
8638: 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");
8639: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 8640:
1.218 brouard 8641: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
8642: pstamp(ficresvij);
8643: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
8644: if(popbased==1)
8645: 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);
8646: else
8647: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
8648: fprintf(ficresvij,"# Age");
8649: for(i=1; i<=nlstate;i++)
8650: for(j=1; j<=nlstate;j++)
8651: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
8652: fprintf(ficresvij,"\n");
8653:
8654: xp=vector(1,npar);
8655: dnewm=matrix(1,nlstate,1,npar);
8656: doldm=matrix(1,nlstate,1,nlstate);
8657: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
8658: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
8659:
8660: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
8661: gpp=vector(nlstate+1,nlstate+ndeath);
8662: gmp=vector(nlstate+1,nlstate+ndeath);
8663: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 8664:
1.218 brouard 8665: if(estepm < stepm){
8666: printf ("Problem %d lower than %d\n",estepm, stepm);
8667: }
8668: else hstepm=estepm;
8669: /* For example we decided to compute the life expectancy with the smallest unit */
8670: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
8671: nhstepm is the number of hstepm from age to agelim
8672: nstepm is the number of stepm from age to agelim.
8673: Look at function hpijx to understand why because of memory size limitations,
8674: we decided (b) to get a life expectancy respecting the most precise curvature of the
8675: survival function given by stepm (the optimization length). Unfortunately it
8676: means that if the survival funtion is printed every two years of age and if
8677: you sum them up and add 1 year (area under the trapezoids) you won't get the same
8678: results. So we changed our mind and took the option of the best precision.
8679: */
8680: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
8681: agelim = AGESUP;
8682: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
8683: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
8684: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
8685: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8686: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
8687: gp=matrix(0,nhstepm,1,nlstate);
8688: gm=matrix(0,nhstepm,1,nlstate);
8689:
8690:
8691: for(theta=1; theta <=npar; theta++){
8692: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
8693: xp[i] = x[i] + (i==theta ?delti[theta]:0);
8694: }
1.279 brouard 8695: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
8696: * returns into prlim .
1.288 brouard 8697: */
1.242 brouard 8698: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 8699:
8700: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 8701: if (popbased==1) {
8702: if(mobilav ==0){
8703: for(i=1; i<=nlstate;i++)
8704: prlim[i][i]=probs[(int)age][i][ij];
8705: }else{ /* mobilav */
8706: for(i=1; i<=nlstate;i++)
8707: prlim[i][i]=mobaverage[(int)age][i][ij];
8708: }
8709: }
1.295 brouard 8710: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 8711: */
8712: 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 8713: /**< 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 8714: * at horizon h in state j including mortality.
8715: */
1.218 brouard 8716: for(j=1; j<= nlstate; j++){
8717: for(h=0; h<=nhstepm; h++){
8718: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
8719: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
8720: }
8721: }
1.279 brouard 8722: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 8723: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 8724: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 8725: */
8726: for(j=nlstate+1;j<=nlstate+ndeath;j++){
8727: for(i=1,gpp[j]=0.; i<= nlstate; i++)
8728: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 8729: }
8730:
8731: /* Again with minus shift */
1.218 brouard 8732:
8733: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
8734: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 8735:
1.242 brouard 8736: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 8737:
8738: if (popbased==1) {
8739: if(mobilav ==0){
8740: for(i=1; i<=nlstate;i++)
8741: prlim[i][i]=probs[(int)age][i][ij];
8742: }else{ /* mobilav */
8743: for(i=1; i<=nlstate;i++)
8744: prlim[i][i]=mobaverage[(int)age][i][ij];
8745: }
8746: }
8747:
1.235 brouard 8748: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 8749:
8750: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
8751: for(h=0; h<=nhstepm; h++){
8752: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
8753: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
8754: }
8755: }
8756: /* This for computing probability of death (h=1 means
8757: computed over hstepm matrices product = hstepm*stepm months)
8758: as a weighted average of prlim.
8759: */
8760: for(j=nlstate+1;j<=nlstate+ndeath;j++){
8761: for(i=1,gmp[j]=0.; i<= nlstate; i++)
8762: gmp[j] += prlim[i][i]*p3mat[i][j][1];
8763: }
1.279 brouard 8764: /* end shifting computations */
8765:
8766: /**< Computing gradient matrix at horizon h
8767: */
1.218 brouard 8768: for(j=1; j<= nlstate; j++) /* vareij */
8769: for(h=0; h<=nhstepm; h++){
8770: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
8771: }
1.279 brouard 8772: /**< Gradient of overall mortality p.3 (or p.j)
8773: */
8774: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 8775: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
8776: }
8777:
8778: } /* End theta */
1.279 brouard 8779:
8780: /* We got the gradient matrix for each theta and state j */
1.218 brouard 8781: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
8782:
8783: for(h=0; h<=nhstepm; h++) /* veij */
8784: for(j=1; j<=nlstate;j++)
8785: for(theta=1; theta <=npar; theta++)
8786: trgradg[h][j][theta]=gradg[h][theta][j];
8787:
8788: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
8789: for(theta=1; theta <=npar; theta++)
8790: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 8791: /**< as well as its transposed matrix
8792: */
1.218 brouard 8793:
8794: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
8795: for(i=1;i<=nlstate;i++)
8796: for(j=1;j<=nlstate;j++)
8797: vareij[i][j][(int)age] =0.;
1.279 brouard 8798:
8799: /* Computing trgradg by matcov by gradg at age and summing over h
8800: * and k (nhstepm) formula 15 of article
8801: * Lievre-Brouard-Heathcote
8802: */
8803:
1.218 brouard 8804: for(h=0;h<=nhstepm;h++){
8805: for(k=0;k<=nhstepm;k++){
8806: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
8807: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
8808: for(i=1;i<=nlstate;i++)
8809: for(j=1;j<=nlstate;j++)
8810: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
8811: }
8812: }
8813:
1.279 brouard 8814: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
1.360 ! brouard 8815: * p.j overall mortality formula 19 but computed directly because
1.279 brouard 8816: * we compute the grad (wix pijx) instead of grad (pijx),even if
8817: * wix is independent of theta.
8818: */
1.218 brouard 8819: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
8820: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
8821: for(j=nlstate+1;j<=nlstate+ndeath;j++)
8822: for(i=nlstate+1;i<=nlstate+ndeath;i++)
8823: varppt[j][i]=doldmp[j][i];
8824: /* end ppptj */
8825: /* x centered again */
8826:
1.242 brouard 8827: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 8828:
8829: if (popbased==1) {
8830: if(mobilav ==0){
8831: for(i=1; i<=nlstate;i++)
8832: prlim[i][i]=probs[(int)age][i][ij];
8833: }else{ /* mobilav */
8834: for(i=1; i<=nlstate;i++)
8835: prlim[i][i]=mobaverage[(int)age][i][ij];
8836: }
8837: }
8838:
8839: /* This for computing probability of death (h=1 means
8840: computed over hstepm (estepm) matrices product = hstepm*stepm months)
8841: as a weighted average of prlim.
8842: */
1.235 brouard 8843: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 8844: for(j=nlstate+1;j<=nlstate+ndeath;j++){
8845: for(i=1,gmp[j]=0.;i<= nlstate; i++)
8846: gmp[j] += prlim[i][i]*p3mat[i][j][1];
8847: }
8848: /* end probability of death */
8849:
8850: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
8851: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
8852: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
8853: for(i=1; i<=nlstate;i++){
8854: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
8855: }
8856: }
8857: fprintf(ficresprobmorprev,"\n");
8858:
8859: fprintf(ficresvij,"%.0f ",age );
8860: for(i=1; i<=nlstate;i++)
8861: for(j=1; j<=nlstate;j++){
8862: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
8863: }
8864: fprintf(ficresvij,"\n");
8865: free_matrix(gp,0,nhstepm,1,nlstate);
8866: free_matrix(gm,0,nhstepm,1,nlstate);
8867: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
8868: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
8869: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8870: } /* End age */
8871: free_vector(gpp,nlstate+1,nlstate+ndeath);
8872: free_vector(gmp,nlstate+1,nlstate+ndeath);
8873: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
8874: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
8875: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
8876: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
8877: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
8878: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
8879: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
8880: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
8881: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
8882: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
8883: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
8884: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
8885: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
8886: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
8887: 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);
8888: /* 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 8889: */
1.218 brouard 8890: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
8891: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 8892:
1.218 brouard 8893: free_vector(xp,1,npar);
8894: free_matrix(doldm,1,nlstate,1,nlstate);
8895: free_matrix(dnewm,1,nlstate,1,npar);
8896: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
8897: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
8898: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
8899: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8900: fclose(ficresprobmorprev);
8901: fflush(ficgp);
8902: fflush(fichtm);
8903: } /* end varevsij */
1.126 brouard 8904:
8905: /************ Variance of prevlim ******************/
1.269 brouard 8906: 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 8907: {
1.205 brouard 8908: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 8909: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 8910:
1.268 brouard 8911: double **dnewmpar,**doldm;
1.126 brouard 8912: int i, j, nhstepm, hstepm;
8913: double *xp;
8914: double *gp, *gm;
8915: double **gradg, **trgradg;
1.208 brouard 8916: double **mgm, **mgp;
1.126 brouard 8917: double age,agelim;
8918: int theta;
8919:
8920: pstamp(ficresvpl);
1.288 brouard 8921: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 8922: fprintf(ficresvpl,"# Age ");
8923: if(nresult >=1)
8924: fprintf(ficresvpl," Result# ");
1.126 brouard 8925: for(i=1; i<=nlstate;i++)
8926: fprintf(ficresvpl," %1d-%1d",i,i);
8927: fprintf(ficresvpl,"\n");
8928:
8929: xp=vector(1,npar);
1.268 brouard 8930: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 8931: doldm=matrix(1,nlstate,1,nlstate);
8932:
8933: hstepm=1*YEARM; /* Every year of age */
8934: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
8935: agelim = AGESUP;
8936: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
8937: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
8938: if (stepm >= YEARM) hstepm=1;
8939: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
8940: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 8941: mgp=matrix(1,npar,1,nlstate);
8942: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 8943: gp=vector(1,nlstate);
8944: gm=vector(1,nlstate);
8945:
8946: for(theta=1; theta <=npar; theta++){
8947: for(i=1; i<=npar; i++){ /* Computes gradient */
8948: xp[i] = x[i] + (i==theta ?delti[theta]:0);
8949: }
1.288 brouard 8950: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
8951: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
8952: /* else */
8953: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 8954: for(i=1;i<=nlstate;i++){
1.126 brouard 8955: gp[i] = prlim[i][i];
1.208 brouard 8956: mgp[theta][i] = prlim[i][i];
8957: }
1.126 brouard 8958: for(i=1; i<=npar; i++) /* Computes gradient */
8959: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 8960: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
8961: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
8962: /* else */
8963: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 8964: for(i=1;i<=nlstate;i++){
1.126 brouard 8965: gm[i] = prlim[i][i];
1.208 brouard 8966: mgm[theta][i] = prlim[i][i];
8967: }
1.126 brouard 8968: for(i=1;i<=nlstate;i++)
8969: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 8970: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 8971: } /* End theta */
8972:
8973: trgradg =matrix(1,nlstate,1,npar);
8974:
8975: for(j=1; j<=nlstate;j++)
8976: for(theta=1; theta <=npar; theta++)
8977: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 8978: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
8979: /* printf("\nmgm mgp %d ",(int)age); */
8980: /* for(j=1; j<=nlstate;j++){ */
8981: /* printf(" %d ",j); */
8982: /* for(theta=1; theta <=npar; theta++) */
8983: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
8984: /* printf("\n "); */
8985: /* } */
8986: /* } */
8987: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
8988: /* printf("\n gradg %d ",(int)age); */
8989: /* for(j=1; j<=nlstate;j++){ */
8990: /* printf("%d ",j); */
8991: /* for(theta=1; theta <=npar; theta++) */
8992: /* printf("%d %lf ",theta,gradg[theta][j]); */
8993: /* printf("\n "); */
8994: /* } */
8995: /* } */
1.126 brouard 8996:
8997: for(i=1;i<=nlstate;i++)
8998: varpl[i][(int)age] =0.;
1.209 brouard 8999: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 9000: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
9001: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 9002: }else{
1.268 brouard 9003: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
9004: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 9005: }
1.126 brouard 9006: for(i=1;i<=nlstate;i++)
9007: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
9008:
9009: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 9010: if(nresult >=1)
9011: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 9012: for(i=1; i<=nlstate;i++){
1.126 brouard 9013: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 9014: /* for(j=1;j<=nlstate;j++) */
9015: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
9016: }
1.126 brouard 9017: fprintf(ficresvpl,"\n");
9018: free_vector(gp,1,nlstate);
9019: free_vector(gm,1,nlstate);
1.208 brouard 9020: free_matrix(mgm,1,npar,1,nlstate);
9021: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 9022: free_matrix(gradg,1,npar,1,nlstate);
9023: free_matrix(trgradg,1,nlstate,1,npar);
9024: } /* End age */
9025:
9026: free_vector(xp,1,npar);
9027: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 9028: free_matrix(dnewmpar,1,nlstate,1,nlstate);
9029:
9030: }
9031:
9032:
9033: /************ Variance of backprevalence limit ******************/
1.269 brouard 9034: 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 9035: {
9036: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
9037: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
9038:
9039: double **dnewmpar,**doldm;
9040: int i, j, nhstepm, hstepm;
9041: double *xp;
9042: double *gp, *gm;
9043: double **gradg, **trgradg;
9044: double **mgm, **mgp;
9045: double age,agelim;
9046: int theta;
9047:
9048: pstamp(ficresvbl);
9049: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
9050: fprintf(ficresvbl,"# Age ");
9051: if(nresult >=1)
9052: fprintf(ficresvbl," Result# ");
9053: for(i=1; i<=nlstate;i++)
9054: fprintf(ficresvbl," %1d-%1d",i,i);
9055: fprintf(ficresvbl,"\n");
9056:
9057: xp=vector(1,npar);
9058: dnewmpar=matrix(1,nlstate,1,npar);
9059: doldm=matrix(1,nlstate,1,nlstate);
9060:
9061: hstepm=1*YEARM; /* Every year of age */
9062: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
9063: agelim = AGEINF;
9064: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
9065: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9066: if (stepm >= YEARM) hstepm=1;
9067: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9068: gradg=matrix(1,npar,1,nlstate);
9069: mgp=matrix(1,npar,1,nlstate);
9070: mgm=matrix(1,npar,1,nlstate);
9071: gp=vector(1,nlstate);
9072: gm=vector(1,nlstate);
9073:
9074: for(theta=1; theta <=npar; theta++){
9075: for(i=1; i<=npar; i++){ /* Computes gradient */
9076: xp[i] = x[i] + (i==theta ?delti[theta]:0);
9077: }
9078: if(mobilavproj > 0 )
9079: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
9080: else
9081: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
9082: for(i=1;i<=nlstate;i++){
9083: gp[i] = bprlim[i][i];
9084: mgp[theta][i] = bprlim[i][i];
9085: }
9086: for(i=1; i<=npar; i++) /* Computes gradient */
9087: xp[i] = x[i] - (i==theta ?delti[theta]:0);
9088: if(mobilavproj > 0 )
9089: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
9090: else
9091: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
9092: for(i=1;i<=nlstate;i++){
9093: gm[i] = bprlim[i][i];
9094: mgm[theta][i] = bprlim[i][i];
9095: }
9096: for(i=1;i<=nlstate;i++)
9097: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
9098: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
9099: } /* End theta */
9100:
9101: trgradg =matrix(1,nlstate,1,npar);
9102:
9103: for(j=1; j<=nlstate;j++)
9104: for(theta=1; theta <=npar; theta++)
9105: trgradg[j][theta]=gradg[theta][j];
9106: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
9107: /* printf("\nmgm mgp %d ",(int)age); */
9108: /* for(j=1; j<=nlstate;j++){ */
9109: /* printf(" %d ",j); */
9110: /* for(theta=1; theta <=npar; theta++) */
9111: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
9112: /* printf("\n "); */
9113: /* } */
9114: /* } */
9115: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
9116: /* printf("\n gradg %d ",(int)age); */
9117: /* for(j=1; j<=nlstate;j++){ */
9118: /* printf("%d ",j); */
9119: /* for(theta=1; theta <=npar; theta++) */
9120: /* printf("%d %lf ",theta,gradg[theta][j]); */
9121: /* printf("\n "); */
9122: /* } */
9123: /* } */
9124:
9125: for(i=1;i<=nlstate;i++)
9126: varbpl[i][(int)age] =0.;
9127: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
9128: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
9129: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
9130: }else{
9131: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
9132: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
9133: }
9134: for(i=1;i<=nlstate;i++)
9135: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
9136:
9137: fprintf(ficresvbl,"%.0f ",age );
9138: if(nresult >=1)
9139: fprintf(ficresvbl,"%d ",nres );
9140: for(i=1; i<=nlstate;i++)
9141: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
9142: fprintf(ficresvbl,"\n");
9143: free_vector(gp,1,nlstate);
9144: free_vector(gm,1,nlstate);
9145: free_matrix(mgm,1,npar,1,nlstate);
9146: free_matrix(mgp,1,npar,1,nlstate);
9147: free_matrix(gradg,1,npar,1,nlstate);
9148: free_matrix(trgradg,1,nlstate,1,npar);
9149: } /* End age */
9150:
9151: free_vector(xp,1,npar);
9152: free_matrix(doldm,1,nlstate,1,npar);
9153: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 9154:
9155: }
9156:
9157: /************ Variance of one-step probabilities ******************/
9158: 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 9159: {
9160: int i, j=0, k1, l1, tj;
9161: int k2, l2, j1, z1;
9162: int k=0, l;
9163: int first=1, first1, first2;
1.326 brouard 9164: int nres=0; /* New */
1.222 brouard 9165: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
9166: double **dnewm,**doldm;
9167: double *xp;
9168: double *gp, *gm;
9169: double **gradg, **trgradg;
9170: double **mu;
9171: double age, cov[NCOVMAX+1];
9172: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
9173: int theta;
9174: char fileresprob[FILENAMELENGTH];
9175: char fileresprobcov[FILENAMELENGTH];
9176: char fileresprobcor[FILENAMELENGTH];
9177: double ***varpij;
9178:
9179: strcpy(fileresprob,"PROB_");
1.356 brouard 9180: strcat(fileresprob,fileresu);
1.222 brouard 9181: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
9182: printf("Problem with resultfile: %s\n", fileresprob);
9183: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
9184: }
9185: strcpy(fileresprobcov,"PROBCOV_");
9186: strcat(fileresprobcov,fileresu);
9187: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
9188: printf("Problem with resultfile: %s\n", fileresprobcov);
9189: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
9190: }
9191: strcpy(fileresprobcor,"PROBCOR_");
9192: strcat(fileresprobcor,fileresu);
9193: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
9194: printf("Problem with resultfile: %s\n", fileresprobcor);
9195: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
9196: }
9197: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
9198: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
9199: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
9200: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
9201: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
9202: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
9203: pstamp(ficresprob);
9204: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
9205: fprintf(ficresprob,"# Age");
9206: pstamp(ficresprobcov);
9207: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
9208: fprintf(ficresprobcov,"# Age");
9209: pstamp(ficresprobcor);
9210: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
9211: fprintf(ficresprobcor,"# Age");
1.126 brouard 9212:
9213:
1.222 brouard 9214: for(i=1; i<=nlstate;i++)
9215: for(j=1; j<=(nlstate+ndeath);j++){
9216: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
9217: fprintf(ficresprobcov," p%1d-%1d ",i,j);
9218: fprintf(ficresprobcor," p%1d-%1d ",i,j);
9219: }
9220: /* fprintf(ficresprob,"\n");
9221: fprintf(ficresprobcov,"\n");
9222: fprintf(ficresprobcor,"\n");
9223: */
9224: xp=vector(1,npar);
9225: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
9226: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
9227: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
9228: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
9229: first=1;
9230: fprintf(ficgp,"\n# Routine varprob");
9231: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
9232: fprintf(fichtm,"\n");
9233:
1.288 brouard 9234: 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 9235: 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);
9236: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 9237: and drawn. It helps understanding how is the covariance between two incidences.\
9238: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 9239: 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 9240: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
9241: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
9242: standard deviations wide on each axis. <br>\
9243: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
9244: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
9245: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
9246:
1.222 brouard 9247: cov[1]=1;
9248: /* tj=cptcoveff; */
1.225 brouard 9249: tj = (int) pow(2,cptcoveff);
1.222 brouard 9250: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
9251: j1=0;
1.332 brouard 9252:
9253: for(nres=1;nres <=nresult; nres++){ /* For each resultline */
9254: for(j1=1; j1<=tj;j1++){ /* For any combination of dummy covariates, fixed and varying */
1.342 brouard 9255: /* 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 9256: if(tj != 1 && TKresult[nres]!= j1)
9257: continue;
9258:
9259: /* for(j1=1; j1<=tj;j1++){ /\* For each valid combination of covariates or only once*\/ */
9260: /* for(nres=1;nres <=1; nres++){ /\* For each resultline *\/ */
9261: /* /\* for(nres=1;nres <=nresult; nres++){ /\\* For each resultline *\\/ *\/ */
1.222 brouard 9262: if (cptcovn>0) {
1.334 brouard 9263: fprintf(ficresprob, "\n#********** Variable ");
9264: fprintf(ficresprobcov, "\n#********** Variable ");
9265: fprintf(ficgp, "\n#********** Variable ");
9266: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
9267: fprintf(ficresprobcor, "\n#********** Variable ");
9268:
9269: /* Including quantitative variables of the resultline to be done */
9270: for (z1=1; z1<=cptcovs; z1++){ /* Loop on each variable of this resultline */
1.343 brouard 9271: /* printf("Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model); */
1.338 brouard 9272: fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model);
9273: /* 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 9274: if(Dummy[modelresult[nres][z1]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to z1 in resultline */
9275: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed referenced to model equation */
9276: 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 */
9277: 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 */
9278: 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 */
9279: 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 */
9280: 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 */
9281: fprintf(ficresprob,"fixed ");
9282: fprintf(ficresprobcov,"fixed ");
9283: fprintf(ficgp,"fixed ");
9284: fprintf(fichtmcov,"fixed ");
9285: fprintf(ficresprobcor,"fixed ");
9286: }else{
9287: fprintf(ficresprob,"varyi ");
9288: fprintf(ficresprobcov,"varyi ");
9289: fprintf(ficgp,"varyi ");
9290: fprintf(fichtmcov,"varyi ");
9291: fprintf(ficresprobcor,"varyi ");
9292: }
9293: }else if(Dummy[modelresult[nres][z1]]==1){ /* Quanti variable */
9294: /* For each selected (single) quantitative value */
1.337 brouard 9295: fprintf(ficresprob," V%d=%lg ",Tvqresult[nres][z1],Tqresult[nres][z1]);
1.334 brouard 9296: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed */
9297: fprintf(ficresprob,"fixed ");
9298: fprintf(ficresprobcov,"fixed ");
9299: fprintf(ficgp,"fixed ");
9300: fprintf(fichtmcov,"fixed ");
9301: fprintf(ficresprobcor,"fixed ");
9302: }else{
9303: fprintf(ficresprob,"varyi ");
9304: fprintf(ficresprobcov,"varyi ");
9305: fprintf(ficgp,"varyi ");
9306: fprintf(fichtmcov,"varyi ");
9307: fprintf(ficresprobcor,"varyi ");
9308: }
9309: }else{
9310: 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 */
9311: 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 */
9312: exit(1);
9313: }
9314: } /* End loop on variable of this resultline */
9315: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]); */
1.222 brouard 9316: fprintf(ficresprob, "**********\n#\n");
9317: fprintf(ficresprobcov, "**********\n#\n");
9318: fprintf(ficgp, "**********\n#\n");
9319: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
9320: fprintf(ficresprobcor, "**********\n#");
9321: if(invalidvarcomb[j1]){
9322: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
9323: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
9324: continue;
9325: }
9326: }
9327: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
9328: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
9329: gp=vector(1,(nlstate)*(nlstate+ndeath));
9330: gm=vector(1,(nlstate)*(nlstate+ndeath));
1.334 brouard 9331: for (age=bage; age<=fage; age ++){ /* Fo each age we feed the model equation with covariates, using precov as in hpxij() ? */
1.222 brouard 9332: cov[2]=age;
9333: if(nagesqr==1)
9334: cov[3]= age*age;
1.334 brouard 9335: /* New code end of combination but for each resultline */
9336: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.349 brouard 9337: if(Typevar[k1]==1 || Typevar[k1] ==3){ /* A product with age */
1.334 brouard 9338: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.326 brouard 9339: }else{
1.334 brouard 9340: cov[2+nagesqr+k1]=precov[nres][k1];
1.326 brouard 9341: }
1.334 brouard 9342: }/* End of loop on model equation */
9343: /* Old code */
9344: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
9345: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; *\/ */
9346: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
9347: /* /\* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates *\/ */
9348: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TnsdVar[TvarsD[k]])]; */
9349: /* /\*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*\//\* j1 1 2 3 4 */
9350: /* * 1 1 1 1 1 */
9351: /* * 2 2 1 1 1 */
9352: /* * 3 1 2 1 1 */
9353: /* *\/ */
9354: /* /\* nbcode[1][1]=0 nbcode[1][2]=1;*\/ */
9355: /* } */
9356: /* /\* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
9357: /* /\* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] *\/ */
9358: /* /\*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; *\/ */
9359: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
9360: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
9361: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,TnsdVar[Tvar[Tage[k]]])]*cov[2]; */
9362: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
9363: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
9364: /* 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]); */
9365: /* /\* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\\* Using the mean of quantitative variable Tvar[Tage[k]] /\\* Tqresult[nres][k]; *\\/ *\/ */
9366: /* /\* exit(1); *\/ */
9367: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
9368: /* } */
9369: /* /\* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
9370: /* } */
9371: /* for (k=1; k<=cptcovprod;k++){/\* For product without age *\/ */
9372: /* if(Dummy[Tvard[k][1]]==0){ */
9373: /* if(Dummy[Tvard[k][2]]==0){ */
9374: /* 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]])]; */
9375: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
9376: /* }else{ /\* Should we use the mean of the quantitative variables? *\/ */
9377: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * Tqresult[nres][resultmodel[nres][k]]; */
9378: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
9379: /* } */
9380: /* }else{ */
9381: /* if(Dummy[Tvard[k][2]]==0){ */
9382: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][TnsdVar[Tvard[k][1]]]; */
9383: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
9384: /* }else{ */
9385: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][TnsdVar[Tvard[k][1]]]* Tqinvresult[nres][TnsdVar[Tvard[k][2]]]; */
9386: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
9387: /* } */
9388: /* } */
9389: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
9390: /* } */
1.326 brouard 9391: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 9392: for(theta=1; theta <=npar; theta++){
9393: for(i=1; i<=npar; i++)
9394: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 9395:
1.222 brouard 9396: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 9397:
1.222 brouard 9398: k=0;
9399: for(i=1; i<= (nlstate); i++){
9400: for(j=1; j<=(nlstate+ndeath);j++){
9401: k=k+1;
9402: gp[k]=pmmij[i][j];
9403: }
9404: }
1.220 brouard 9405:
1.222 brouard 9406: for(i=1; i<=npar; i++)
9407: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 9408:
1.222 brouard 9409: pmij(pmmij,cov,ncovmodel,xp,nlstate);
9410: k=0;
9411: for(i=1; i<=(nlstate); i++){
9412: for(j=1; j<=(nlstate+ndeath);j++){
9413: k=k+1;
9414: gm[k]=pmmij[i][j];
9415: }
9416: }
1.220 brouard 9417:
1.222 brouard 9418: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
9419: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
9420: }
1.126 brouard 9421:
1.222 brouard 9422: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
9423: for(theta=1; theta <=npar; theta++)
9424: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 9425:
1.222 brouard 9426: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
9427: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 9428:
1.222 brouard 9429: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 9430:
1.222 brouard 9431: k=0;
9432: for(i=1; i<=(nlstate); i++){
9433: for(j=1; j<=(nlstate+ndeath);j++){
9434: k=k+1;
9435: mu[k][(int) age]=pmmij[i][j];
9436: }
9437: }
9438: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
9439: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
9440: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 9441:
1.222 brouard 9442: /*printf("\n%d ",(int)age);
9443: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
9444: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
9445: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
9446: }*/
1.220 brouard 9447:
1.222 brouard 9448: fprintf(ficresprob,"\n%d ",(int)age);
9449: fprintf(ficresprobcov,"\n%d ",(int)age);
9450: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 9451:
1.222 brouard 9452: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
9453: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
9454: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
9455: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
9456: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
9457: }
9458: i=0;
9459: for (k=1; k<=(nlstate);k++){
9460: for (l=1; l<=(nlstate+ndeath);l++){
9461: i++;
9462: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
9463: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
9464: for (j=1; j<=i;j++){
9465: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
9466: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
9467: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
9468: }
9469: }
9470: }/* end of loop for state */
9471: } /* end of loop for age */
9472: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
9473: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
9474: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
9475: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
9476:
9477: /* Confidence intervalle of pij */
9478: /*
9479: fprintf(ficgp,"\nunset parametric;unset label");
9480: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
9481: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
9482: 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);
9483: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
9484: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
9485: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
9486: */
9487:
9488: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
9489: first1=1;first2=2;
9490: for (k2=1; k2<=(nlstate);k2++){
9491: for (l2=1; l2<=(nlstate+ndeath);l2++){
9492: if(l2==k2) continue;
9493: j=(k2-1)*(nlstate+ndeath)+l2;
9494: for (k1=1; k1<=(nlstate);k1++){
9495: for (l1=1; l1<=(nlstate+ndeath);l1++){
9496: if(l1==k1) continue;
9497: i=(k1-1)*(nlstate+ndeath)+l1;
9498: if(i<=j) continue;
9499: for (age=bage; age<=fage; age ++){
9500: if ((int)age %5==0){
9501: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
9502: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
9503: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
9504: mu1=mu[i][(int) age]/stepm*YEARM ;
9505: mu2=mu[j][(int) age]/stepm*YEARM;
9506: c12=cv12/sqrt(v1*v2);
9507: /* Computing eigen value of matrix of covariance */
9508: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
9509: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
9510: if ((lc2 <0) || (lc1 <0) ){
9511: if(first2==1){
9512: first1=0;
9513: 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);
9514: }
9515: 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);
9516: /* lc1=fabs(lc1); */ /* If we want to have them positive */
9517: /* lc2=fabs(lc2); */
9518: }
1.220 brouard 9519:
1.222 brouard 9520: /* Eigen vectors */
1.280 brouard 9521: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
9522: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
9523: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
9524: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
9525: }else
9526: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 9527: /*v21=sqrt(1.-v11*v11); *//* error */
9528: v21=(lc1-v1)/cv12*v11;
9529: v12=-v21;
9530: v22=v11;
9531: tnalp=v21/v11;
9532: if(first1==1){
9533: first1=0;
9534: 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);
9535: }
9536: 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);
9537: /*printf(fignu*/
9538: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
9539: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
9540: if(first==1){
9541: first=0;
9542: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
9543: fprintf(ficgp,"\nset parametric;unset label");
9544: 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);
9545: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 9546: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 9547: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 9548: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 9549: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
9550: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
9551: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
9552: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
9553: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
9554: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
9555: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
9556: 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 9557: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
9558: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 9559: }else{
9560: first=0;
9561: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
9562: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
9563: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
9564: 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 9565: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
9566: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 9567: }/* if first */
9568: } /* age mod 5 */
9569: } /* end loop age */
9570: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
9571: first=1;
9572: } /*l12 */
9573: } /* k12 */
9574: } /*l1 */
9575: }/* k1 */
1.332 brouard 9576: } /* loop on combination of covariates j1 */
1.326 brouard 9577: } /* loop on nres */
1.222 brouard 9578: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
9579: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
9580: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
9581: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
9582: free_vector(xp,1,npar);
9583: fclose(ficresprob);
9584: fclose(ficresprobcov);
9585: fclose(ficresprobcor);
9586: fflush(ficgp);
9587: fflush(fichtmcov);
9588: }
1.126 brouard 9589:
9590:
9591: /******************* Printing html file ***********/
1.201 brouard 9592: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9593: int lastpass, int stepm, int weightopt, char model[],\
9594: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 9595: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
9596: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
9597: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.359 brouard 9598: int jj1, k1, cpt, nres;
1.319 brouard 9599: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 9600: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
9601: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
9602: </ul>");
1.319 brouard 9603: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
9604: /* </ul>", model); */
1.214 brouard 9605: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
9606: 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",
9607: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
1.332 brouard 9608: 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 9609: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
9610: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 9611: fprintf(fichtm,"\
9612: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 9613: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 9614: fprintf(fichtm,"\
1.217 brouard 9615: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
9616: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
9617: fprintf(fichtm,"\
1.288 brouard 9618: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 9619: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 9620: fprintf(fichtm,"\
1.288 brouard 9621: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 9622: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
9623: fprintf(fichtm,"\
1.211 brouard 9624: - (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 9625: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 9626: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 9627: if(prevfcast==1){
9628: fprintf(fichtm,"\
9629: - Prevalence projections by age and states: \
1.201 brouard 9630: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 9631: }
1.126 brouard 9632:
9633:
1.225 brouard 9634: m=pow(2,cptcoveff);
1.222 brouard 9635: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 9636:
1.317 brouard 9637: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 9638:
9639: jj1=0;
9640:
9641: fprintf(fichtm," \n<ul>");
1.337 brouard 9642: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9643: /* k1=nres; */
1.338 brouard 9644: k1=TKresult[nres];
9645: if(TKresult[nres]==0)k1=1; /* To be checked for no result */
1.337 brouard 9646: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
9647: /* if(m != 1 && TKresult[nres]!= k1) */
9648: /* continue; */
1.264 brouard 9649: jj1++;
9650: if (cptcovn > 0) {
9651: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
1.337 brouard 9652: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
9653: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 9654: }
1.337 brouard 9655: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
9656: /* fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
9657: /* } */
9658: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
9659: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
9660: /* } */
1.264 brouard 9661: fprintf(fichtm,"\">");
9662:
9663: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
9664: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 9665: for (cpt=1; cpt<=cptcovs;cpt++){
9666: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 9667: }
1.337 brouard 9668: /* fprintf(fichtm,"************ Results for covariates"); */
9669: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
9670: /* fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
9671: /* } */
9672: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
9673: /* fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
9674: /* } */
1.264 brouard 9675: if(invalidvarcomb[k1]){
9676: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
9677: continue;
9678: }
9679: fprintf(fichtm,"</a></li>");
9680: } /* cptcovn >0 */
9681: }
1.317 brouard 9682: fprintf(fichtm," \n</ul>");
1.264 brouard 9683:
1.222 brouard 9684: jj1=0;
1.237 brouard 9685:
1.337 brouard 9686: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9687: /* k1=nres; */
1.338 brouard 9688: k1=TKresult[nres];
9689: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 9690: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
9691: /* if(m != 1 && TKresult[nres]!= k1) */
9692: /* continue; */
1.220 brouard 9693:
1.222 brouard 9694: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
9695: jj1++;
9696: if (cptcovn > 0) {
1.264 brouard 9697: fprintf(fichtm,"\n<p><a name=\"rescov");
1.337 brouard 9698: for (cpt=1; cpt<=cptcovs;cpt++){
9699: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 9700: }
1.337 brouard 9701: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
9702: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
9703: /* } */
1.264 brouard 9704: fprintf(fichtm,"\"</a>");
9705:
1.222 brouard 9706: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 9707: for (cpt=1; cpt<=cptcovs;cpt++){
9708: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
9709: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 9710: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
9711: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 9712: }
1.230 brouard 9713: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.338 brouard 9714: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 9715: if(invalidvarcomb[k1]){
9716: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
9717: printf("\nCombination (%d) ignored because no cases \n",k1);
9718: continue;
9719: }
9720: }
9721: /* aij, bij */
1.259 brouard 9722: 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 9723: <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 9724: /* Pij */
1.241 brouard 9725: 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> \
9726: <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 9727: /* Quasi-incidences */
9728: 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 9729: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 9730: 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 9731: 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> \
9732: <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 9733: /* Survival functions (period) in state j */
9734: for(cpt=1; cpt<=nlstate;cpt++){
1.359 brouard 9735: 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 9736: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
9737: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 9738: }
9739: /* State specific survival functions (period) */
9740: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 9741: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
1.359 brouard 9742: 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 9743: <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);
9744: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
9745: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 9746: }
1.288 brouard 9747: /* Period (forward stable) prevalence in each health state */
1.222 brouard 9748: for(cpt=1; cpt<=nlstate;cpt++){
1.359 brouard 9749: 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 9750: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
1.329 brouard 9751: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 9752: }
1.296 brouard 9753: if(prevbcast==1){
1.288 brouard 9754: /* Backward prevalence in each health state */
1.222 brouard 9755: for(cpt=1; cpt<=nlstate;cpt++){
1.338 brouard 9756: 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);
9757: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJB_"),subdirf2(optionfilefiname,"PIJB_"));
9758: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.222 brouard 9759: }
1.217 brouard 9760: }
1.222 brouard 9761: if(prevfcast==1){
1.288 brouard 9762: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 9763: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 9764: 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);
9765: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
9766: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
9767: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 9768: }
9769: }
1.296 brouard 9770: if(prevbcast==1){
1.268 brouard 9771: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
9772: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 9773: 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 9774: 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 \
9775: 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 9776: 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);
9777: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
9778: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 9779: }
9780: }
1.220 brouard 9781:
1.222 brouard 9782: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 9783: 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);
9784: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
9785: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 9786: }
9787: /* } /\* end i1 *\/ */
1.337 brouard 9788: }/* End k1=nres */
1.222 brouard 9789: fprintf(fichtm,"</ul>");
1.126 brouard 9790:
1.222 brouard 9791: fprintf(fichtm,"\
1.126 brouard 9792: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 9793: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 9794: - 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 9795: But because parameters are usually highly correlated (a higher incidence of disability \
9796: and a higher incidence of recovery can give very close observed transition) it might \
9797: be very useful to look not only at linear confidence intervals estimated from the \
9798: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
9799: (parameters) of the logistic regression, it might be more meaningful to visualize the \
9800: covariance matrix of the one-step probabilities. \
9801: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 9802:
1.222 brouard 9803: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
9804: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
9805: fprintf(fichtm,"\
1.126 brouard 9806: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 9807: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 9808:
1.222 brouard 9809: fprintf(fichtm,"\
1.126 brouard 9810: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 9811: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
9812: fprintf(fichtm,"\
1.126 brouard 9813: - 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): \
9814: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 9815: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 9816: fprintf(fichtm,"\
1.126 brouard 9817: - (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): \
9818: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 9819: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 9820: fprintf(fichtm,"\
1.288 brouard 9821: - 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 9822: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
9823: fprintf(fichtm,"\
1.128 brouard 9824: - 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 9825: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
9826: fprintf(fichtm,"\
1.288 brouard 9827: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 9828: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 9829:
9830: /* if(popforecast==1) fprintf(fichtm,"\n */
9831: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
9832: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
9833: /* <br>",fileres,fileres,fileres,fileres); */
9834: /* else */
1.338 brouard 9835: /* 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 9836: fflush(fichtm);
1.126 brouard 9837:
1.225 brouard 9838: m=pow(2,cptcoveff);
1.222 brouard 9839: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 9840:
1.317 brouard 9841: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
9842:
9843: jj1=0;
9844:
9845: fprintf(fichtm," \n<ul>");
1.337 brouard 9846: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9847: /* k1=nres; */
1.338 brouard 9848: k1=TKresult[nres];
1.337 brouard 9849: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
9850: /* if(m != 1 && TKresult[nres]!= k1) */
9851: /* continue; */
1.317 brouard 9852: jj1++;
9853: if (cptcovn > 0) {
9854: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
1.337 brouard 9855: for (cpt=1; cpt<=cptcovs;cpt++){
9856: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 9857: }
9858: fprintf(fichtm,"\">");
9859:
9860: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
9861: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 9862: for (cpt=1; cpt<=cptcovs;cpt++){
9863: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 9864: }
9865: if(invalidvarcomb[k1]){
9866: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
9867: continue;
9868: }
9869: fprintf(fichtm,"</a></li>");
9870: } /* cptcovn >0 */
1.337 brouard 9871: } /* End nres */
1.317 brouard 9872: fprintf(fichtm," \n</ul>");
9873:
1.222 brouard 9874: jj1=0;
1.237 brouard 9875:
1.241 brouard 9876: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 9877: /* k1=nres; */
1.338 brouard 9878: k1=TKresult[nres];
9879: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 9880: /* for(k1=1; k1<=m;k1++){ */
9881: /* if(m != 1 && TKresult[nres]!= k1) */
9882: /* continue; */
1.222 brouard 9883: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
9884: jj1++;
1.126 brouard 9885: if (cptcovn > 0) {
1.317 brouard 9886: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
1.337 brouard 9887: for (cpt=1; cpt<=cptcovs;cpt++){
9888: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 9889: }
9890: fprintf(fichtm,"\"</a>");
9891:
1.126 brouard 9892: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 9893: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcoveff number of variables */
9894: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
9895: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 9896: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 9897: }
1.237 brouard 9898:
1.338 brouard 9899: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 9900:
1.222 brouard 9901: if(invalidvarcomb[k1]){
9902: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
9903: continue;
9904: }
1.337 brouard 9905: } /* If cptcovn >0 */
1.126 brouard 9906: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 9907: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 9908: 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);
9909: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
9910: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 9911: }
9912: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.360 ! brouard 9913: health expectancies in each live state (1 to %d) with confidence intervals \
! 9914: on left y-scale as well as proportions of time spent in each live state \
! 9915: (with confidence intervals) on right y-scale 0 to 100%%.\
! 9916: If popbased=1 the smooth (due to the model) \
1.128 brouard 9917: true period expectancies (those weighted with period prevalences are also\
9918: drawn in addition to the population based expectancies computed using\
1.314 brouard 9919: 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);
9920: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
9921: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 9922: /* } /\* end i1 *\/ */
1.241 brouard 9923: }/* End nres */
1.222 brouard 9924: fprintf(fichtm,"</ul>");
9925: fflush(fichtm);
1.126 brouard 9926: }
9927:
9928: /******************* Gnuplot file **************/
1.296 brouard 9929: 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 9930:
1.354 brouard 9931: char dirfileres[256],optfileres[256];
9932: char gplotcondition[256], gplotlabel[256];
1.343 brouard 9933: 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;
1.211 brouard 9934: int lv=0, vlv=0, kl=0;
1.130 brouard 9935: int ng=0;
1.201 brouard 9936: int vpopbased;
1.223 brouard 9937: int ioffset; /* variable offset for columns */
1.270 brouard 9938: int iyearc=1; /* variable column for year of projection */
9939: int iagec=1; /* variable column for age of projection */
1.235 brouard 9940: int nres=0; /* Index of resultline */
1.266 brouard 9941: int istart=1; /* For starting graphs in projections */
1.219 brouard 9942:
1.126 brouard 9943: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
9944: /* printf("Problem with file %s",optionfilegnuplot); */
9945: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
9946: /* } */
9947:
9948: /*#ifdef windows */
9949: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 9950: /*#endif */
1.225 brouard 9951: m=pow(2,cptcoveff);
1.126 brouard 9952:
1.274 brouard 9953: /* diagram of the model */
9954: fprintf(ficgp,"\n#Diagram of the model \n");
9955: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
9956: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
9957: 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);
9958:
1.343 brouard 9959: 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 9960: fprintf(ficgp,"\n#show arrow\nunset label\n");
9961: 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);
9962: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
9963: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
9964: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
9965: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
9966:
1.202 brouard 9967: /* Contribution to likelihood */
9968: /* Plot the probability implied in the likelihood */
1.223 brouard 9969: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
9970: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
9971: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
9972: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 9973: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 9974: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
9975: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 9976: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
9977: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
9978: 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));
9979: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
9980: 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));
9981: for (i=1; i<= nlstate ; i ++) {
9982: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
9983: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
9984: 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);
9985: for (j=2; j<= nlstate+ndeath ; j ++) {
9986: 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);
9987: }
9988: fprintf(ficgp,";\nset out; unset ylabel;\n");
9989: }
9990: /* 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 */
9991: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
9992: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
9993: fprintf(ficgp,"\nset out;unset log\n");
9994: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 9995:
1.343 brouard 9996: /* Plot the probability implied in the likelihood by covariate value */
9997: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
9998: /* if(debugILK==1){ */
9999: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
1.347 brouard 10000: kvar=Tvar[TvarFind[kf]]; /* variable name */
10001: /* k=18+Tvar[TvarFind[kf]];/\*offset because there are 18 columns in the ILK_ file but could be placed else where *\/ */
1.350 brouard 10002: /* k=18+kf;/\*offset because there are 18 columns in the ILK_ file *\/ */
1.356 brouard 10003: /* k=19+kf;/\*offset because there are 19 columns in the ILK_ file *\/ */
1.355 brouard 10004: 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 10005: for (i=1; i<= nlstate ; i ++) {
10006: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
10007: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
1.348 brouard 10008: if(gnuplotversion >=5.2){ /* Former gnuplot versions do not have variable pointsize!! */
10009: 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);
10010: for (j=2; j<= nlstate+ndeath ; j ++) {
10011: 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);
10012: }
10013: }else{
10014: 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);
10015: for (j=2; j<= nlstate+ndeath ; j ++) {
10016: 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);
10017: }
1.343 brouard 10018: }
10019: fprintf(ficgp,";\nset out; unset ylabel;\n");
10020: }
10021: } /* End of each covariate dummy */
10022: for(ncovv=1, iposold=0, kk=0; ncovv <= ncovvt ; ncovv++){
10023: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
10024: * kmodel = 1 2 3 4 5 6 7 8 9
10025: * varying 1 2 3 4 5
10026: * ncovv 1 2 3 4 5 6 7 8
10027: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
10028: * TvarVVind[ncovv]=kmodel 2 3 7 7 8 8 9 9
10029: * TvarFind[kmodel] 1 0 0 0 0 0 0 0 0
10030: * kdata ncovcol=[V1 V2] nqv=0 ntv=[V3 V4] nqtv=V5
10031: * Dummy[kmodel] 0 0 1 2 2 3 1 1 1
10032: */
10033: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
10034: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
10035: /* 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]); */
10036: if(ipos!=iposold){ /* Not a product or first of a product */
10037: /* printf(" %d",ipos); */
10038: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
10039: /* printf(" DebugILK ficgp suite ipos=%d != iposold=%d\n", ipos, iposold); */
10040: kk++; /* Position of the ncovv column in ILK_ */
10041: k=18+ncovf+kk; /*offset because there are 18 columns in the ILK_ file plus ncovf fixed covariate */
10042: 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) */
10043: for (i=1; i<= nlstate ; i ++) {
10044: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
10045: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
10046:
1.348 brouard 10047: /* printf("Before DebugILK gnuplotversion=%g >=5.2\n",gnuplotversion); */
1.343 brouard 10048: if(gnuplotversion >=5.2){ /* Former gnuplot versions do not have variable pointsize!! */
10049: /* printf("DebugILK gnuplotversion=%g >=5.2\n",gnuplotversion); */
10050: 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);
10051: for (j=2; j<= nlstate+ndeath ; j ++) {
10052: 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);
10053: }
10054: }else{
10055: /* printf("DebugILK gnuplotversion=%g <5.2\n",gnuplotversion); */
10056: 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);
10057: for (j=2; j<= nlstate+ndeath ; j ++) {
10058: 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);
10059: }
10060: }
10061: fprintf(ficgp,";\nset out; unset ylabel;\n");
10062: }
10063: }/* End if dummy varying */
10064: }else{ /*Product */
10065: /* printf("*"); */
10066: /* fprintf(ficresilk,"*"); */
10067: }
10068: iposold=ipos;
10069: } /* For each time varying covariate */
10070: /* } /\* debugILK==1 *\/ */
10071: /* 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 */
10072: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
10073: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
10074: fprintf(ficgp,"\nset out;unset log\n");
10075: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
10076:
10077:
10078:
1.126 brouard 10079: strcpy(dirfileres,optionfilefiname);
10080: strcpy(optfileres,"vpl");
1.223 brouard 10081: /* 1eme*/
1.238 brouard 10082: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
1.337 brouard 10083: /* for (k1=1; k1<= m ; k1 ++){ /\* For each valid combination of covariate *\/ */
1.236 brouard 10084: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10085: k1=TKresult[nres];
1.338 brouard 10086: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.238 brouard 10087: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.337 brouard 10088: /* if(m != 1 && TKresult[nres]!= k1) */
10089: /* continue; */
1.238 brouard 10090: /* We are interested in selected combination by the resultline */
1.246 brouard 10091: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 10092: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 10093: strcpy(gplotlabel,"(");
1.337 brouard 10094: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10095: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10096: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10097:
10098: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate k get corresponding value lv for combination k1 *\/ */
10099: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the value of the covariate corresponding to k1 combination *\\/ *\/ */
10100: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10101: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10102: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10103: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10104: /* vlv= nbcode[Tvaraff[k]][lv]; /\* vlv is the value of the covariate lv, 0 or 1 *\/ */
10105: /* /\* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv *\/ */
10106: /* /\* printf(" V%d=%d ",Tvaraff[k],vlv); *\/ */
10107: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10108: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10109: /* } */
10110: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10111: /* /\* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); *\/ */
10112: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
10113: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.264 brouard 10114: }
10115: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 10116: /* printf("\n#\n"); */
1.238 brouard 10117: fprintf(ficgp,"\n#\n");
10118: if(invalidvarcomb[k1]){
1.260 brouard 10119: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 10120: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10121: continue;
10122: }
1.235 brouard 10123:
1.241 brouard 10124: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
10125: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 10126: /* 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 10127: fprintf(ficgp,"set title \"Alive state %d %s model=1+age+%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 10128: 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_"),nres-1,nres-1,nres);
10129: /* 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); */
10130: /* k1-1 error should be nres-1*/
1.238 brouard 10131: for (i=1; i<= nlstate ; i ++) {
10132: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10133: else fprintf(ficgp," %%*lf (%%*lf)");
10134: }
1.288 brouard 10135: 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 10136: for (i=1; i<= nlstate ; i ++) {
10137: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10138: else fprintf(ficgp," %%*lf (%%*lf)");
10139: }
1.260 brouard 10140: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 10141: for (i=1; i<= nlstate ; i ++) {
10142: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10143: else fprintf(ficgp," %%*lf (%%*lf)");
10144: }
1.265 brouard 10145: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
10146:
10147: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
10148: if(cptcoveff ==0){
1.271 brouard 10149: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 10150: }else{
10151: kl=0;
10152: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 10153: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
10154: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.265 brouard 10155: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
10156: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
10157: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
10158: vlv= nbcode[Tvaraff[k]][lv];
10159: kl++;
10160: /* 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 *\/ */
10161: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
10162: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
10163: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
10164: if(k==cptcoveff){
10165: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
10166: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
10167: }else{
10168: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
10169: kl++;
10170: }
10171: } /* end covariate */
10172: } /* end if no covariate */
10173:
1.296 brouard 10174: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 10175: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 10176: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 10177: if(cptcoveff ==0){
1.245 brouard 10178: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 10179: }else{
10180: kl=0;
10181: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 10182: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
10183: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 10184: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
10185: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
10186: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 10187: /* vlv= nbcode[Tvaraff[k]][lv]; */
10188: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 10189: kl++;
1.238 brouard 10190: /* 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 *\/ */
10191: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
10192: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
10193: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
10194: if(k==cptcoveff){
1.245 brouard 10195: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 10196: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 10197: }else{
1.332 brouard 10198: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]);
1.238 brouard 10199: kl++;
10200: }
10201: } /* end covariate */
10202: } /* end if no covariate */
1.296 brouard 10203: if(prevbcast == 1){
1.268 brouard 10204: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
10205: /* k1-1 error should be nres-1*/
10206: for (i=1; i<= nlstate ; i ++) {
10207: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10208: else fprintf(ficgp," %%*lf (%%*lf)");
10209: }
1.271 brouard 10210: 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 10211: for (i=1; i<= nlstate ; i ++) {
10212: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10213: else fprintf(ficgp," %%*lf (%%*lf)");
10214: }
1.276 brouard 10215: 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 10216: for (i=1; i<= nlstate ; i ++) {
10217: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
10218: else fprintf(ficgp," %%*lf (%%*lf)");
10219: }
1.274 brouard 10220: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 10221: } /* end if backprojcast */
1.296 brouard 10222: } /* end if prevbcast */
1.276 brouard 10223: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
10224: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 10225: } /* nres */
1.337 brouard 10226: /* } /\* k1 *\/ */
1.201 brouard 10227: } /* cpt */
1.235 brouard 10228:
10229:
1.126 brouard 10230: /*2 eme*/
1.337 brouard 10231: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 10232: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10233: k1=TKresult[nres];
1.338 brouard 10234: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10235: /* if(m != 1 && TKresult[nres]!= k1) */
10236: /* continue; */
1.238 brouard 10237: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 10238: strcpy(gplotlabel,"(");
1.337 brouard 10239: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10240: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10241: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10242: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10243: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10244: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10245: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10246: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10247: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10248: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10249: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10250: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10251: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10252: /* } */
10253: /* /\* for(k=1; k <= ncovds; k++){ *\/ */
10254: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10255: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
10256: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
10257: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 10258: }
1.264 brouard 10259: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 10260: fprintf(ficgp,"\n#\n");
1.223 brouard 10261: if(invalidvarcomb[k1]){
10262: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10263: continue;
10264: }
1.219 brouard 10265:
1.241 brouard 10266: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 10267: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 10268: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
10269: if(vpopbased==0){
1.360 ! brouard 10270: 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 10271: }else
1.238 brouard 10272: fprintf(ficgp,"\nreplot ");
1.360 ! brouard 10273: 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 10274: k=2*i;
1.360 ! brouard 10275: 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 */
! 10276: for (j=1; j<= nlstate+1 ; j ++) { /* e.. e.1 e.2 again j-1 is the state of end, wlim_i eij*/
! 10277: 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 */
! 10278: else fprintf(ficgp," %%*lf (%%*lf)"); /* skipping that field with a star */
1.238 brouard 10279: }
10280: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
1.360 ! brouard 10281: 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 10282: 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 10283: for (j=1; j<= nlstate+1 ; j ++) {
10284: if (j==i) fprintf(ficgp," %%lf (%%lf)");
10285: else fprintf(ficgp," %%*lf (%%*lf)");
10286: }
10287: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 10288: 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 10289: for (j=1; j<= nlstate+1 ; j ++) {
10290: if (j==i) fprintf(ficgp," %%lf (%%lf)");
10291: else fprintf(ficgp," %%*lf (%%*lf)");
10292: }
1.360 ! brouard 10293: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0,\\\n"); /* ,\\\n added for th percentage graphs */
1.238 brouard 10294: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
10295: } /* state */
1.360 ! brouard 10296: /* again for the percentag spent in state i-1=1 to i-1=nlstate */
! 10297: for (i=2; i<= nlstate+1 ; i ++) { /* For state i-1=0 is LE, while i-1=1 to nlstate are origin state */
! 10298: k=2*i;
! 10299: 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 */
! 10300: for (j=1; j<= nlstate ; j ++)
! 10301: fprintf(ficgp," %%*lf (%%*lf)"); /* Skipping TLE and LE to read %LE only */
! 10302: for (j=1; j<= nlstate+1 ; j ++) { /* e.. e.1 e.2 again j-1 is the state of end, wlim_i eij*/
! 10303: 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 */
! 10304: else fprintf(ficgp," %%*lf (%%*lf)"); /* skipping that field with a star */
! 10305: }
! 10306: if (i== 1) fprintf(ficgp,"\" t\"%%TLE\" w l lt %d axis x1y2, \\\n",i); /* Not used */
! 10307: 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 */
! 10308: 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);
! 10309: for (j=1; j<= nlstate ; j ++)
! 10310: fprintf(ficgp," %%*lf (%%*lf)"); /* Skipping TLE and LE to read %LE only */
! 10311: for (j=1; j<= nlstate+1 ; j ++) {
! 10312: if (j==i) fprintf(ficgp," %%lf (%%lf)");
! 10313: else fprintf(ficgp," %%*lf (%%*lf)");
! 10314: }
! 10315: fprintf(ficgp,"\" t\"\" w l lt 0 axis x1y2,");
! 10316: 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);
! 10317: for (j=1; j<= nlstate ; j ++)
! 10318: fprintf(ficgp," %%*lf (%%*lf)"); /* Skipping TLE and LE to read %LE only */
! 10319: for (j=1; j<= nlstate+1 ; j ++) {
! 10320: if (j==i) fprintf(ficgp," %%lf (%%lf)");
! 10321: else fprintf(ficgp," %%*lf (%%*lf)");
! 10322: }
! 10323: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0 axis x1y2");
! 10324: else fprintf(ficgp,"\" t\"\" w l lt 0 axis x1y2,\\\n");
! 10325: } /* state for percent */
1.238 brouard 10326: } /* vpopbased */
1.264 brouard 10327: 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 10328: } /* end nres */
1.337 brouard 10329: /* } /\* k1 end 2 eme*\/ */
1.238 brouard 10330:
10331:
10332: /*3eme*/
1.337 brouard 10333: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 10334: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10335: k1=TKresult[nres];
1.338 brouard 10336: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10337: /* if(m != 1 && TKresult[nres]!= k1) */
10338: /* continue; */
1.238 brouard 10339:
1.332 brouard 10340: for (cpt=1; cpt<= nlstate ; cpt ++) { /* Fragile no verification of covariate values */
1.261 brouard 10341: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 10342: strcpy(gplotlabel,"(");
1.337 brouard 10343: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10344: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10345: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10346: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10347: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10348: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
10349: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10350: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10351: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10352: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10353: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10354: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10355: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10356: /* } */
10357: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10358: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
10359: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
10360: }
1.264 brouard 10361: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 10362: fprintf(ficgp,"\n#\n");
10363: if(invalidvarcomb[k1]){
10364: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10365: continue;
10366: }
10367:
10368: /* k=2+nlstate*(2*cpt-2); */
10369: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 10370: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 10371: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 10372: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 10373: 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 10374: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
10375: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
10376: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
10377: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
10378: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
10379: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 10380:
1.238 brouard 10381: */
10382: for (i=1; i< nlstate ; i ++) {
1.261 brouard 10383: 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 10384: /* 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 10385:
1.238 brouard 10386: }
1.261 brouard 10387: 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 10388: }
1.264 brouard 10389: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 10390: } /* end nres */
1.337 brouard 10391: /* } /\* end kl 3eme *\/ */
1.126 brouard 10392:
1.223 brouard 10393: /* 4eme */
1.201 brouard 10394: /* Survival functions (period) from state i in state j by initial state i */
1.337 brouard 10395: /* for (k1=1; k1<=m; k1++){ /\* For each covariate and each value *\/ */
1.238 brouard 10396: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10397: k1=TKresult[nres];
1.338 brouard 10398: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10399: /* if(m != 1 && TKresult[nres]!= k1) */
10400: /* continue; */
1.238 brouard 10401: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 10402: strcpy(gplotlabel,"(");
1.337 brouard 10403: fprintf(ficgp,"\n#\n#\n# Survival functions in state %d : 'LIJ_' files, cov=%d state=%d", cpt, k1, cpt);
10404: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10405: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10406: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10407: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10408: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10409: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10410: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10411: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10412: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10413: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10414: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10415: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10416: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10417: /* } */
10418: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10419: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10420: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 10421: }
1.264 brouard 10422: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 10423: fprintf(ficgp,"\n#\n");
10424: if(invalidvarcomb[k1]){
10425: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10426: continue;
1.223 brouard 10427: }
1.238 brouard 10428:
1.241 brouard 10429: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 10430: 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 10431: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
10432: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
10433: k=3;
10434: for (i=1; i<= nlstate ; i ++){
10435: if(i==1){
10436: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
10437: }else{
10438: fprintf(ficgp,", '' ");
10439: }
10440: l=(nlstate+ndeath)*(i-1)+1;
10441: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
10442: for (j=2; j<= nlstate+ndeath ; j ++)
10443: fprintf(ficgp,"+$%d",k+l+j-1);
10444: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
10445: } /* nlstate */
1.264 brouard 10446: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 10447: } /* end cpt state*/
10448: } /* end nres */
1.337 brouard 10449: /* } /\* end covariate k1 *\/ */
1.238 brouard 10450:
1.220 brouard 10451: /* 5eme */
1.201 brouard 10452: /* Survival functions (period) from state i in state j by final state j */
1.337 brouard 10453: /* for (k1=1; k1<= m ; k1++){ /\* For each covariate combination if any *\/ */
1.238 brouard 10454: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10455: k1=TKresult[nres];
1.338 brouard 10456: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10457: /* if(m != 1 && TKresult[nres]!= k1) */
10458: /* continue; */
1.238 brouard 10459: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 10460: strcpy(gplotlabel,"(");
1.238 brouard 10461: 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 10462: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10463: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10464: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10465: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10466: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10467: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10468: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10469: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10470: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10471: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10472: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10473: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10474: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10475: /* } */
10476: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10477: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10478: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 10479: }
1.264 brouard 10480: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 10481: fprintf(ficgp,"\n#\n");
10482: if(invalidvarcomb[k1]){
10483: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10484: continue;
10485: }
1.227 brouard 10486:
1.241 brouard 10487: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 10488: 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 10489: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
10490: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
10491: k=3;
10492: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
10493: if(j==1)
10494: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
10495: else
10496: fprintf(ficgp,", '' ");
10497: l=(nlstate+ndeath)*(cpt-1) +j;
10498: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
10499: /* for (i=2; i<= nlstate+ndeath ; i ++) */
10500: /* fprintf(ficgp,"+$%d",k+l+i-1); */
10501: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
10502: } /* nlstate */
10503: fprintf(ficgp,", '' ");
10504: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
10505: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
10506: l=(nlstate+ndeath)*(cpt-1) +j;
10507: if(j < nlstate)
10508: fprintf(ficgp,"$%d +",k+l);
10509: else
10510: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
10511: }
1.264 brouard 10512: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 10513: } /* end cpt state*/
1.337 brouard 10514: /* } /\* end covariate *\/ */
1.238 brouard 10515: } /* end nres */
1.227 brouard 10516:
1.220 brouard 10517: /* 6eme */
1.202 brouard 10518: /* CV preval stable (period) for each covariate */
1.337 brouard 10519: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 10520: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10521: k1=TKresult[nres];
1.338 brouard 10522: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10523: /* if(m != 1 && TKresult[nres]!= k1) */
10524: /* continue; */
1.255 brouard 10525: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 10526: strcpy(gplotlabel,"(");
1.288 brouard 10527: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 10528: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10529: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10530: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10531: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10532: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10533: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10534: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10535: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10536: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10537: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10538: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10539: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10540: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10541: /* } */
10542: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10543: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10544: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 10545: }
1.264 brouard 10546: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 10547: fprintf(ficgp,"\n#\n");
1.223 brouard 10548: if(invalidvarcomb[k1]){
1.227 brouard 10549: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10550: continue;
1.223 brouard 10551: }
1.227 brouard 10552:
1.241 brouard 10553: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 10554: 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 10555: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 10556: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 10557: k=3; /* Offset */
1.255 brouard 10558: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 10559: if(i==1)
10560: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
10561: else
10562: fprintf(ficgp,", '' ");
1.255 brouard 10563: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 10564: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
10565: for (j=2; j<= nlstate ; j ++)
10566: fprintf(ficgp,"+$%d",k+l+j-1);
10567: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 10568: } /* nlstate */
1.264 brouard 10569: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 10570: } /* end cpt state*/
10571: } /* end covariate */
1.227 brouard 10572:
10573:
1.220 brouard 10574: /* 7eme */
1.296 brouard 10575: if(prevbcast == 1){
1.288 brouard 10576: /* CV backward prevalence for each covariate */
1.337 brouard 10577: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 10578: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10579: k1=TKresult[nres];
1.338 brouard 10580: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10581: /* if(m != 1 && TKresult[nres]!= k1) */
10582: /* continue; */
1.268 brouard 10583: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 10584: strcpy(gplotlabel,"(");
1.288 brouard 10585: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 10586: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10587: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10588: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10589: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
10590: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
10591: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10592: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10593: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10594: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10595: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10596: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10597: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10598: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10599: /* } */
10600: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10601: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10602: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 10603: }
1.264 brouard 10604: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 10605: fprintf(ficgp,"\n#\n");
10606: if(invalidvarcomb[k1]){
10607: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10608: continue;
10609: }
10610:
1.241 brouard 10611: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 10612: 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 10613: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 10614: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 10615: k=3; /* Offset */
1.268 brouard 10616: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 10617: if(i==1)
10618: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
10619: else
10620: fprintf(ficgp,", '' ");
10621: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 10622: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 10623: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
10624: /* 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 10625: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 10626: /* for (j=2; j<= nlstate ; j ++) */
10627: /* fprintf(ficgp,"+$%d",k+l+j-1); */
10628: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 10629: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 10630: } /* nlstate */
1.264 brouard 10631: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 10632: } /* end cpt state*/
10633: } /* end covariate */
1.296 brouard 10634: } /* End if prevbcast */
1.218 brouard 10635:
1.223 brouard 10636: /* 8eme */
1.218 brouard 10637: if(prevfcast==1){
1.288 brouard 10638: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 10639:
1.337 brouard 10640: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 10641: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10642: k1=TKresult[nres];
1.338 brouard 10643: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10644: /* if(m != 1 && TKresult[nres]!= k1) */
10645: /* continue; */
1.211 brouard 10646: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 10647: strcpy(gplotlabel,"(");
1.288 brouard 10648: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 10649: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10650: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10651: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10652: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
10653: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
10654: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10655: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10656: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10657: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10658: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10659: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10660: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10661: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10662: /* } */
10663: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10664: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10665: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 10666: }
1.264 brouard 10667: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 10668: fprintf(ficgp,"\n#\n");
10669: if(invalidvarcomb[k1]){
10670: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10671: continue;
10672: }
10673:
10674: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 10675: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 10676: 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 10677: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 10678: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 10679:
10680: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
10681: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
10682: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
10683: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 10684: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10685: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
10686: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10687: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 10688: if(i==istart){
1.227 brouard 10689: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
10690: }else{
10691: fprintf(ficgp,",\\\n '' ");
10692: }
10693: if(cptcoveff ==0){ /* No covariate */
10694: ioffset=2; /* Age is in 2 */
10695: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
10696: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
10697: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
10698: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
10699: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 10700: if(i==nlstate+1){
1.270 brouard 10701: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 10702: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
10703: fprintf(ficgp,",\\\n '' ");
10704: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 10705: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 10706: offyear, \
1.268 brouard 10707: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 10708: }else
1.227 brouard 10709: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
10710: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
10711: }else{ /* more than 2 covariates */
1.270 brouard 10712: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
10713: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10714: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
10715: iyearc=ioffset-1;
10716: iagec=ioffset;
1.227 brouard 10717: fprintf(ficgp," u %d:(",ioffset);
10718: kl=0;
10719: strcpy(gplotcondition,"(");
1.351 brouard 10720: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate writing the chain of conditions *\/ */
1.332 brouard 10721: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
1.351 brouard 10722: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10723: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
10724: lv=Tvresult[nres][k];
10725: vlv=TinvDoQresult[nres][Tvresult[nres][k]];
1.227 brouard 10726: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
10727: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
10728: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 10729: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
1.351 brouard 10730: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
1.227 brouard 10731: kl++;
1.351 brouard 10732: /* sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]); */
10733: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,lv, kl+1, vlv );
1.227 brouard 10734: kl++;
1.351 brouard 10735: if(k <cptcovs && cptcovs>1)
1.227 brouard 10736: sprintf(gplotcondition+strlen(gplotcondition)," && ");
10737: }
10738: strcpy(gplotcondition+strlen(gplotcondition),")");
10739: /* 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 *\/ */
10740: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
10741: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
10742: /* '' 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*/
10743: if(i==nlstate+1){
1.270 brouard 10744: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
10745: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 10746: fprintf(ficgp,",\\\n '' ");
1.270 brouard 10747: fprintf(ficgp," u %d:(",iagec);
10748: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
10749: iyearc, iagec, offyear, \
10750: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 10751: /* '' 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 10752: }else{
10753: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
10754: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
10755: }
10756: } /* end if covariate */
10757: } /* nlstate */
1.264 brouard 10758: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 10759: } /* end cpt state*/
10760: } /* end covariate */
10761: } /* End if prevfcast */
1.227 brouard 10762:
1.296 brouard 10763: if(prevbcast==1){
1.268 brouard 10764: /* Back projection from cross-sectional to stable (mixed) for each covariate */
10765:
1.337 brouard 10766: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.268 brouard 10767: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10768: k1=TKresult[nres];
1.338 brouard 10769: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10770: /* if(m != 1 && TKresult[nres]!= k1) */
10771: /* continue; */
1.268 brouard 10772: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
10773: strcpy(gplotlabel,"(");
10774: 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 10775: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
10776: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10777: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10778: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
10779: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
10780: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
10781: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10782: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10783: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10784: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10785: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10786: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10787: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10788: /* } */
10789: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10790: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10791: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.268 brouard 10792: }
10793: strcpy(gplotlabel+strlen(gplotlabel),")");
10794: fprintf(ficgp,"\n#\n");
10795: if(invalidvarcomb[k1]){
10796: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
10797: continue;
10798: }
10799:
10800: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
10801: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
10802: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
10803: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
10804: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
10805:
10806: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
10807: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
10808: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
10809: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
10810: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10811: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
10812: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10813: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
10814: if(i==istart){
10815: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
10816: }else{
10817: fprintf(ficgp,",\\\n '' ");
10818: }
1.351 brouard 10819: /* if(cptcoveff ==0){ /\* No covariate *\/ */
10820: if(cptcovs ==0){ /* No covariate */
1.268 brouard 10821: ioffset=2; /* Age is in 2 */
10822: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
10823: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
10824: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
10825: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
10826: fprintf(ficgp," u %d:(", ioffset);
10827: if(i==nlstate+1){
1.270 brouard 10828: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 10829: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
10830: fprintf(ficgp,",\\\n '' ");
10831: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 10832: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 10833: offbyear, \
10834: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
10835: }else
10836: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
10837: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
10838: }else{ /* more than 2 covariates */
1.270 brouard 10839: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
10840: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
10841: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
10842: iyearc=ioffset-1;
10843: iagec=ioffset;
1.268 brouard 10844: fprintf(ficgp," u %d:(",ioffset);
10845: kl=0;
10846: strcpy(gplotcondition,"(");
1.337 brouard 10847: for (k=1; k<=cptcovs; k++){ /* For each covariate k of the resultline, get corresponding value lv for combination k1 */
1.338 brouard 10848: if(Dummy[modelresult[nres][k]]==0){ /* To be verified */
1.337 brouard 10849: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate writing the chain of conditions *\/ */
10850: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
10851: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
10852: lv=Tvresult[nres][k];
10853: vlv=TinvDoQresult[nres][Tvresult[nres][k]];
10854: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
10855: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
10856: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
10857: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
10858: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10859: kl++;
10860: /* sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]); */
10861: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%lg " ,kl,Tvresult[nres][k], kl+1,TinvDoQresult[nres][Tvresult[nres][k]]);
10862: kl++;
1.338 brouard 10863: if(k <cptcovs && cptcovs>1)
1.337 brouard 10864: sprintf(gplotcondition+strlen(gplotcondition)," && ");
10865: }
1.268 brouard 10866: }
10867: strcpy(gplotcondition+strlen(gplotcondition),")");
10868: /* 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 *\/ */
10869: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
10870: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
10871: /* '' 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*/
10872: if(i==nlstate+1){
1.270 brouard 10873: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
10874: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 10875: fprintf(ficgp,",\\\n '' ");
1.270 brouard 10876: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 10877: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 10878: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
10879: iyearc,iagec,offbyear, \
10880: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 10881: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
10882: }else{
10883: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
10884: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
10885: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
10886: }
10887: } /* end if covariate */
10888: } /* nlstate */
10889: fprintf(ficgp,"\nset out; unset label;\n");
10890: } /* end cpt state*/
10891: } /* end covariate */
1.296 brouard 10892: } /* End if prevbcast */
1.268 brouard 10893:
1.227 brouard 10894:
1.238 brouard 10895: /* 9eme writing MLE parameters */
10896: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 10897: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 10898: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 10899: for(k=1; k <=(nlstate+ndeath); k++){
10900: if (k != i) {
1.227 brouard 10901: fprintf(ficgp,"# current state %d\n",k);
10902: for(j=1; j <=ncovmodel; j++){
10903: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
10904: jk++;
10905: }
10906: fprintf(ficgp,"\n");
1.126 brouard 10907: }
10908: }
1.223 brouard 10909: }
1.187 brouard 10910: fprintf(ficgp,"##############\n#\n");
1.227 brouard 10911:
1.145 brouard 10912: /*goto avoid;*/
1.238 brouard 10913: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
10914: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 10915: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
10916: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
10917: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
10918: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
10919: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
10920: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
10921: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
10922: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
10923: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
10924: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
10925: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
10926: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
10927: fprintf(ficgp,"#\n");
1.223 brouard 10928: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 10929: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.338 brouard 10930: fprintf(ficgp,"#model=1+age+%s \n",model);
1.238 brouard 10931: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.351 brouard 10932: /* fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/\* to be checked *\/ */
10933: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcovs,m);/* to be checked */
1.337 brouard 10934: /* for(k1=1; k1 <=m; k1++) /\* For each combination of covariate *\/ */
1.237 brouard 10935: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 10936: /* k1=nres; */
1.338 brouard 10937: k1=TKresult[nres];
10938: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 10939: fprintf(ficgp,"\n\n# Resultline k1=%d ",k1);
1.264 brouard 10940: strcpy(gplotlabel,"(");
1.276 brouard 10941: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.337 brouard 10942: for (k=1; k<=cptcovs; k++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
10943: /* for each resultline nres, and position k, Tvresult[nres][k] gives the name of the variable and
10944: TinvDoQresult[nres][Tvresult[nres][k]] gives its value double or integer) */
10945: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10946: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
10947: }
10948: /* if(m != 1 && TKresult[nres]!= k1) */
10949: /* continue; */
10950: /* fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1); */
10951: /* strcpy(gplotlabel,"("); */
10952: /* /\*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*\/ */
10953: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
10954: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
10955: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
10956: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
10957: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
10958: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
10959: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
10960: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
10961: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
10962: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
10963: /* } */
10964: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
10965: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10966: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
10967: /* } */
1.264 brouard 10968: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 10969: fprintf(ficgp,"\n#\n");
1.264 brouard 10970: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 10971: fprintf(ficgp,"\nset key outside ");
10972: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
10973: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 10974: fprintf(ficgp,"\nset ter svg size 640, 480 ");
10975: if (ng==1){
10976: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
10977: fprintf(ficgp,"\nunset log y");
10978: }else if (ng==2){
10979: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
10980: fprintf(ficgp,"\nset log y");
10981: }else if (ng==3){
10982: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
10983: fprintf(ficgp,"\nset log y");
10984: }else
10985: fprintf(ficgp,"\nunset title ");
10986: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
10987: i=1;
10988: for(k2=1; k2<=nlstate; k2++) {
10989: k3=i;
10990: for(k=1; k<=(nlstate+ndeath); k++) {
10991: if (k != k2){
10992: switch( ng) {
10993: case 1:
10994: if(nagesqr==0)
10995: fprintf(ficgp," p%d+p%d*x",i,i+1);
10996: else /* nagesqr =1 */
10997: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
10998: break;
10999: case 2: /* ng=2 */
11000: if(nagesqr==0)
11001: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
11002: else /* nagesqr =1 */
11003: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
11004: break;
11005: case 3:
11006: if(nagesqr==0)
11007: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
11008: else /* nagesqr =1 */
11009: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
11010: break;
11011: }
11012: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 11013: ijp=1; /* product no age */
11014: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
11015: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 11016: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 11017: switch(Typevar[j]){
11018: case 1:
11019: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
11020: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
11021: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
11022: if(DummyV[j]==0){/* Bug valgrind */
11023: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
11024: }else{ /* quantitative */
11025: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
11026: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
11027: }
11028: ij++;
1.268 brouard 11029: }
1.237 brouard 11030: }
1.329 brouard 11031: }
11032: break;
11033: case 2:
11034: if(cptcovprod >0){
11035: if(j==Tprod[ijp]) { /* */
11036: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
11037: if(ijp <=cptcovprod) { /* Product */
11038: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
11039: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
11040: /* 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)]); */
11041: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
11042: }else{ /* Vn is dummy and Vm is quanti */
11043: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
11044: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
11045: }
11046: }else{ /* Vn*Vm Vn is quanti */
11047: if(DummyV[Tvard[ijp][2]]==0){
11048: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
11049: }else{ /* Both quanti */
11050: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
11051: }
1.268 brouard 11052: }
1.329 brouard 11053: ijp++;
1.237 brouard 11054: }
1.329 brouard 11055: } /* end Tprod */
11056: }
11057: break;
1.349 brouard 11058: case 3:
11059: if(cptcovdageprod >0){
11060: /* if(j==Tprod[ijp]) { */ /* not necessary */
11061: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
1.350 brouard 11062: if(ijp <=cptcovprod) { /* Product Vn*Vm and age*VN*Vm*/
11063: if(DummyV[Tvardk[ijp][1]]==0){/* Vn is dummy */
11064: if(DummyV[Tvardk[ijp][2]]==0){/* Vn and Vm are dummy */
1.349 brouard 11065: /* 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)]); */
11066: fprintf(ficgp,"+p%d*%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
11067: }else{ /* Vn is dummy and Vm is quanti */
11068: /* 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 11069: 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 11070: }
1.350 brouard 11071: }else{ /* age* Vn*Vm Vn is quanti HERE */
1.349 brouard 11072: if(DummyV[Tvard[ijp][2]]==0){
1.350 brouard 11073: 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 11074: }else{ /* Both quanti */
1.350 brouard 11075: 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 11076: }
11077: }
11078: ijp++;
11079: }
11080: /* } */ /* end Tprod */
11081: }
11082: break;
1.329 brouard 11083: case 0:
11084: /* simple covariate */
1.264 brouard 11085: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 11086: if(Dummy[j]==0){
11087: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
11088: }else{ /* quantitative */
11089: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 11090: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 11091: }
1.329 brouard 11092: /* end simple */
11093: break;
11094: default:
11095: break;
11096: } /* end switch */
1.237 brouard 11097: } /* end j */
1.329 brouard 11098: }else{ /* k=k2 */
11099: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
11100: fprintf(ficgp," (1.");i=i-ncovmodel;
11101: }else
11102: i=i-ncovmodel;
1.223 brouard 11103: }
1.227 brouard 11104:
1.223 brouard 11105: if(ng != 1){
11106: fprintf(ficgp,")/(1");
1.227 brouard 11107:
1.264 brouard 11108: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 11109: if(nagesqr==0)
1.264 brouard 11110: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 11111: else /* nagesqr =1 */
1.264 brouard 11112: 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 11113:
1.223 brouard 11114: ij=1;
1.329 brouard 11115: ijp=1;
11116: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
11117: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
11118: switch(Typevar[j]){
11119: case 1:
11120: if(cptcovage >0){
11121: if(j==Tage[ij]) { /* Bug valgrind */
11122: if(ij <=cptcovage) { /* Bug valgrind */
11123: if(DummyV[j]==0){/* Bug valgrind */
11124: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
11125: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
11126: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
11127: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
11128: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
11129: }else{ /* quantitative */
11130: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
11131: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
11132: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
11133: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
11134: }
11135: ij++;
11136: }
11137: }
11138: }
11139: break;
11140: case 2:
11141: if(cptcovprod >0){
11142: if(j==Tprod[ijp]) { /* */
11143: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
11144: if(ijp <=cptcovprod) { /* Product */
11145: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
11146: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
11147: /* 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)]); */
11148: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
11149: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
11150: }else{ /* Vn is dummy and Vm is quanti */
11151: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
11152: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
11153: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
11154: }
11155: }else{ /* Vn*Vm Vn is quanti */
11156: if(DummyV[Tvard[ijp][2]]==0){
11157: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
11158: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
11159: }else{ /* Both quanti */
11160: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
11161: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
11162: }
11163: }
11164: ijp++;
11165: }
11166: } /* end Tprod */
11167: } /* end if */
11168: break;
1.349 brouard 11169: case 3:
11170: if(cptcovdageprod >0){
11171: /* if(j==Tprod[ijp]) { /\* *\/ */
11172: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
11173: if(ijp <=cptcovprod) { /* Product */
1.350 brouard 11174: if(DummyV[Tvardk[ijp][1]]==0){/* Vn is dummy */
11175: if(DummyV[Tvardk[ijp][2]]==0){/* Vn and Vm are dummy */
1.349 brouard 11176: /* 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 11177: 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 11178: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
11179: }else{ /* Vn is dummy and Vm is quanti */
11180: /* 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 11181: 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 11182: /* fprintf(ficgp,"+p%d*%d*%f*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
11183: }
11184: }else{ /* Vn*Vm Vn is quanti */
1.350 brouard 11185: if(DummyV[Tvardk[ijp][2]]==0){
11186: 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 11187: /* fprintf(ficgp,"+p%d*%d*%f*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
11188: }else{ /* Both quanti */
1.350 brouard 11189: 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 11190: /* fprintf(ficgp,"+p%d*%f*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
11191: }
11192: }
11193: ijp++;
11194: }
11195: /* } /\* end Tprod *\/ */
11196: } /* end if */
11197: break;
1.329 brouard 11198: case 0:
11199: /* simple covariate */
11200: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
11201: if(Dummy[j]==0){
11202: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
11203: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
11204: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
11205: }else{ /* quantitative */
11206: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
11207: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
11208: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
11209: }
11210: /* end simple */
11211: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
11212: break;
11213: default:
11214: break;
11215: } /* end switch */
1.223 brouard 11216: }
11217: fprintf(ficgp,")");
11218: }
11219: fprintf(ficgp,")");
11220: if(ng ==2)
1.276 brouard 11221: 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 11222: else /* ng= 3 */
1.276 brouard 11223: 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 11224: }else{ /* end ng <> 1 */
1.223 brouard 11225: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 11226: 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 11227: }
11228: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
11229: fprintf(ficgp,",");
11230: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
11231: fprintf(ficgp,",");
11232: i=i+ncovmodel;
11233: } /* end k */
11234: } /* end k2 */
1.276 brouard 11235: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
11236: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.337 brouard 11237: } /* end resultline */
1.223 brouard 11238: } /* end ng */
11239: /* avoid: */
11240: fflush(ficgp);
1.126 brouard 11241: } /* end gnuplot */
11242:
11243:
11244: /*************** Moving average **************/
1.219 brouard 11245: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 11246: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 11247:
1.222 brouard 11248: int i, cpt, cptcod;
11249: int modcovmax =1;
11250: int mobilavrange, mob;
11251: int iage=0;
1.288 brouard 11252: int firstA1=0, firstA2=0;
1.222 brouard 11253:
1.266 brouard 11254: double sum=0., sumr=0.;
1.222 brouard 11255: double age;
1.266 brouard 11256: double *sumnewp, *sumnewm, *sumnewmr;
11257: double *agemingood, *agemaxgood;
11258: double *agemingoodr, *agemaxgoodr;
1.222 brouard 11259:
11260:
1.278 brouard 11261: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
11262: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 11263:
11264: sumnewp = vector(1,ncovcombmax);
11265: sumnewm = vector(1,ncovcombmax);
1.266 brouard 11266: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 11267: agemingood = vector(1,ncovcombmax);
1.266 brouard 11268: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 11269: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 11270: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 11271:
11272: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 11273: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 11274: sumnewp[cptcod]=0.;
1.266 brouard 11275: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
11276: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 11277: }
11278: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
11279:
1.266 brouard 11280: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
11281: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 11282: else mobilavrange=mobilav;
11283: for (age=bage; age<=fage; age++)
11284: for (i=1; i<=nlstate;i++)
11285: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
11286: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
11287: /* We keep the original values on the extreme ages bage, fage and for
11288: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
11289: we use a 5 terms etc. until the borders are no more concerned.
11290: */
11291: for (mob=3;mob <=mobilavrange;mob=mob+2){
11292: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 11293: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
11294: sumnewm[cptcod]=0.;
11295: for (i=1; i<=nlstate;i++){
1.222 brouard 11296: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
11297: for (cpt=1;cpt<=(mob-1)/2;cpt++){
11298: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
11299: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
11300: }
11301: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 11302: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
11303: } /* end i */
11304: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
11305: } /* end cptcod */
1.222 brouard 11306: }/* end age */
11307: }/* end mob */
1.266 brouard 11308: }else{
11309: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 11310: return -1;
1.266 brouard 11311: }
11312:
11313: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 11314: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
11315: if(invalidvarcomb[cptcod]){
11316: printf("\nCombination (%d) ignored because no cases \n",cptcod);
11317: continue;
11318: }
1.219 brouard 11319:
1.266 brouard 11320: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
11321: sumnewm[cptcod]=0.;
11322: sumnewmr[cptcod]=0.;
11323: for (i=1; i<=nlstate;i++){
11324: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
11325: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
11326: }
11327: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
11328: agemingoodr[cptcod]=age;
11329: }
11330: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
11331: agemingood[cptcod]=age;
11332: }
11333: } /* age */
11334: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 11335: sumnewm[cptcod]=0.;
1.266 brouard 11336: sumnewmr[cptcod]=0.;
1.222 brouard 11337: for (i=1; i<=nlstate;i++){
11338: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 11339: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
11340: }
11341: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
11342: agemaxgoodr[cptcod]=age;
1.222 brouard 11343: }
11344: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 11345: agemaxgood[cptcod]=age;
11346: }
11347: } /* age */
11348: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
11349: /* but they will change */
1.288 brouard 11350: firstA1=0;firstA2=0;
1.266 brouard 11351: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
11352: sumnewm[cptcod]=0.;
11353: sumnewmr[cptcod]=0.;
11354: for (i=1; i<=nlstate;i++){
11355: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
11356: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
11357: }
11358: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
11359: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
11360: agemaxgoodr[cptcod]=age; /* age min */
11361: for (i=1; i<=nlstate;i++)
11362: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
11363: }else{ /* bad we change the value with the values of good ages */
11364: for (i=1; i<=nlstate;i++){
11365: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
11366: } /* i */
11367: } /* end bad */
11368: }else{
11369: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
11370: agemaxgood[cptcod]=age;
11371: }else{ /* bad we change the value with the values of good ages */
11372: for (i=1; i<=nlstate;i++){
11373: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
11374: } /* i */
11375: } /* end bad */
11376: }/* end else */
11377: sum=0.;sumr=0.;
11378: for (i=1; i<=nlstate;i++){
11379: sum+=mobaverage[(int)age][i][cptcod];
11380: sumr+=probs[(int)age][i][cptcod];
11381: }
11382: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 11383: if(!firstA1){
11384: firstA1=1;
11385: 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);
11386: }
11387: 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 11388: } /* end bad */
11389: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
11390: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 11391: if(!firstA2){
11392: firstA2=1;
11393: 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);
11394: }
11395: 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 11396: } /* end bad */
11397: }/* age */
1.266 brouard 11398:
11399: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 11400: sumnewm[cptcod]=0.;
1.266 brouard 11401: sumnewmr[cptcod]=0.;
1.222 brouard 11402: for (i=1; i<=nlstate;i++){
11403: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 11404: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
11405: }
11406: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
11407: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
11408: agemingoodr[cptcod]=age;
11409: for (i=1; i<=nlstate;i++)
11410: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
11411: }else{ /* bad we change the value with the values of good ages */
11412: for (i=1; i<=nlstate;i++){
11413: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
11414: } /* i */
11415: } /* end bad */
11416: }else{
11417: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
11418: agemingood[cptcod]=age;
11419: }else{ /* bad */
11420: for (i=1; i<=nlstate;i++){
11421: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
11422: } /* i */
11423: } /* end bad */
11424: }/* end else */
11425: sum=0.;sumr=0.;
11426: for (i=1; i<=nlstate;i++){
11427: sum+=mobaverage[(int)age][i][cptcod];
11428: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 11429: }
1.266 brouard 11430: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 11431: 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 11432: } /* end bad */
11433: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
11434: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 11435: 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 11436: } /* end bad */
11437: }/* age */
1.266 brouard 11438:
1.222 brouard 11439:
11440: for (age=bage; age<=fage; age++){
1.235 brouard 11441: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 11442: sumnewp[cptcod]=0.;
11443: sumnewm[cptcod]=0.;
11444: for (i=1; i<=nlstate;i++){
11445: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
11446: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
11447: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
11448: }
11449: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
11450: }
11451: /* printf("\n"); */
11452: /* } */
1.266 brouard 11453:
1.222 brouard 11454: /* brutal averaging */
1.266 brouard 11455: /* for (i=1; i<=nlstate;i++){ */
11456: /* for (age=1; age<=bage; age++){ */
11457: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
11458: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
11459: /* } */
11460: /* for (age=fage; age<=AGESUP; age++){ */
11461: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
11462: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
11463: /* } */
11464: /* } /\* end i status *\/ */
11465: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
11466: /* for (age=1; age<=AGESUP; age++){ */
11467: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
11468: /* mobaverage[(int)age][i][cptcod]=0.; */
11469: /* } */
11470: /* } */
1.222 brouard 11471: }/* end cptcod */
1.266 brouard 11472: free_vector(agemaxgoodr,1, ncovcombmax);
11473: free_vector(agemaxgood,1, ncovcombmax);
11474: free_vector(agemingood,1, ncovcombmax);
11475: free_vector(agemingoodr,1, ncovcombmax);
11476: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 11477: free_vector(sumnewm,1, ncovcombmax);
11478: free_vector(sumnewp,1, ncovcombmax);
11479: return 0;
11480: }/* End movingaverage */
1.218 brouard 11481:
1.126 brouard 11482:
1.296 brouard 11483:
1.126 brouard 11484: /************** Forecasting ******************/
1.296 brouard 11485: /* 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)*/
11486: 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){
11487: /* dateintemean, mean date of interviews
11488: dateprojd, year, month, day of starting projection
11489: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 11490: agemin, agemax range of age
11491: dateprev1 dateprev2 range of dates during which prevalence is computed
11492: */
1.296 brouard 11493: /* double anprojd, mprojd, jprojd; */
11494: /* double anprojf, mprojf, jprojf; */
1.359 brouard 11495: int yearp, stepsize, hstepm, nhstepm, j, k, i, h, nres=0;
1.126 brouard 11496: double agec; /* generic age */
1.359 brouard 11497: double agelim, ppij;
11498: /*double *popcount;*/
1.126 brouard 11499: double ***p3mat;
1.218 brouard 11500: /* double ***mobaverage; */
1.126 brouard 11501: char fileresf[FILENAMELENGTH];
11502:
11503: agelim=AGESUP;
1.211 brouard 11504: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
11505: in each health status at the date of interview (if between dateprev1 and dateprev2).
11506: We still use firstpass and lastpass as another selection.
11507: */
1.214 brouard 11508: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
11509: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 11510:
1.201 brouard 11511: strcpy(fileresf,"F_");
11512: strcat(fileresf,fileresu);
1.126 brouard 11513: if((ficresf=fopen(fileresf,"w"))==NULL) {
11514: printf("Problem with forecast resultfile: %s\n", fileresf);
11515: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
11516: }
1.235 brouard 11517: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
11518: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 11519:
1.225 brouard 11520: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 11521:
11522:
11523: stepsize=(int) (stepm+YEARM-1)/YEARM;
11524: if (stepm<=12) stepsize=1;
11525: if(estepm < stepm){
11526: printf ("Problem %d lower than %d\n",estepm, stepm);
11527: }
1.270 brouard 11528: else{
11529: hstepm=estepm;
11530: }
11531: if(estepm > stepm){ /* Yes every two year */
11532: stepsize=2;
11533: }
1.296 brouard 11534: hstepm=hstepm/stepm;
1.126 brouard 11535:
1.296 brouard 11536:
11537: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
11538: /* fractional in yp1 *\/ */
11539: /* aintmean=yp; */
11540: /* yp2=modf((yp1*12),&yp); */
11541: /* mintmean=yp; */
11542: /* yp1=modf((yp2*30.5),&yp); */
11543: /* jintmean=yp; */
11544: /* if(jintmean==0) jintmean=1; */
11545: /* if(mintmean==0) mintmean=1; */
1.126 brouard 11546:
1.296 brouard 11547:
11548: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
11549: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
11550: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.351 brouard 11551: /* i1=pow(2,cptcoveff); */
11552: /* if (cptcovn < 1){i1=1;} */
1.126 brouard 11553:
1.296 brouard 11554: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 11555:
11556: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 11557:
1.126 brouard 11558: /* if (h==(int)(YEARM*yearp)){ */
1.351 brouard 11559: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11560: k=TKresult[nres];
11561: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
11562: /* 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) *\/ */
11563: /* if(i1 != 1 && TKresult[nres]!= k) */
11564: /* continue; */
11565: /* if(invalidvarcomb[k]){ */
11566: /* printf("\nCombination (%d) projection ignored because no cases \n",k); */
11567: /* continue; */
11568: /* } */
1.227 brouard 11569: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
1.351 brouard 11570: for(j=1;j<=cptcovs;j++){
11571: /* for(j=1;j<=cptcoveff;j++) { */
11572: /* /\* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); *\/ */
11573: /* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11574: /* } */
11575: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
11576: /* fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11577: /* } */
11578: fprintf(ficresf," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.235 brouard 11579: }
1.351 brouard 11580:
1.227 brouard 11581: fprintf(ficresf," yearproj age");
11582: for(j=1; j<=nlstate+ndeath;j++){
11583: for(i=1; i<=nlstate;i++)
11584: fprintf(ficresf," p%d%d",i,j);
11585: fprintf(ficresf," wp.%d",j);
11586: }
1.296 brouard 11587: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 11588: fprintf(ficresf,"\n");
1.296 brouard 11589: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 11590: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
11591: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 11592: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
11593: nhstepm = nhstepm/hstepm;
11594: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11595: oldm=oldms;savm=savms;
1.268 brouard 11596: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 11597: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 11598: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 11599: for (h=0; h<=nhstepm; h++){
11600: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 11601: break;
11602: }
11603: }
11604: fprintf(ficresf,"\n");
1.351 brouard 11605: /* for(j=1;j<=cptcoveff;j++) */
11606: for(j=1;j<=cptcovs;j++)
11607: fprintf(ficresf,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.332 brouard 11608: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Tvaraff not correct *\/ */
1.351 brouard 11609: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* TnsdVar[Tvaraff] correct *\/ */
1.296 brouard 11610: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 11611:
11612: for(j=1; j<=nlstate+ndeath;j++) {
11613: ppij=0.;
11614: for(i=1; i<=nlstate;i++) {
1.278 brouard 11615: if (mobilav>=1)
11616: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
11617: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
11618: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
11619: }
1.268 brouard 11620: fprintf(ficresf," %.3f", p3mat[i][j][h]);
11621: } /* end i */
11622: fprintf(ficresf," %.3f", ppij);
11623: }/* end j */
1.227 brouard 11624: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11625: } /* end agec */
1.266 brouard 11626: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
11627: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 11628: } /* end yearp */
11629: } /* end k */
1.219 brouard 11630:
1.126 brouard 11631: fclose(ficresf);
1.215 brouard 11632: printf("End of Computing forecasting \n");
11633: fprintf(ficlog,"End of Computing forecasting\n");
11634:
1.126 brouard 11635: }
11636:
1.269 brouard 11637: /************** Back Forecasting ******************/
1.296 brouard 11638: /* 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){ */
11639: 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){
11640: /* back1, year, month, day of starting backprojection
1.267 brouard 11641: agemin, agemax range of age
11642: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 11643: anback2 year of end of backprojection (same day and month as back1).
11644: prevacurrent and prev are prevalences.
1.267 brouard 11645: */
1.359 brouard 11646: int yearp, stepsize, hstepm, nhstepm, j, k, i, h, nres=0;
1.267 brouard 11647: double agec; /* generic age */
1.359 brouard 11648: double agelim, ppij, ppi; /* ,jintmean,mintmean,aintmean;*/
11649: /*double *popcount;*/
1.267 brouard 11650: double ***p3mat;
11651: /* double ***mobaverage; */
11652: char fileresfb[FILENAMELENGTH];
11653:
1.268 brouard 11654: agelim=AGEINF;
1.267 brouard 11655: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
11656: in each health status at the date of interview (if between dateprev1 and dateprev2).
11657: We still use firstpass and lastpass as another selection.
11658: */
11659: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
11660: /* firstpass, lastpass, stepm, weightopt, model); */
11661:
11662: /*Do we need to compute prevalence again?*/
11663:
11664: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11665:
11666: strcpy(fileresfb,"FB_");
11667: strcat(fileresfb,fileresu);
11668: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
11669: printf("Problem with back forecast resultfile: %s\n", fileresfb);
11670: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
11671: }
11672: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
11673: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
11674:
11675: if (cptcoveff==0) ncodemax[cptcoveff]=1;
11676:
11677:
11678: stepsize=(int) (stepm+YEARM-1)/YEARM;
11679: if (stepm<=12) stepsize=1;
11680: if(estepm < stepm){
11681: printf ("Problem %d lower than %d\n",estepm, stepm);
11682: }
1.270 brouard 11683: else{
11684: hstepm=estepm;
11685: }
11686: if(estepm >= stepm){ /* Yes every two year */
11687: stepsize=2;
11688: }
1.267 brouard 11689:
11690: hstepm=hstepm/stepm;
1.296 brouard 11691: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
11692: /* fractional in yp1 *\/ */
11693: /* aintmean=yp; */
11694: /* yp2=modf((yp1*12),&yp); */
11695: /* mintmean=yp; */
11696: /* yp1=modf((yp2*30.5),&yp); */
11697: /* jintmean=yp; */
11698: /* if(jintmean==0) jintmean=1; */
11699: /* if(mintmean==0) jintmean=1; */
1.267 brouard 11700:
1.351 brouard 11701: /* i1=pow(2,cptcoveff); */
11702: /* if (cptcovn < 1){i1=1;} */
1.267 brouard 11703:
1.296 brouard 11704: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
11705: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 11706:
11707: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
11708:
1.351 brouard 11709: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11710: k=TKresult[nres];
11711: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
11712: /* for(k=1; k<=i1;k++){ */
11713: /* if(i1 != 1 && TKresult[nres]!= k) */
11714: /* continue; */
11715: /* if(invalidvarcomb[k]){ */
11716: /* printf("\nCombination (%d) projection ignored because no cases \n",k); */
11717: /* continue; */
11718: /* } */
1.268 brouard 11719: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.351 brouard 11720: for(j=1;j<=cptcovs;j++){
11721: /* for(j=1;j<=cptcoveff;j++) { */
11722: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11723: /* } */
11724: fprintf(ficresfb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.267 brouard 11725: }
1.351 brouard 11726: /* fprintf(ficrespij,"******\n"); */
11727: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
11728: /* fprintf(ficresfb," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11729: /* } */
1.267 brouard 11730: fprintf(ficresfb," yearbproj age");
11731: for(j=1; j<=nlstate+ndeath;j++){
11732: for(i=1; i<=nlstate;i++)
1.268 brouard 11733: fprintf(ficresfb," b%d%d",i,j);
11734: fprintf(ficresfb," b.%d",j);
1.267 brouard 11735: }
1.296 brouard 11736: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 11737: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
11738: fprintf(ficresfb,"\n");
1.296 brouard 11739: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 11740: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 11741: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
11742: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 11743: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 11744: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 11745: nhstepm = nhstepm/hstepm;
11746: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11747: oldm=oldms;savm=savms;
1.268 brouard 11748: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 11749: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 11750: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 11751: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
11752: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
11753: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 11754: for (h=0; h<=nhstepm; h++){
1.268 brouard 11755: if (h*hstepm/YEARM*stepm ==-yearp) {
11756: break;
11757: }
11758: }
11759: fprintf(ficresfb,"\n");
1.351 brouard 11760: /* for(j=1;j<=cptcoveff;j++) */
11761: for(j=1;j<=cptcovs;j++)
11762: fprintf(ficresfb,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11763: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.296 brouard 11764: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 11765: for(i=1; i<=nlstate+ndeath;i++) {
11766: ppij=0.;ppi=0.;
11767: for(j=1; j<=nlstate;j++) {
11768: /* if (mobilav==1) */
1.269 brouard 11769: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
11770: ppi=ppi+prevacurrent[(int)agec][j][k];
11771: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
11772: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 11773: /* else { */
11774: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
11775: /* } */
1.268 brouard 11776: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
11777: } /* end j */
11778: if(ppi <0.99){
11779: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
11780: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
11781: }
11782: fprintf(ficresfb," %.3f", ppij);
11783: }/* end j */
1.267 brouard 11784: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11785: } /* end agec */
11786: } /* end yearp */
11787: } /* end k */
1.217 brouard 11788:
1.267 brouard 11789: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 11790:
1.267 brouard 11791: fclose(ficresfb);
11792: printf("End of Computing Back forecasting \n");
11793: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 11794:
1.267 brouard 11795: }
1.217 brouard 11796:
1.269 brouard 11797: /* Variance of prevalence limit: varprlim */
11798: 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 11799: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 11800:
11801: char fileresvpl[FILENAMELENGTH];
11802: FILE *ficresvpl;
11803: double **oldm, **savm;
11804: double **varpl; /* Variances of prevalence limits by age */
11805: int i1, k, nres, j ;
11806:
11807: strcpy(fileresvpl,"VPL_");
11808: strcat(fileresvpl,fileresu);
11809: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 11810: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 11811: exit(0);
11812: }
1.288 brouard 11813: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11814: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 11815:
11816: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11817: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11818:
11819: i1=pow(2,cptcoveff);
11820: if (cptcovn < 1){i1=1;}
11821:
1.337 brouard 11822: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11823: k=TKresult[nres];
1.338 brouard 11824: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 11825: /* for(k=1; k<=i1;k++){ /\* We find the combination equivalent to result line values of dummies *\/ */
1.269 brouard 11826: if(i1 != 1 && TKresult[nres]!= k)
11827: continue;
11828: fprintf(ficresvpl,"\n#****** ");
11829: printf("\n#****** ");
11830: fprintf(ficlog,"\n#****** ");
1.337 brouard 11831: for(j=1;j<=cptcovs;j++) {
11832: fprintf(ficresvpl,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11833: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11834: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11835: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11836: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.269 brouard 11837: }
1.337 brouard 11838: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
11839: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
11840: /* fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
11841: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
11842: /* } */
1.269 brouard 11843: fprintf(ficresvpl,"******\n");
11844: printf("******\n");
11845: fprintf(ficlog,"******\n");
11846:
11847: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11848: oldm=oldms;savm=savms;
11849: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
11850: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
11851: /*}*/
11852: }
11853:
11854: fclose(ficresvpl);
1.288 brouard 11855: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
11856: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 11857:
11858: }
11859: /* Variance of back prevalence: varbprlim */
11860: 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){
11861: /*------- Variance of back (stable) prevalence------*/
11862:
11863: char fileresvbl[FILENAMELENGTH];
11864: FILE *ficresvbl;
11865:
11866: double **oldm, **savm;
11867: double **varbpl; /* Variances of back prevalence limits by age */
11868: int i1, k, nres, j ;
11869:
11870: strcpy(fileresvbl,"VBL_");
11871: strcat(fileresvbl,fileresu);
11872: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
11873: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
11874: exit(0);
11875: }
11876: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
11877: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
11878:
11879:
11880: i1=pow(2,cptcoveff);
11881: if (cptcovn < 1){i1=1;}
11882:
1.337 brouard 11883: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11884: k=TKresult[nres];
1.338 brouard 11885: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 11886: /* for(k=1; k<=i1;k++){ */
11887: /* if(i1 != 1 && TKresult[nres]!= k) */
11888: /* continue; */
1.269 brouard 11889: fprintf(ficresvbl,"\n#****** ");
11890: printf("\n#****** ");
11891: fprintf(ficlog,"\n#****** ");
1.337 brouard 11892: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */
1.338 brouard 11893: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
11894: fprintf(ficresvbl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
11895: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
1.337 brouard 11896: /* for(j=1;j<=cptcoveff;j++) { */
11897: /* fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11898: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11899: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11900: /* } */
11901: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
11902: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
11903: /* fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
11904: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
1.269 brouard 11905: }
11906: fprintf(ficresvbl,"******\n");
11907: printf("******\n");
11908: fprintf(ficlog,"******\n");
11909:
11910: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
11911: oldm=oldms;savm=savms;
11912:
11913: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
11914: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
11915: /*}*/
11916: }
11917:
11918: fclose(ficresvbl);
11919: printf("done variance-covariance of back prevalence\n");fflush(stdout);
11920: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
11921:
11922: } /* End of varbprlim */
11923:
1.126 brouard 11924: /************** Forecasting *****not tested NB*************/
1.227 brouard 11925: /* 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 11926:
1.227 brouard 11927: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
11928: /* int *popage; */
11929: /* double calagedatem, agelim, kk1, kk2; */
11930: /* double *popeffectif,*popcount; */
11931: /* double ***p3mat,***tabpop,***tabpopprev; */
11932: /* /\* double ***mobaverage; *\/ */
11933: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 11934:
1.227 brouard 11935: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
11936: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
11937: /* agelim=AGESUP; */
11938: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 11939:
1.227 brouard 11940: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 11941:
11942:
1.227 brouard 11943: /* strcpy(filerespop,"POP_"); */
11944: /* strcat(filerespop,fileresu); */
11945: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
11946: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
11947: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
11948: /* } */
11949: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
11950: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 11951:
1.227 brouard 11952: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 11953:
1.227 brouard 11954: /* /\* if (mobilav!=0) { *\/ */
11955: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
11956: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
11957: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
11958: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
11959: /* /\* } *\/ */
11960: /* /\* } *\/ */
1.126 brouard 11961:
1.227 brouard 11962: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
11963: /* if (stepm<=12) stepsize=1; */
1.126 brouard 11964:
1.227 brouard 11965: /* agelim=AGESUP; */
1.126 brouard 11966:
1.227 brouard 11967: /* hstepm=1; */
11968: /* hstepm=hstepm/stepm; */
1.218 brouard 11969:
1.227 brouard 11970: /* if (popforecast==1) { */
11971: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
11972: /* printf("Problem with population file : %s\n",popfile);exit(0); */
11973: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
11974: /* } */
11975: /* popage=ivector(0,AGESUP); */
11976: /* popeffectif=vector(0,AGESUP); */
11977: /* popcount=vector(0,AGESUP); */
1.126 brouard 11978:
1.227 brouard 11979: /* i=1; */
11980: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 11981:
1.227 brouard 11982: /* imx=i; */
11983: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
11984: /* } */
1.218 brouard 11985:
1.227 brouard 11986: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
11987: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
11988: /* k=k+1; */
11989: /* fprintf(ficrespop,"\n#******"); */
11990: /* for(j=1;j<=cptcoveff;j++) { */
11991: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
11992: /* } */
11993: /* fprintf(ficrespop,"******\n"); */
11994: /* fprintf(ficrespop,"# Age"); */
11995: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
11996: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 11997:
1.227 brouard 11998: /* for (cpt=0; cpt<=0;cpt++) { */
11999: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 12000:
1.227 brouard 12001: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
12002: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
12003: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 12004:
1.227 brouard 12005: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
12006: /* oldm=oldms;savm=savms; */
12007: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 12008:
1.227 brouard 12009: /* for (h=0; h<=nhstepm; h++){ */
12010: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
12011: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
12012: /* } */
12013: /* for(j=1; j<=nlstate+ndeath;j++) { */
12014: /* kk1=0.;kk2=0; */
12015: /* for(i=1; i<=nlstate;i++) { */
12016: /* if (mobilav==1) */
12017: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
12018: /* else { */
12019: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
12020: /* } */
12021: /* } */
12022: /* if (h==(int)(calagedatem+12*cpt)){ */
12023: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
12024: /* /\*fprintf(ficrespop," %.3f", kk1); */
12025: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
12026: /* } */
12027: /* } */
12028: /* for(i=1; i<=nlstate;i++){ */
12029: /* kk1=0.; */
12030: /* for(j=1; j<=nlstate;j++){ */
12031: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
12032: /* } */
12033: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
12034: /* } */
1.218 brouard 12035:
1.227 brouard 12036: /* if (h==(int)(calagedatem+12*cpt)) */
12037: /* for(j=1; j<=nlstate;j++) */
12038: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
12039: /* } */
12040: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
12041: /* } */
12042: /* } */
1.218 brouard 12043:
1.227 brouard 12044: /* /\******\/ */
1.218 brouard 12045:
1.227 brouard 12046: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
12047: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
12048: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
12049: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
12050: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 12051:
1.227 brouard 12052: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
12053: /* oldm=oldms;savm=savms; */
12054: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
12055: /* for (h=0; h<=nhstepm; h++){ */
12056: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
12057: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
12058: /* } */
12059: /* for(j=1; j<=nlstate+ndeath;j++) { */
12060: /* kk1=0.;kk2=0; */
12061: /* for(i=1; i<=nlstate;i++) { */
12062: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
12063: /* } */
12064: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
12065: /* } */
12066: /* } */
12067: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
12068: /* } */
12069: /* } */
12070: /* } */
12071: /* } */
1.218 brouard 12072:
1.227 brouard 12073: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 12074:
1.227 brouard 12075: /* if (popforecast==1) { */
12076: /* free_ivector(popage,0,AGESUP); */
12077: /* free_vector(popeffectif,0,AGESUP); */
12078: /* free_vector(popcount,0,AGESUP); */
12079: /* } */
12080: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
12081: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
12082: /* fclose(ficrespop); */
12083: /* } /\* End of popforecast *\/ */
1.218 brouard 12084:
1.126 brouard 12085: int fileappend(FILE *fichier, char *optionfich)
12086: {
12087: if((fichier=fopen(optionfich,"a"))==NULL) {
12088: printf("Problem with file: %s\n", optionfich);
12089: fprintf(ficlog,"Problem with file: %s\n", optionfich);
12090: return (0);
12091: }
12092: fflush(fichier);
12093: return (1);
12094: }
12095:
12096:
12097: /**************** function prwizard **********************/
12098: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
12099: {
12100:
12101: /* Wizard to print covariance matrix template */
12102:
1.164 brouard 12103: char ca[32], cb[32];
12104: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 12105: int numlinepar;
12106:
12107: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12108: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12109: for(i=1; i <=nlstate; i++){
12110: jj=0;
12111: for(j=1; j <=nlstate+ndeath; j++){
12112: if(j==i) continue;
12113: jj++;
12114: /*ca[0]= k+'a'-1;ca[1]='\0';*/
12115: printf("%1d%1d",i,j);
12116: fprintf(ficparo,"%1d%1d",i,j);
12117: for(k=1; k<=ncovmodel;k++){
12118: /* printf(" %lf",param[i][j][k]); */
12119: /* fprintf(ficparo," %lf",param[i][j][k]); */
12120: printf(" 0.");
12121: fprintf(ficparo," 0.");
12122: }
12123: printf("\n");
12124: fprintf(ficparo,"\n");
12125: }
12126: }
12127: printf("# Scales (for hessian or gradient estimation)\n");
12128: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
12129: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
12130: for(i=1; i <=nlstate; i++){
12131: jj=0;
12132: for(j=1; j <=nlstate+ndeath; j++){
12133: if(j==i) continue;
12134: jj++;
12135: fprintf(ficparo,"%1d%1d",i,j);
12136: printf("%1d%1d",i,j);
12137: fflush(stdout);
12138: for(k=1; k<=ncovmodel;k++){
12139: /* printf(" %le",delti3[i][j][k]); */
12140: /* fprintf(ficparo," %le",delti3[i][j][k]); */
12141: printf(" 0.");
12142: fprintf(ficparo," 0.");
12143: }
12144: numlinepar++;
12145: printf("\n");
12146: fprintf(ficparo,"\n");
12147: }
12148: }
12149: printf("# Covariance matrix\n");
12150: /* # 121 Var(a12)\n\ */
12151: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12152: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12153: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12154: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12155: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12156: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12157: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12158: fflush(stdout);
12159: fprintf(ficparo,"# Covariance matrix\n");
12160: /* # 121 Var(a12)\n\ */
12161: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12162: /* # ...\n\ */
12163: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12164:
12165: for(itimes=1;itimes<=2;itimes++){
12166: jj=0;
12167: for(i=1; i <=nlstate; i++){
12168: for(j=1; j <=nlstate+ndeath; j++){
12169: if(j==i) continue;
12170: for(k=1; k<=ncovmodel;k++){
12171: jj++;
12172: ca[0]= k+'a'-1;ca[1]='\0';
12173: if(itimes==1){
12174: printf("#%1d%1d%d",i,j,k);
12175: fprintf(ficparo,"#%1d%1d%d",i,j,k);
12176: }else{
12177: printf("%1d%1d%d",i,j,k);
12178: fprintf(ficparo,"%1d%1d%d",i,j,k);
12179: /* printf(" %.5le",matcov[i][j]); */
12180: }
12181: ll=0;
12182: for(li=1;li <=nlstate; li++){
12183: for(lj=1;lj <=nlstate+ndeath; lj++){
12184: if(lj==li) continue;
12185: for(lk=1;lk<=ncovmodel;lk++){
12186: ll++;
12187: if(ll<=jj){
12188: cb[0]= lk +'a'-1;cb[1]='\0';
12189: if(ll<jj){
12190: if(itimes==1){
12191: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12192: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12193: }else{
12194: printf(" 0.");
12195: fprintf(ficparo," 0.");
12196: }
12197: }else{
12198: if(itimes==1){
12199: printf(" Var(%s%1d%1d)",ca,i,j);
12200: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
12201: }else{
12202: printf(" 0.");
12203: fprintf(ficparo," 0.");
12204: }
12205: }
12206: }
12207: } /* end lk */
12208: } /* end lj */
12209: } /* end li */
12210: printf("\n");
12211: fprintf(ficparo,"\n");
12212: numlinepar++;
12213: } /* end k*/
12214: } /*end j */
12215: } /* end i */
12216: } /* end itimes */
12217:
12218: } /* end of prwizard */
12219: /******************* Gompertz Likelihood ******************************/
12220: double gompertz(double x[])
12221: {
1.302 brouard 12222: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 12223: int i,n=0; /* n is the size of the sample */
12224:
1.220 brouard 12225: for (i=1;i<=imx ; i++) {
1.126 brouard 12226: sump=sump+weight[i];
12227: /* sump=sump+1;*/
12228: num=num+1;
12229: }
1.302 brouard 12230: L=0.0;
12231: /* agegomp=AGEGOMP; */
1.126 brouard 12232: /* for (i=0; i<=imx; i++)
12233: 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]);*/
12234:
1.302 brouard 12235: for (i=1;i<=imx ; i++) {
12236: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
12237: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
12238: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
12239: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
12240: * +
12241: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
12242: */
12243: if (wav[i] > 1 || agedc[i] < AGESUP) {
12244: if (cens[i] == 1){
12245: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
12246: } else if (cens[i] == 0){
1.126 brouard 12247: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 12248: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
12249: } else
12250: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 12251: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 12252: L=L+A*weight[i];
1.126 brouard 12253: /* 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 12254: }
12255: }
1.126 brouard 12256:
1.302 brouard 12257: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 12258:
12259: return -2*L*num/sump;
12260: }
12261:
1.136 brouard 12262: #ifdef GSL
12263: /******************* Gompertz_f Likelihood ******************************/
12264: double gompertz_f(const gsl_vector *v, void *params)
12265: {
1.302 brouard 12266: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 12267: double *x= (double *) v->data;
12268: int i,n=0; /* n is the size of the sample */
12269:
12270: for (i=0;i<=imx-1 ; i++) {
12271: sump=sump+weight[i];
12272: /* sump=sump+1;*/
12273: num=num+1;
12274: }
12275:
12276:
12277: /* for (i=0; i<=imx; i++)
12278: 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]);*/
12279: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
12280: for (i=1;i<=imx ; i++)
12281: {
12282: if (cens[i] == 1 && wav[i]>1)
12283: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
12284:
12285: if (cens[i] == 0 && wav[i]>1)
12286: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
12287: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
12288:
12289: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
12290: if (wav[i] > 1 ) { /* ??? */
12291: LL=LL+A*weight[i];
12292: /* 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]);*/
12293: }
12294: }
12295:
12296: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
12297: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
12298:
12299: return -2*LL*num/sump;
12300: }
12301: #endif
12302:
1.126 brouard 12303: /******************* Printing html file ***********/
1.201 brouard 12304: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 12305: int lastpass, int stepm, int weightopt, char model[],\
12306: int imx, double p[],double **matcov,double agemortsup){
12307: int i,k;
12308:
12309: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
12310: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
12311: for (i=1;i<=2;i++)
12312: 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 12313: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 12314: fprintf(fichtm,"</ul>");
12315:
12316: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
12317:
12318: 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>");
12319:
12320: for (k=agegomp;k<(agemortsup-2);k++)
12321: 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]);
12322:
12323:
12324: fflush(fichtm);
12325: }
12326:
12327: /******************* Gnuplot file **************/
1.201 brouard 12328: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 12329:
12330: char dirfileres[132],optfileres[132];
1.164 brouard 12331:
1.359 brouard 12332: /*int ng;*/
1.126 brouard 12333:
12334:
12335: /*#ifdef windows */
12336: fprintf(ficgp,"cd \"%s\" \n",pathc);
12337: /*#endif */
12338:
12339:
12340: strcpy(dirfileres,optionfilefiname);
12341: strcpy(optfileres,"vpl");
1.199 brouard 12342: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 12343: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 12344: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 12345: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 12346: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
12347:
12348: }
12349:
1.136 brouard 12350: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
12351: {
1.126 brouard 12352:
1.136 brouard 12353: /*-------- data file ----------*/
12354: FILE *fic;
12355: char dummy[]=" ";
1.359 brouard 12356: int i = 0, j = 0, n = 0, iv = 0;/* , v;*/
1.223 brouard 12357: int lstra;
1.136 brouard 12358: int linei, month, year,iout;
1.302 brouard 12359: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 12360: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 12361: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 12362: char *stratrunc;
1.223 brouard 12363:
1.349 brouard 12364: /* DummyV=ivector(-1,NCOVMAX); /\* 1 to 3 *\/ */
12365: /* FixedV=ivector(-1,NCOVMAX); /\* 1 to 3 *\/ */
1.339 brouard 12366:
12367: ncovcolt=ncovcol+nqv+ntv+nqtv; /* total of covariates in the data, not in the model equation */
12368:
1.136 brouard 12369: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 12370: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
12371: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 12372: }
1.126 brouard 12373:
1.302 brouard 12374: /* Is it a BOM UTF-8 Windows file? */
12375: /* First data line */
12376: linei=0;
12377: while(fgets(line, MAXLINE, fic)) {
12378: noffset=0;
12379: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
12380: {
12381: noffset=noffset+3;
12382: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
12383: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
12384: fflush(ficlog); return 1;
12385: }
12386: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
12387: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
12388: {
12389: noffset=noffset+2;
1.304 brouard 12390: 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);
12391: 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 12392: fflush(ficlog); return 1;
12393: }
12394: else if( line[0] == 0 && line[1] == 0)
12395: {
12396: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
12397: noffset=noffset+4;
1.304 brouard 12398: 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);
12399: 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 12400: fflush(ficlog); return 1;
12401: }
12402: } else{
12403: ;/*printf(" Not a BOM file\n");*/
12404: }
12405: /* If line starts with a # it is a comment */
12406: if (line[noffset] == '#') {
12407: linei=linei+1;
12408: break;
12409: }else{
12410: break;
12411: }
12412: }
12413: fclose(fic);
12414: if((fic=fopen(datafile,"r"))==NULL) {
12415: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
12416: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
12417: }
12418: /* Not a Bom file */
12419:
1.136 brouard 12420: i=1;
12421: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
12422: linei=linei+1;
12423: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
12424: if(line[j] == '\t')
12425: line[j] = ' ';
12426: }
12427: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
12428: ;
12429: };
12430: line[j+1]=0; /* Trims blanks at end of line */
12431: if(line[0]=='#'){
12432: fprintf(ficlog,"Comment line\n%s\n",line);
12433: printf("Comment line\n%s\n",line);
12434: continue;
12435: }
12436: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 12437: strcpy(line, linetmp);
1.223 brouard 12438:
12439: /* Loops on waves */
12440: for (j=maxwav;j>=1;j--){
12441: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 12442: cutv(stra, strb, line, ' ');
12443: if(strb[0]=='.') { /* Missing value */
12444: lval=-1;
12445: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
1.341 brouard 12446: cotvar[j][ncovcol+nqv+ntv+iv][i]=-1; /* For performance reasons */
1.238 brouard 12447: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
12448: 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);
12449: 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);
12450: return 1;
12451: }
12452: }else{
12453: errno=0;
12454: /* what_kind_of_number(strb); */
12455: dval=strtod(strb,&endptr);
12456: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
12457: /* if(strb != endptr && *endptr == '\0') */
12458: /* dval=dlval; */
12459: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
12460: if( strb[0]=='\0' || (*endptr != '\0')){
12461: 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);
12462: 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);
12463: return 1;
12464: }
12465: cotqvar[j][iv][i]=dval;
1.341 brouard 12466: cotvar[j][ncovcol+nqv+ntv+iv][i]=dval; /* because cotvar starts now at first ntv */
1.238 brouard 12467: }
12468: strcpy(line,stra);
1.223 brouard 12469: }/* end loop ntqv */
1.225 brouard 12470:
1.223 brouard 12471: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 12472: cutv(stra, strb, line, ' ');
12473: if(strb[0]=='.') { /* Missing value */
12474: lval=-1;
12475: }else{
12476: errno=0;
12477: lval=strtol(strb,&endptr,10);
12478: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
12479: if( strb[0]=='\0' || (*endptr != '\0')){
12480: 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);
12481: 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);
12482: return 1;
12483: }
12484: }
12485: if(lval <-1 || lval >1){
12486: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 12487: 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 12488: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 12489: For example, for multinomial values like 1, 2 and 3,\n \
12490: build V1=0 V2=0 for the reference value (1),\n \
12491: V1=1 V2=0 for (2) \n \
1.223 brouard 12492: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 12493: output of IMaCh is often meaningless.\n \
1.319 brouard 12494: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 12495: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 12496: 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 12497: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 12498: For example, for multinomial values like 1, 2 and 3,\n \
12499: build V1=0 V2=0 for the reference value (1),\n \
12500: V1=1 V2=0 for (2) \n \
1.223 brouard 12501: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 12502: output of IMaCh is often meaningless.\n \
1.319 brouard 12503: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 12504: return 1;
12505: }
1.341 brouard 12506: cotvar[j][ncovcol+nqv+iv][i]=(double)(lval);
1.238 brouard 12507: strcpy(line,stra);
1.223 brouard 12508: }/* end loop ntv */
1.225 brouard 12509:
1.223 brouard 12510: /* Statuses at wave */
1.137 brouard 12511: cutv(stra, strb, line, ' ');
1.223 brouard 12512: if(strb[0]=='.') { /* Missing value */
1.238 brouard 12513: lval=-1;
1.136 brouard 12514: }else{
1.238 brouard 12515: errno=0;
12516: lval=strtol(strb,&endptr,10);
12517: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
1.347 brouard 12518: if( strb[0]=='\0' || (*endptr != '\0' )){
12519: 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);
12520: 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);
12521: return 1;
12522: }else if( lval==0 || lval > nlstate+ndeath){
1.348 brouard 12523: 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);
12524: 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 12525: return 1;
12526: }
1.136 brouard 12527: }
1.225 brouard 12528:
1.136 brouard 12529: s[j][i]=lval;
1.225 brouard 12530:
1.223 brouard 12531: /* Date of Interview */
1.136 brouard 12532: strcpy(line,stra);
12533: cutv(stra, strb,line,' ');
1.169 brouard 12534: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 12535: }
1.169 brouard 12536: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 12537: month=99;
12538: year=9999;
1.136 brouard 12539: }else{
1.225 brouard 12540: 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);
12541: 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);
12542: return 1;
1.136 brouard 12543: }
12544: anint[j][i]= (double) year;
1.302 brouard 12545: mint[j][i]= (double)month;
12546: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
12547: /* 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]); */
12548: /* 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]); */
12549: /* } */
1.136 brouard 12550: strcpy(line,stra);
1.223 brouard 12551: } /* End loop on waves */
1.225 brouard 12552:
1.223 brouard 12553: /* Date of death */
1.136 brouard 12554: cutv(stra, strb,line,' ');
1.169 brouard 12555: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 12556: }
1.169 brouard 12557: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 12558: month=99;
12559: year=9999;
12560: }else{
1.141 brouard 12561: 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 12562: 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);
12563: return 1;
1.136 brouard 12564: }
12565: andc[i]=(double) year;
12566: moisdc[i]=(double) month;
12567: strcpy(line,stra);
12568:
1.223 brouard 12569: /* Date of birth */
1.136 brouard 12570: cutv(stra, strb,line,' ');
1.169 brouard 12571: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 12572: }
1.169 brouard 12573: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 12574: month=99;
12575: year=9999;
12576: }else{
1.141 brouard 12577: 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);
12578: 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 12579: return 1;
1.136 brouard 12580: }
12581: if (year==9999) {
1.141 brouard 12582: 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);
12583: 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 12584: return 1;
12585:
1.136 brouard 12586: }
12587: annais[i]=(double)(year);
1.302 brouard 12588: moisnais[i]=(double)(month);
12589: for (j=1;j<=maxwav;j++){
12590: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
12591: 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]);
12592: 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]);
12593: }
12594: }
12595:
1.136 brouard 12596: strcpy(line,stra);
1.225 brouard 12597:
1.223 brouard 12598: /* Sample weight */
1.136 brouard 12599: cutv(stra, strb,line,' ');
12600: errno=0;
12601: dval=strtod(strb,&endptr);
12602: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 12603: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
12604: 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 12605: fflush(ficlog);
12606: return 1;
12607: }
12608: weight[i]=dval;
12609: strcpy(line,stra);
1.225 brouard 12610:
1.223 brouard 12611: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
12612: cutv(stra, strb, line, ' ');
12613: if(strb[0]=='.') { /* Missing value */
1.225 brouard 12614: lval=-1;
1.311 brouard 12615: coqvar[iv][i]=NAN;
12616: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 12617: }else{
1.225 brouard 12618: errno=0;
12619: /* what_kind_of_number(strb); */
12620: dval=strtod(strb,&endptr);
12621: /* if(strb != endptr && *endptr == '\0') */
12622: /* dval=dlval; */
12623: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
12624: if( strb[0]=='\0' || (*endptr != '\0')){
12625: 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);
12626: 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);
12627: return 1;
12628: }
12629: coqvar[iv][i]=dval;
1.226 brouard 12630: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 12631: }
12632: strcpy(line,stra);
12633: }/* end loop nqv */
1.136 brouard 12634:
1.223 brouard 12635: /* Covariate values */
1.136 brouard 12636: for (j=ncovcol;j>=1;j--){
12637: cutv(stra, strb,line,' ');
1.223 brouard 12638: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 12639: lval=-1;
1.136 brouard 12640: }else{
1.225 brouard 12641: errno=0;
12642: lval=strtol(strb,&endptr,10);
12643: if( strb[0]=='\0' || (*endptr != '\0')){
12644: 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);
12645: 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);
12646: return 1;
12647: }
1.136 brouard 12648: }
12649: if(lval <-1 || lval >1){
1.225 brouard 12650: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 12651: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
12652: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 12653: For example, for multinomial values like 1, 2 and 3,\n \
12654: build V1=0 V2=0 for the reference value (1),\n \
12655: V1=1 V2=0 for (2) \n \
1.136 brouard 12656: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 12657: output of IMaCh is often meaningless.\n \
1.136 brouard 12658: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 12659: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 12660: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
12661: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 12662: For example, for multinomial values like 1, 2 and 3,\n \
12663: build V1=0 V2=0 for the reference value (1),\n \
12664: V1=1 V2=0 for (2) \n \
1.136 brouard 12665: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 12666: output of IMaCh is often meaningless.\n \
1.136 brouard 12667: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 12668: return 1;
1.136 brouard 12669: }
12670: covar[j][i]=(double)(lval);
12671: strcpy(line,stra);
12672: }
12673: lstra=strlen(stra);
1.225 brouard 12674:
1.136 brouard 12675: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
12676: stratrunc = &(stra[lstra-9]);
12677: num[i]=atol(stratrunc);
12678: }
12679: else
12680: num[i]=atol(stra);
12681: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
12682: 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;}*/
12683:
12684: i=i+1;
12685: } /* End loop reading data */
1.225 brouard 12686:
1.136 brouard 12687: *imax=i-1; /* Number of individuals */
12688: fclose(fic);
1.225 brouard 12689:
1.136 brouard 12690: return (0);
1.164 brouard 12691: /* endread: */
1.225 brouard 12692: printf("Exiting readdata: ");
12693: fclose(fic);
12694: return (1);
1.223 brouard 12695: }
1.126 brouard 12696:
1.234 brouard 12697: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 12698: char *p1 = *stri, *p2 = *stri;
1.235 brouard 12699: while (*p2 == ' ')
1.234 brouard 12700: p2++;
12701: /* while ((*p1++ = *p2++) !=0) */
12702: /* ; */
12703: /* do */
12704: /* while (*p2 == ' ') */
12705: /* p2++; */
12706: /* while (*p1++ == *p2++); */
12707: *stri=p2;
1.145 brouard 12708: }
12709:
1.330 brouard 12710: int decoderesult( char resultline[], int nres)
1.230 brouard 12711: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
12712: {
1.235 brouard 12713: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 12714: char resultsav[MAXLINE];
1.330 brouard 12715: /* int resultmodel[MAXLINE]; */
1.334 brouard 12716: /* int modelresult[MAXLINE]; */
1.230 brouard 12717: char stra[80], strb[80], strc[80], strd[80],stre[80];
12718:
1.234 brouard 12719: removefirstspace(&resultline);
1.332 brouard 12720: printf("decoderesult:%s\n",resultline);
1.230 brouard 12721:
1.332 brouard 12722: strcpy(resultsav,resultline);
1.342 brouard 12723: /* printf("Decoderesult resultsav=\"%s\" resultline=\"%s\"\n", resultsav, resultline); */
1.230 brouard 12724: if (strlen(resultsav) >1){
1.334 brouard 12725: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' in this resultline */
1.230 brouard 12726: }
1.353 brouard 12727: if(j == 0 && cptcovs== 0){ /* Resultline but no = and no covariate in the model */
1.253 brouard 12728: TKresult[nres]=0; /* Combination for the nresult and the model */
12729: return (0);
12730: }
1.234 brouard 12731: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.353 brouard 12732: 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);
12733: 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);
12734: if(j==0)
12735: return 1;
1.234 brouard 12736: }
1.334 brouard 12737: for(k=1; k<=j;k++){ /* Loop on any covariate of the RESULT LINE */
1.234 brouard 12738: if(nbocc(resultsav,'=') >1){
1.318 brouard 12739: 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 12740: /* 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 12741: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.332 brouard 12742: /* If a blank, then strc="V4=" and strd='\0' */
12743: if(strc[0]=='\0'){
12744: printf("Error in resultline, probably a blank after the \"%s\", \"result:%s\", stra=\"%s\" resultsav=\"%s\"\n",strb,resultline, stra, resultsav);
12745: fprintf(ficlog,"Error in resultline, probably a blank after the \"V%s=\", resultline=%s\n",strb,resultline);
12746: return 1;
12747: }
1.234 brouard 12748: }else
12749: cutl(strc,strd,resultsav,'=');
1.318 brouard 12750: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 12751:
1.230 brouard 12752: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 12753: 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 12754: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
12755: /* cptcovsel++; */
12756: if (nbocc(stra,'=') >0)
12757: strcpy(resultsav,stra); /* and analyzes it */
12758: }
1.235 brouard 12759: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 12760: /* 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 12761: 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 12762: if(Typevar[k1]==0){ /* Single covariate in model */
12763: /* 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 12764: match=0;
1.318 brouard 12765: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
12766: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 12767: modelresult[nres][k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 12768: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 12769: break;
12770: }
12771: }
12772: if(match == 0){
1.338 brouard 12773: 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]);
12774: 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 12775: return 1;
1.234 brouard 12776: }
1.332 brouard 12777: }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*/
12778: /* We feed resultmodel[k1]=k2; */
12779: match=0;
12780: 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 */
12781: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 12782: 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 12783: resultmodel[nres][k1]=k2; /* Added here */
1.342 brouard 12784: /* printf("Decoderesult first modelresult[k2=%d]=%d (k1) V%d*AGE\n",k2,k1,Tvar[k1]); */
1.332 brouard 12785: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
12786: break;
12787: }
12788: }
12789: if(match == 0){
1.338 brouard 12790: 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]);
12791: 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 12792: return 1;
12793: }
1.349 brouard 12794: }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 12795: /* resultmodel[nres][of such a Vn * Vm product k1] is not unique, so can't exist, we feed Tvard[k1][1] and [2] */
12796: match=0;
1.342 brouard 12797: /* 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 12798: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
12799: if(Tvardk[k1][1]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
12800: /* modelresult[k2]=k1; */
1.342 brouard 12801: /* printf("Decoderesult first Product modelresult[k2=%d]=%d (k1) V%d * \n",k2,k1,Tvarsel[k2]); */
1.332 brouard 12802: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
12803: }
12804: }
12805: if(match == 0){
1.349 brouard 12806: 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);
12807: 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 12808: return 1;
12809: }
12810: match=0;
12811: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
12812: if(Tvardk[k1][2]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
12813: /* modelresult[k2]=k1;*/
1.342 brouard 12814: /* printf("Decoderesult second Product modelresult[k2=%d]=%d (k1) * V%d \n ",k2,k1,Tvarsel[k2]); */
1.332 brouard 12815: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
12816: break;
12817: }
12818: }
12819: if(match == 0){
1.349 brouard 12820: 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);
12821: 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 12822: return 1;
12823: }
12824: }/* End of testing */
1.333 brouard 12825: }/* End loop cptcovt */
1.235 brouard 12826: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 12827: /* Feeds resultmodel[nres][k1]=k2 for single covariate (k1) in the model equation */
1.334 brouard 12828: 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)
12829: * Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 12830: match=0;
1.318 brouard 12831: 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 12832: if(Typevar[k1]==0){ /* Single only */
1.349 brouard 12833: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 What if a product? */
1.330 brouard 12834: 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 12835: 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 12836: ++match;
12837: }
12838: }
12839: }
12840: if(match == 0){
1.338 brouard 12841: printf("Error in result line: variable V%d is missing in model; result: %s, model=1+age+%s\n",Tvarsel[k2], resultline, model);
12842: 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 12843: return 1;
1.234 brouard 12844: }else if(match > 1){
1.338 brouard 12845: printf("Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
12846: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
1.310 brouard 12847: return 1;
1.234 brouard 12848: }
12849: }
1.334 brouard 12850: /* cptcovres=j /\* Number of variables in the resultline is equal to cptcovs and thus useless *\/ */
1.234 brouard 12851: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 12852: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 brouard 12853: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
12854: /* 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*/
12855: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 12856: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
12857: /* 1 0 0 0 */
12858: /* 2 1 0 0 */
12859: /* 3 0 1 0 */
1.330 brouard 12860: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 12861: /* 5 0 0 1 */
1.330 brouard 12862: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 12863: /* 7 0 1 1 */
12864: /* 8 1 1 1 */
1.237 brouard 12865: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
12866: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
12867: /* V5*age V5 known which value for nres? */
12868: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.334 brouard 12869: 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.
12870: * loop on position k1 in the MODEL LINE */
1.331 brouard 12871: /* k counting number of combination of single dummies in the equation model */
12872: /* k4 counting single dummies in the equation model */
12873: /* k4q counting single quantitatives in the equation model */
1.344 brouard 12874: 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 12875: /* 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 12876: /* 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 12877: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
1.332 brouard 12878: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
12879: /* k3 is the position in the nres result line of the k1th variable of the model equation */
12880: /* Tvarsel[k3]: Name of the variable at the k3th position in the result line. */
12881: /* Tvalsel[k3]: Value of the variable at the k3th position in the result line. */
12882: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.334 brouard 12883: /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 12884: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1.330 brouard 12885: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.332 brouard 12886: k3= resultmodel[nres][k1]; /* From position k1 in model get position k3 in result line */
12887: /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
12888: 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 12889: 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 12890: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][Name]=Value; stores the value into the name of the variable. */
1.332 brouard 12891: /* Tinvresult[nres][4]=1 */
1.334 brouard 12892: /* Tresult[nres][k4+1]=Tvalsel[k3];/\* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) *\/ */
12893: Tresult[nres][k3]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
12894: /* Tvresult[nres][k4+1]=(int)Tvarsel[k3];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
12895: Tvresult[nres][k3]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
1.237 brouard 12896: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.334 brouard 12897: precov[nres][k1]=Tvalsel[k3]; /* Value from resultline of the variable at the k1 position in the model */
1.342 brouard 12898: /* 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 12899: k4++;;
1.331 brouard 12900: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Quantitative and single */
1.330 brouard 12901: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.332 brouard 12902: /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.330 brouard 12903: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
1.332 brouard 12904: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 5 =k3q */
12905: k2q=(int)Tvarsel[k3q]; /* Name of variable at k3q th position in the resultline */
12906: /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 12907: /* Tqresult[nres][k4q+1]=Tvalsel[k3q]; /\* Tqresult[nres][1]=25.1 *\/ */
12908: /* Tvresult[nres][k4q+1]=(int)Tvarsel[k3q];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
12909: /* Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /\* Tvqresult[nres][1]=5 *\/ */
12910: Tqresult[nres][k3q]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
12911: Tvresult[nres][k3q]=(int)Tvarsel[k3q];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
12912: Tvqresult[nres][k3q]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
1.237 brouard 12913: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 brouard 12914: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 12915: precov[nres][k1]=Tvalsel[k3q];
1.342 brouard 12916: /* 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 12917: k4q++;;
1.350 brouard 12918: }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"*/
12919: /* Tvar[k1]; */ /* Age variable */ /* 17 age*V6*V2 ?*/
1.332 brouard 12920: /* Wrong we want the value of variable name Tvar[k1] */
1.350 brouard 12921: if(Typevar[k1]==2 || Typevar[k1]==3 ){ /* For product quant or dummy (with or without age) */
12922: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
12923: /* 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]]); */
12924: }else{
12925: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
12926: 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)*/
12927: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][4]=1 */
12928: precov[nres][k1]=Tvalsel[k3];
12929: }
1.342 brouard 12930: /* 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 12931: }else if( Dummy[k1]==3 ){ /* For quant with age product */
1.350 brouard 12932: if(Typevar[k1]==2 || Typevar[k1]==3 ){ /* For product quant or dummy (with or without age) */
12933: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
12934: /* 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]]); */
12935: }else{
12936: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
12937: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
12938: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* TinvDoQresult[nres][5]=25.1 */
12939: precov[nres][k1]=Tvalsel[k3q];
12940: }
1.342 brouard 12941: /* 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 12942: }else if(Typevar[k1]==2 || Typevar[k1]==3 ){ /* For product quant or dummy (with or without age) */
1.332 brouard 12943: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
1.342 brouard 12944: /* 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 12945: }else{
1.332 brouard 12946: printf("Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
12947: fprintf(ficlog,"Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 12948: }
12949: }
1.234 brouard 12950:
1.334 brouard 12951: TKresult[nres]=++k; /* Number of combinations of dummies for the nresult and the model =Tvalsel[k3]*pow(2,k4) + 1*/
1.230 brouard 12952: return (0);
12953: }
1.235 brouard 12954:
1.230 brouard 12955: int decodemodel( char model[], int lastobs)
12956: /**< This routine decodes the model and returns:
1.224 brouard 12957: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
12958: * - nagesqr = 1 if age*age in the model, otherwise 0.
12959: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
12960: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
12961: * - cptcovage number of covariates with age*products =2
12962: * - cptcovs number of simple covariates
1.339 brouard 12963: * ncovcolt=ncovcol+nqv+ntv+nqtv total of covariates in the data, not in the model equation
1.224 brouard 12964: * - 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 12965: * which is a new column after the 9 (ncovcol+nqv+ntv+nqtv) variables.
1.319 brouard 12966: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 12967: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
12968: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
12969: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
12970: */
1.319 brouard 12971: /* 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 12972: {
1.359 brouard 12973: int i, j, k, ks;/* , v;*/
1.349 brouard 12974: int n,m;
12975: int j1, k1, k11, k12, k2, k3, k4;
12976: char modelsav[300];
12977: char stra[300], strb[300], strc[300], strd[300],stre[300],strf[300];
1.187 brouard 12978: char *strpt;
1.349 brouard 12979: int **existcomb;
12980:
12981: existcomb=imatrix(1,NCOVMAX,1,NCOVMAX);
12982: for(i=1;i<=NCOVMAX;i++)
12983: for(j=1;j<=NCOVMAX;j++)
12984: existcomb[i][j]=0;
12985:
1.145 brouard 12986: /*removespace(model);*/
1.136 brouard 12987: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.349 brouard 12988: j=0, j1=0, k1=0, k12=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 12989: if (strstr(model,"AGE") !=0){
1.192 brouard 12990: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
12991: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 12992: return 1;
12993: }
1.141 brouard 12994: if (strstr(model,"v") !=0){
1.338 brouard 12995: printf("Error. 'v' must be in upper case 'V' model=1+age+%s ",model);
12996: fprintf(ficlog,"Error. 'v' must be in upper case model=1+age+%s ",model);fflush(ficlog);
1.141 brouard 12997: return 1;
12998: }
1.187 brouard 12999: strcpy(modelsav,model);
13000: if ((strpt=strstr(model,"age*age")) !=0){
1.338 brouard 13001: printf(" strpt=%s, model=1+age+%s\n",strpt, model);
1.187 brouard 13002: if(strpt != model){
1.338 brouard 13003: printf("Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 13004: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 13005: corresponding column of parameters.\n",model);
1.338 brouard 13006: fprintf(ficlog,"Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 13007: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 13008: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 13009: return 1;
1.225 brouard 13010: }
1.187 brouard 13011: nagesqr=1;
13012: if (strstr(model,"+age*age") !=0)
1.234 brouard 13013: substrchaine(modelsav, model, "+age*age");
1.187 brouard 13014: else if (strstr(model,"age*age+") !=0)
1.234 brouard 13015: substrchaine(modelsav, model, "age*age+");
1.187 brouard 13016: else
1.234 brouard 13017: substrchaine(modelsav, model, "age*age");
1.187 brouard 13018: }else
13019: nagesqr=0;
1.349 brouard 13020: 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 13021: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
13022: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.351 brouard 13023: cptcovs=0; /**< Number of simple covariates V1 +V1*age +V3 +V3*V4 +age*age => V1 + V3 =4+1-3=2 Wrong */
1.187 brouard 13024: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 13025: * cst, age and age*age
13026: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
13027: /* including age products which are counted in cptcovage.
13028: * but the covariates which are products must be treated
13029: * separately: ncovn=4- 2=2 (V1+V3). */
1.349 brouard 13030: cptcovprod=0; /**< Number of products V1*V2 +v3*age = 2 */
13031: cptcovdageprod=0; /* Number of doouble products with age age*Vn*VM or Vn*age*Vm or Vn*Vm*age */
1.187 brouard 13032: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.349 brouard 13033: cptcovprodage=0;
13034: /* cptcovprodage=nboccstr(modelsav,"age");*/
1.225 brouard 13035:
1.187 brouard 13036: /* Design
13037: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
13038: * < ncovcol=8 >
13039: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
13040: * k= 1 2 3 4 5 6 7 8
13041: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
1.345 brouard 13042: * covar[k,i], are for fixed covariates, value of kth covariate if not including age for individual i:
1.224 brouard 13043: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
13044: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 13045: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
13046: * Tage[++cptcovage]=k
1.345 brouard 13047: * if products, new covar are created after ncovcol + nqv (quanti fixed) with k1
1.187 brouard 13048: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
13049: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
13050: * 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
13051: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
13052: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
13053: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
1.345 brouard 13054: * < ncovcol=8 8 fixed covariate. Additional starts at 9 (V5*V6) and 10(V7*V8) >
1.187 brouard 13055: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
13056: * k= 1 2 3 4 5 6 7 8 9 10 11 12
1.345 brouard 13057: * Tvard[k]= 2 1 3 3 10 11 8 8 5 6 7 8
13058: * p Tvar[1]@12={2, 1, 3, 3, 9, 10, 8, 8}
1.187 brouard 13059: * p Tprod[1]@2={ 6, 5}
13060: *p Tvard[1][1]@4= {7, 8, 5, 6}
13061: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
13062: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 13063: *How to reorganize? Tvars(orted)
1.187 brouard 13064: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
13065: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
13066: * {2, 1, 4, 8, 5, 6, 3, 7}
13067: * Struct []
13068: */
1.225 brouard 13069:
1.187 brouard 13070: /* This loop fills the array Tvar from the string 'model'.*/
13071: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
13072: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
13073: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
13074: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
13075: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
13076: /* k=1 Tvar[1]=2 (from V2) */
13077: /* k=5 Tvar[5] */
13078: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 13079: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 13080: /* } */
1.198 brouard 13081: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 13082: /*
13083: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 13084: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
13085: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
13086: }
1.187 brouard 13087: cptcovage=0;
1.351 brouard 13088:
13089: /* First loop in order to calculate */
13090: /* for age*VN*Vm
13091: * Provides, Typevar[k], Tage[cptcovage], existcomb[n][m], FixedV[ncovcolt+k12]
13092: * Tprod[k1]=k Tposprod[k]=k1; Tvard[k1][1] =m;
13093: */
13094: /* Needs FixedV[Tvardk[k][1]] */
13095: /* For others:
13096: * Sets Typevar[k];
13097: * Tvar[k]=ncovcol+nqv+ntv+nqtv+k11;
13098: * Tposprod[k]=k11;
13099: * Tprod[k11]=k;
13100: * Tvardk[k][1] =m;
13101: * Needs FixedV[Tvardk[k][1]] == 0
13102: */
13103:
1.319 brouard 13104: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
13105: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
13106: 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" */
13107: if (nbocc(modelsav,'+')==0)
13108: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 13109: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
13110: /*scanf("%d",i);*/
1.349 brouard 13111: 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 */
13112: 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 */
13113: 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 */
13114: Typevar[k]=3; /* 3 for age and double product age*Vn*Vm varying of fixed */
13115: if(strstr(strc,"age")!=0) { /* It means that strc=V2*age or age*V2 and thus that strd=Vn */
13116: cutl(stre,strf,strc,'*') ; /* strf=age or Vm, stre=Vm or age. If strc=V6*V2 then strf=V6 and stre=V2 */
13117: strcpy(strc,strb); /* save strb(=age*Vn*Vm) into strc */
13118: /* We want strb=Vn*Vm */
13119: if(strcmp(strf,"age")==0){ /* strf is "age" so that stre=Vm =V2 . */
13120: strcpy(strb,strd);
13121: strcat(strb,"*");
13122: strcat(strb,stre);
13123: }else{ /* strf=Vm If strf=V6 then stre=V2 */
13124: strcpy(strb,strf);
13125: strcat(strb,"*");
13126: strcat(strb,stre);
13127: strcpy(strd,strb); /* in order for strd to not be "age" for next test (will be Vn*Vm */
13128: }
1.351 brouard 13129: /* 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]]]); */
13130: /* FixedV[Tvar[Tage[k]]]=0; /\* HERY not sure if V7*V4*age Fixed might not exist yet*\/ */
1.349 brouard 13131: }else{ /* strc=Vn*Vm (and strd=age) and should be strb=Vn*Vm but want to keep original strb double product */
13132: strcpy(stre,strb); /* save full b in stre */
13133: strcpy(strb,strc); /* save short c in new short b for next block strb=Vn*Vm*/
13134: strcpy(strf,strc); /* save short c in new short f */
13135: cutl(strc,strd,strf,'*'); /* We get strd=Vn and strc=Vm for next block (strb=Vn*Vm)*/
13136: /* strcpy(strc,stre);*/ /* save full e in c for future */
13137: }
13138: cptcovdageprod++; /* double product with age Which product is it? */
13139: /* 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 *\/ */
13140: /* cutl(strc,strd,strb,'*'); /\* strd= V6 or V2 or age and strc= V2 or age or V2 *\/ */
1.234 brouard 13141: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
1.349 brouard 13142: n=atoi(stre);
1.234 brouard 13143: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
1.349 brouard 13144: m=atoi(strc);
13145: cptcovage++; /* Counts the number of covariates which include age as a product */
13146: Tage[cptcovage]=k; /* For age*V3*V2 gives the position in model of covariates associated with age Tage[1]=6 HERY too*/
13147: if(existcomb[n][m] == 0){
13148: /* r /home/brouard/Documents/Recherches/REVES/Zachary/Zach-2022/Feinuo_Sun/Feinuo-threeway/femV12V15_3wayintNBe.imach */
13149: 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);
13150: 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);
13151: fflush(ficlog);
13152: k1++; /* The combination Vn*Vm will be in the model so we create it at k1 */
13153: k12++;
13154: existcomb[n][m]=k1;
13155: existcomb[m][n]=k1;
13156: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1;
13157: 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*/
13158: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 Gives the k1 double product Vn*Vm or age*Vn*Vm at the k position */
13159: Tvard[k1][1] =m; /* m 1 for V1*/
13160: Tvardk[k][1] =m; /* m 1 for V1*/
13161: Tvard[k1][2] =n; /* n 4 for V4*/
13162: Tvardk[k][2] =n; /* n 4 for V4*/
1.351 brouard 13163: /* Tvar[Tage[cptcovage]]=k1;*/ /* Tvar[6=age*V3*V2]=9 (new fixed covariate) */ /* We don't know about Fixed yet HERE */
1.349 brouard 13164: 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 */
13165: for (i=1; i<=lastobs;i++){/* For fixed product */
13166: /* Computes the new covariate which is a product of
13167: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
13168: covar[ncovcolt+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
13169: }
13170: cptcovprodage++; /* Counting the number of fixed covariate with age */
13171: FixedV[ncovcolt+k12]=0; /* We expand Vn*Vm */
13172: k12++;
13173: FixedV[ncovcolt+k12]=0;
13174: }else{ /*End of FixedV */
13175: cptcovprodvage++; /* Counting the number of varying covariate with age */
13176: FixedV[ncovcolt+k12]=1; /* We expand Vn*Vm */
13177: k12++;
13178: FixedV[ncovcolt+k12]=1;
13179: }
13180: }else{ /* k1 Vn*Vm already exists */
13181: k11=existcomb[n][m];
13182: Tposprod[k]=k11; /* OK */
13183: Tvar[k]=Tvar[Tprod[k11]]; /* HERY */
13184: Tvardk[k][1]=m;
13185: Tvardk[k][2]=n;
13186: 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 */
13187: /*cptcovage++;*/ /* Counts the number of covariates which include age as a product */
13188: cptcovprodage++; /* Counting the number of fixed covariate with age */
13189: /*Tage[cptcovage]=k;*/ /* For age*V3*V2 Tage[1]=V3*V3=9 HERY too*/
13190: Tvar[Tage[cptcovage]]=k1;
13191: FixedV[ncovcolt+k12]=0; /* We expand Vn*Vm */
13192: k12++;
13193: FixedV[ncovcolt+k12]=0;
13194: }else{ /* Already exists but time varying (and age) */
13195: /*cptcovage++;*/ /* Counts the number of covariates which include age as a product */
13196: /*Tage[cptcovage]=k;*/ /* For age*V3*V2 Tage[1]=V3*V3=9 HERY too*/
13197: /* Tvar[Tage[cptcovage]]=k1; */
13198: cptcovprodvage++;
13199: FixedV[ncovcolt+k12]=1; /* We expand Vn*Vm */
13200: k12++;
13201: FixedV[ncovcolt+k12]=1;
13202: }
13203: }
13204: /* Tage[cptcovage]=k; /\* V2+V1+V4+V3*age Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
13205: /* Tvar[k]=k11; /\* HERY *\/ */
13206: } 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 */
13207: cptcovprod++;
13208: if (strcmp(strc,"age")==0) { /**< Model includes age: strb= Vn*age c=age d=Vn*/
13209: /* covar is not filled and then is empty */
13210: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
13211: 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 */
13212: Typevar[k]=1; /* 1 for age product */
13213: cptcovage++; /* Counts the number of covariates which include age as a product */
13214: Tage[cptcovage]=k; /* V2+V1+V4+V3*age Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
13215: if( FixedV[Tvar[k]] == 0){
13216: cptcovprodage++; /* Counting the number of fixed covariate with age */
13217: }else{
13218: cptcovprodvage++; /* Counting the number of fixedvarying covariate with age */
13219: }
13220: /*printf("stre=%s ", stre);*/
13221: } else if (strcmp(strd,"age")==0) { /* strb= age*Vn c=Vn */
13222: cutl(stre,strb,strc,'V');
13223: Tvar[k]=atoi(stre);
13224: Typevar[k]=1; /* 1 for age product */
13225: cptcovage++;
13226: Tage[cptcovage]=k;
13227: if( FixedV[Tvar[k]] == 0){
13228: cptcovprodage++; /* Counting the number of fixed covariate with age */
13229: }else{
13230: cptcovprodvage++; /* Counting the number of fixedvarying covariate with age */
1.339 brouard 13231: }
1.349 brouard 13232: }else{ /* for product Vn*Vm */
13233: Typevar[k]=2; /* 2 for product Vn*Vm */
13234: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
13235: n=atoi(stre);
13236: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
13237: m=atoi(strc);
13238: k1++;
13239: cptcovprodnoage++;
13240: if(existcomb[n][m] != 0 || existcomb[m][n] != 0){
13241: printf("Warning in model combination V%d*V%d already exists in the model in position k1=%d!\n",n,m,existcomb[n][m]);
13242: 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]);
13243: fflush(ficlog);
13244: k11=existcomb[n][m];
13245: Tvar[k]=ncovcol+nqv+ntv+nqtv+k11;
13246: Tposprod[k]=k11;
13247: Tprod[k11]=k;
13248: Tvardk[k][1] =m; /* m 1 for V1*/
13249: /* Tvard[k11][1] =m; /\* n 4 for V4*\/ */
13250: Tvardk[k][2] =n; /* n 4 for V4*/
13251: /* Tvard[k11][2] =n; /\* n 4 for V4*\/ */
13252: }else{ /* combination Vn*Vm doesn't exist we create it (no age)*/
13253: existcomb[n][m]=k1;
13254: existcomb[m][n]=k1;
13255: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* ncovcolt+k1; For model-covariate k tells which data-covariate to use but
13256: because this model-covariate is a construction we invent a new column
13257: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
13258: If already ncovcol=4 and model= V2 + V1 + V1*V4 + age*V3 + V3*V2
13259: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
13260: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=3 etc */
13261: /* Please remark that the new variables are model dependent */
13262: /* If we have 4 variable but the model uses only 3, like in
13263: * model= V1 + age*V1 + V2 + V3 + age*V2 + age*V3 + V1*V2 + V1*V3
13264: * k= 1 2 3 4 5 6 7 8
13265: * Tvar[k]=1 1 2 3 2 3 (5 6) (and not 4 5 because of V4 missing)
13266: * Tage[kk] [1]= 2 [2]=5 [3]=6 kk=1 to cptcovage=3
13267: * Tvar[Tage[kk]][1]=2 [2]=2 [3]=3
13268: */
13269: /* We need to feed some variables like TvarVV, but later on next loop because of ncovv (k2) is not correct */
13270: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 +V6*V2*age */
13271: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
13272: Tvard[k1][1] =m; /* m 1 for V1*/
13273: Tvardk[k][1] =m; /* m 1 for V1*/
13274: Tvard[k1][2] =n; /* n 4 for V4*/
13275: Tvardk[k][2] =n; /* n 4 for V4*/
13276: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
13277: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
13278: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
13279: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
13280: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
13281: 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 */
13282: for (i=1; i<=lastobs;i++){/* For fixed product */
13283: /* Computes the new covariate which is a product of
13284: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
13285: covar[ncovcolt+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
13286: }
13287: /* TvarVV[k2]=n; */
13288: /* FixedV[ncovcolt+k2]=0; /\* or FixedV[Tvar[k]]=0; FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13289: /* TvarVV[k2+1]=m; */
13290: /* FixedV[ncovcolt+k2]=0; /\* or FixedV[Tvar[k]]=0; FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13291: }else{ /* not FixedV */
13292: /* TvarVV[k2]=n; */
13293: /* FixedV[ncovcolt+k2]=0; /\* or FixedV[Tvar[k]]=0; FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13294: /* TvarVV[k2+1]=m; */
13295: /* FixedV[ncovcolt+k2]=0; /\* or FixedV[Tvar[k]]=0; FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13296: }
13297: } /* End of creation of Vn*Vm if not created by age*Vn*Vm earlier */
13298: } /* End of product Vn*Vm */
13299: } /* End of age*double product or simple product */
13300: }else { /* not a product */
1.234 brouard 13301: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
13302: /* scanf("%d",i);*/
13303: cutl(strd,strc,strb,'V');
13304: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
13305: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
13306: Tvar[k]=atoi(strd);
13307: Typevar[k]=0; /* 0 for simple covariates */
13308: }
13309: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 13310: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 13311: scanf("%d",i);*/
1.187 brouard 13312: } /* end of loop + on total covariates */
1.351 brouard 13313:
13314:
1.187 brouard 13315: } /* end if strlen(modelsave == 0) age*age might exist */
13316: } /* end if strlen(model == 0) */
1.349 brouard 13317: 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 */
13318:
1.136 brouard 13319: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
13320: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 13321:
1.136 brouard 13322: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 13323: printf("cptcovprod=%d ", cptcovprod);
13324: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
13325: scanf("%d ",i);*/
13326:
13327:
1.230 brouard 13328: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
13329: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 13330: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
13331: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
13332: k = 1 2 3 4 5 6 7 8 9
13333: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 13334: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 13335: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
13336: Dummy[k] 1 0 0 0 3 1 1 2 3
13337: Tmodelind[combination of covar]=k;
1.225 brouard 13338: */
13339: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 13340: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 13341: /* 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 13342: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 13343: printf("Model=1+age+%s\n\
1.349 brouard 13344: 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 13345: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
13346: 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 13347: fprintf(ficlog,"Model=1+age+%s\n\
1.349 brouard 13348: 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 13349: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
13350: 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 13351: for(k=-1;k<=NCOVMAX; k++){ Fixed[k]=0; Dummy[k]=0;}
13352: for(k=1;k<=NCOVMAX; k++){TvarFind[k]=0; TvarVind[k]=0;}
1.351 brouard 13353:
13354:
13355: /* Second loop for calculating Fixed[k], Dummy[k]*/
13356:
13357:
1.349 brouard 13358: 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 13359: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 13360: Fixed[k]= 0;
13361: Dummy[k]= 0;
1.225 brouard 13362: ncoveff++;
1.232 brouard 13363: ncovf++;
1.234 brouard 13364: nsd++;
13365: modell[k].maintype= FTYPE;
13366: TvarsD[nsd]=Tvar[k];
13367: TvarsDind[nsd]=k;
1.330 brouard 13368: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 13369: TvarF[ncovf]=Tvar[k];
13370: TvarFind[ncovf]=k;
13371: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
13372: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.339 brouard 13373: /* }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 13374: }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 13375: Fixed[k]= 0;
13376: Dummy[k]= 1;
1.230 brouard 13377: nqfveff++;
1.234 brouard 13378: modell[k].maintype= FTYPE;
13379: modell[k].subtype= FQ;
13380: nsq++;
1.334 brouard 13381: TvarsQ[nsq]=Tvar[k]; /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary see below */
13382: TvarsQind[nsq]=k; /* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.232 brouard 13383: ncovf++;
1.234 brouard 13384: TvarF[ncovf]=Tvar[k];
13385: TvarFind[ncovf]=k;
1.231 brouard 13386: 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 13387: 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 13388: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.339 brouard 13389: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
13390: /* model V1+V3+age*V1+age*V3+V1*V3 */
13391: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
13392: ncovvt++;
13393: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
13394: TvarVVind[ncovvt]=k; /* TvarVVind[1]=2 (second position in the model equation */
13395:
1.227 brouard 13396: Fixed[k]= 1;
13397: Dummy[k]= 0;
1.225 brouard 13398: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 13399: modell[k].maintype= VTYPE;
13400: modell[k].subtype= VD;
13401: nsd++;
13402: TvarsD[nsd]=Tvar[k];
13403: TvarsDind[nsd]=k;
1.330 brouard 13404: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 13405: ncovv++; /* Only simple time varying variables */
13406: TvarV[ncovv]=Tvar[k];
1.242 brouard 13407: 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 13408: 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 */
13409: 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 13410: 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);
13411: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 13412: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.339 brouard 13413: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
13414: /* model V1+V3+age*V1+age*V3+V1*V3 */
13415: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
13416: ncovvt++;
13417: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
13418: TvarVVind[ncovvt]=k; /* TvarVV[1]=V3 (first time varying in the model equation */
13419:
1.234 brouard 13420: Fixed[k]= 1;
13421: Dummy[k]= 1;
13422: nqtveff++;
13423: modell[k].maintype= VTYPE;
13424: modell[k].subtype= VQ;
13425: ncovv++; /* Only simple time varying variables */
13426: nsq++;
1.334 brouard 13427: 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) */
13428: 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 13429: TvarV[ncovv]=Tvar[k];
1.242 brouard 13430: 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 13431: 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 */
13432: 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 13433: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
13434: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
1.349 brouard 13435: /* 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 13436: /* printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv); */
1.227 brouard 13437: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 13438: ncova++;
13439: TvarA[ncova]=Tvar[k];
13440: TvarAind[ncova]=k;
1.349 brouard 13441: /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
13442: /** 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 13443: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 13444: Fixed[k]= 2;
13445: Dummy[k]= 2;
13446: modell[k].maintype= ATYPE;
13447: modell[k].subtype= APFD;
1.349 brouard 13448: ncovta++;
13449: TvarAVVA[ncovta]=Tvar[k]; /* (2)age*V3 */
13450: TvarAVVAind[ncovta]=k;
1.240 brouard 13451: /* ncoveff++; */
1.227 brouard 13452: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 13453: Fixed[k]= 2;
13454: Dummy[k]= 3;
13455: modell[k].maintype= ATYPE;
13456: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
1.349 brouard 13457: ncovta++;
13458: TvarAVVA[ncovta]=Tvar[k]; /* */
13459: TvarAVVAind[ncovta]=k;
1.240 brouard 13460: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 13461: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 13462: Fixed[k]= 3;
13463: Dummy[k]= 2;
13464: modell[k].maintype= ATYPE;
13465: modell[k].subtype= APVD; /* Product age * varying dummy */
1.349 brouard 13466: ncovva++;
13467: TvarVVA[ncovva]=Tvar[k]; /* (1)+age*V6 + (2)age*V7 */
13468: TvarVVAind[ncovva]=k;
13469: ncovta++;
13470: TvarAVVA[ncovta]=Tvar[k]; /* */
13471: TvarAVVAind[ncovta]=k;
1.240 brouard 13472: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 13473: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 13474: Fixed[k]= 3;
13475: Dummy[k]= 3;
13476: modell[k].maintype= ATYPE;
13477: modell[k].subtype= APVQ; /* Product age * varying quantitative */
1.349 brouard 13478: ncovva++;
13479: TvarVVA[ncovva]=Tvar[k]; /* */
13480: TvarVVAind[ncovva]=k;
13481: ncovta++;
13482: TvarAVVA[ncovta]=Tvar[k]; /* (1)+age*V6 + (2)age*V7 */
13483: TvarAVVAind[ncovta]=k;
1.240 brouard 13484: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 13485: }
1.349 brouard 13486: }else if( Tposprod[k]>0 && Typevar[k]==2){ /* Detects if fixed product no age Vm*Vn */
13487: printf("MEMORY ERRORR k=%d Tposprod[k]=%d, Typevar[k]=%d\n ",k, Tposprod[k], Typevar[k]);
13488: 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 */
13489: 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]]);
13490: Fixed[k]= 0;
13491: Dummy[k]= 0;
13492: ncoveff++;
13493: ncovf++;
13494: /* ncovv++; */
13495: /* TvarVV[ncovv]=Tvardk[k][1]; */
13496: /* FixedV[ncovcolt+ncovv]=0; /\* or FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13497: /* ncovv++; */
13498: /* TvarVV[ncovv]=Tvardk[k][2]; */
13499: /* FixedV[ncovcolt+ncovv]=0; /\* or FixedV[TvarVV[ncovv]]=0 HERE *\/ */
13500: modell[k].maintype= FTYPE;
13501: TvarF[ncovf]=Tvar[k];
13502: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
13503: TvarFind[ncovf]=k;
13504: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
13505: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
13506: }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 */
13507: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
13508: /* model V1+V3+age*V1+age*V3+V1*V3 + V1*V3*age*/
13509: /* Tvar={1, 3, 1, 3, 6, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
13510: 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 */
13511: ncovvt++;
13512: TvarVV[ncovvt]=Tvard[k1][1]; /* TvarVV[2]=V1 (because TvarVV[1] was V3, first time varying covariates */
13513: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
13514: ncovvt++;
13515: TvarVV[ncovvt]=Tvard[k1][2]; /* TvarVV[3]=V3 */
13516: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
13517:
13518: /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
13519: /** Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product */
13520:
13521: if(Tvard[k1][1] <=ncovcol){ /* Vn is dummy fixed, (Tvard[1][1]=V1), (Tvard[1][1]=V3 time varying) */
13522: if(Tvard[k1][2] <=ncovcol){ /* Vm is dummy fixed */
13523: Fixed[k]= 1;
13524: Dummy[k]= 0;
13525: modell[k].maintype= FTYPE;
13526: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
13527: ncovf++; /* Fixed variables without age */
13528: TvarF[ncovf]=Tvar[k];
13529: TvarFind[ncovf]=k;
13530: }else if(Tvard[k1][2] <=ncovcol+nqv){ /* Vm is quanti fixed */
13531: Fixed[k]= 0; /* Fixed product */
13532: Dummy[k]= 1;
13533: modell[k].maintype= FTYPE;
13534: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
13535: ncovf++; /* Varying variables without age */
13536: TvarF[ncovf]=Tvar[k];
13537: TvarFind[ncovf]=k;
13538: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is a time varying dummy covariate */
13539: Fixed[k]= 1;
13540: Dummy[k]= 0;
13541: modell[k].maintype= VTYPE;
13542: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
13543: ncovv++; /* Varying variables without age */
13544: TvarV[ncovv]=Tvar[k]; /* TvarV[1]=Tvar[5]=5 because there is a V4 */
13545: TvarVind[ncovv]=k;/* TvarVind[1]=5 */
13546: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is a time varying quantitative covariate */
13547: Fixed[k]= 1;
13548: Dummy[k]= 1;
13549: modell[k].maintype= VTYPE;
13550: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
13551: ncovv++; /* Varying variables without age */
13552: TvarV[ncovv]=Tvar[k];
13553: TvarVind[ncovv]=k;
13554: }
13555: }else if(Tvard[k1][1] <=ncovcol+nqv){ /* Vn is fixed quanti */
13556: if(Tvard[k1][2] <=ncovcol){ /* Vm is fixed dummy */
13557: Fixed[k]= 0; /* Fixed product */
13558: Dummy[k]= 1;
13559: modell[k].maintype= FTYPE;
13560: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
13561: ncovf++; /* Fixed variables without age */
13562: TvarF[ncovf]=Tvar[k];
13563: TvarFind[ncovf]=k;
13564: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is time varying */
13565: Fixed[k]= 1;
13566: Dummy[k]= 1;
13567: modell[k].maintype= VTYPE;
13568: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
13569: ncovv++; /* Varying variables without age */
13570: TvarV[ncovv]=Tvar[k];
13571: TvarVind[ncovv]=k;
13572: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is time varying quanti */
13573: Fixed[k]= 1;
13574: Dummy[k]= 1;
13575: modell[k].maintype= VTYPE;
13576: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
13577: ncovv++; /* Varying variables without age */
13578: TvarV[ncovv]=Tvar[k];
13579: TvarVind[ncovv]=k;
13580: ncovv++; /* Varying variables without age */
13581: TvarV[ncovv]=Tvar[k];
13582: TvarVind[ncovv]=k;
13583: }
13584: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){ /* Vn is time varying dummy */
13585: if(Tvard[k1][2] <=ncovcol){
13586: Fixed[k]= 1;
13587: Dummy[k]= 1;
13588: modell[k].maintype= VTYPE;
13589: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
13590: ncovv++; /* Varying variables without age */
13591: TvarV[ncovv]=Tvar[k];
13592: TvarVind[ncovv]=k;
13593: }else if(Tvard[k1][2] <=ncovcol+nqv){
13594: Fixed[k]= 1;
13595: Dummy[k]= 1;
13596: modell[k].maintype= VTYPE;
13597: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
13598: ncovv++; /* Varying variables without age */
13599: TvarV[ncovv]=Tvar[k];
13600: TvarVind[ncovv]=k;
13601: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
13602: Fixed[k]= 1;
13603: Dummy[k]= 0;
13604: modell[k].maintype= VTYPE;
13605: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
13606: ncovv++; /* Varying variables without age */
13607: TvarV[ncovv]=Tvar[k];
13608: TvarVind[ncovv]=k;
13609: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
13610: Fixed[k]= 1;
13611: Dummy[k]= 1;
13612: modell[k].maintype= VTYPE;
13613: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
13614: ncovv++; /* Varying variables without age */
13615: TvarV[ncovv]=Tvar[k];
13616: TvarVind[ncovv]=k;
13617: }
13618: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){ /* Vn is time varying quanti */
13619: if(Tvard[k1][2] <=ncovcol){
13620: Fixed[k]= 1;
13621: Dummy[k]= 1;
13622: modell[k].maintype= VTYPE;
13623: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
13624: ncovv++; /* Varying variables without age */
13625: TvarV[ncovv]=Tvar[k];
13626: TvarVind[ncovv]=k;
13627: }else if(Tvard[k1][2] <=ncovcol+nqv){
13628: Fixed[k]= 1;
13629: Dummy[k]= 1;
13630: modell[k].maintype= VTYPE;
13631: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
13632: ncovv++; /* Varying variables without age */
13633: TvarV[ncovv]=Tvar[k];
13634: TvarVind[ncovv]=k;
13635: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
13636: Fixed[k]= 1;
13637: Dummy[k]= 1;
13638: modell[k].maintype= VTYPE;
13639: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
13640: ncovv++; /* Varying variables without age */
13641: TvarV[ncovv]=Tvar[k];
13642: TvarVind[ncovv]=k;
13643: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
13644: Fixed[k]= 1;
13645: Dummy[k]= 1;
13646: modell[k].maintype= VTYPE;
13647: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
13648: ncovv++; /* Varying variables without age */
13649: TvarV[ncovv]=Tvar[k];
13650: TvarVind[ncovv]=k;
13651: }
13652: }else{
13653: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
13654: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
13655: } /*end k1*/
13656: }
13657: }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 13658: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
1.349 brouard 13659: /* model V1+V3+age*V1+age*V3+V1*V3 + V1*V3*age*/
13660: /* Tvar={1, 3, 1, 3, 6, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
13661: 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 */
13662: ncova++;
13663: TvarA[ncova]=Tvard[k1][1]; /* TvarVV[2]=V1 (because TvarVV[1] was V3, first time varying covariates */
13664: TvarAind[ncova]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
13665: ncova++;
13666: TvarA[ncova]=Tvard[k1][2]; /* TvarVV[3]=V3 */
13667: TvarAind[ncova]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
1.339 brouard 13668:
1.349 brouard 13669: /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
13670: /** Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product */
13671: if( FixedV[Tvardk[k][1]] == 0 && FixedV[Tvardk[k][2]] == 0){
13672: ncovta++;
13673: TvarAVVA[ncovta]=Tvard[k1][1]; /* age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 */
13674: TvarAVVAind[ncovta]=k;
13675: ncovta++;
13676: TvarAVVA[ncovta]=Tvard[k1][2]; /* age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 */
13677: TvarAVVAind[ncovta]=k;
13678: }else{
13679: ncovva++; /* HERY reached */
13680: TvarVVA[ncovva]=Tvard[k1][1]; /* age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 */
13681: TvarVVAind[ncovva]=k;
13682: ncovva++;
13683: TvarVVA[ncovva]=Tvard[k1][2]; /* */
13684: TvarVVAind[ncovva]=k;
13685: ncovta++;
13686: TvarAVVA[ncovta]=Tvard[k1][1]; /* age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 */
13687: TvarAVVAind[ncovta]=k;
13688: ncovta++;
13689: TvarAVVA[ncovta]=Tvard[k1][2]; /* age*V6*V3 +age*V7*V3 + age*V6*V4 +age*V7*V4 */
13690: TvarAVVAind[ncovta]=k;
13691: }
1.339 brouard 13692: if(Tvard[k1][1] <=ncovcol){ /* Vn is dummy fixed, (Tvard[1][1]=V1), (Tvard[1][1]=V3 time varying) */
13693: if(Tvard[k1][2] <=ncovcol){ /* Vm is dummy fixed */
1.349 brouard 13694: Fixed[k]= 2;
13695: Dummy[k]= 2;
1.240 brouard 13696: modell[k].maintype= FTYPE;
13697: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
1.349 brouard 13698: /* TvarF[ncova]=Tvar[k]; /\* Problem to solve *\/ */
13699: /* TvarFind[ncova]=k; */
1.339 brouard 13700: }else if(Tvard[k1][2] <=ncovcol+nqv){ /* Vm is quanti fixed */
1.349 brouard 13701: Fixed[k]= 2; /* Fixed product */
13702: Dummy[k]= 3;
1.240 brouard 13703: modell[k].maintype= FTYPE;
13704: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
1.349 brouard 13705: /* TvarF[ncova]=Tvar[k]; */
13706: /* TvarFind[ncova]=k; */
1.339 brouard 13707: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is a time varying dummy covariate */
1.349 brouard 13708: Fixed[k]= 3;
13709: Dummy[k]= 2;
1.240 brouard 13710: modell[k].maintype= VTYPE;
13711: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
1.349 brouard 13712: TvarV[ncova]=Tvar[k]; /* TvarV[1]=Tvar[5]=5 because there is a V4 */
13713: TvarVind[ncova]=k;/* TvarVind[1]=5 */
1.339 brouard 13714: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is a time varying quantitative covariate */
1.349 brouard 13715: Fixed[k]= 3;
13716: Dummy[k]= 3;
1.240 brouard 13717: modell[k].maintype= VTYPE;
13718: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
1.349 brouard 13719: /* ncovv++; /\* Varying variables without age *\/ */
13720: /* TvarV[ncovv]=Tvar[k]; */
13721: /* TvarVind[ncovv]=k; */
1.240 brouard 13722: }
1.339 brouard 13723: }else if(Tvard[k1][1] <=ncovcol+nqv){ /* Vn is fixed quanti */
13724: if(Tvard[k1][2] <=ncovcol){ /* Vm is fixed dummy */
1.349 brouard 13725: Fixed[k]= 2; /* Fixed product */
13726: Dummy[k]= 2;
1.240 brouard 13727: modell[k].maintype= FTYPE;
13728: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
1.349 brouard 13729: /* ncova++; /\* Fixed variables with age *\/ */
13730: /* TvarF[ncovf]=Tvar[k]; */
13731: /* TvarFind[ncovf]=k; */
1.339 brouard 13732: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is time varying */
1.349 brouard 13733: Fixed[k]= 2;
13734: Dummy[k]= 3;
1.240 brouard 13735: modell[k].maintype= VTYPE;
13736: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
1.349 brouard 13737: /* ncova++; /\* Varying variables with age *\/ */
13738: /* TvarV[ncova]=Tvar[k]; */
13739: /* TvarVind[ncova]=k; */
1.339 brouard 13740: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is time varying quanti */
1.349 brouard 13741: Fixed[k]= 3;
13742: Dummy[k]= 2;
1.240 brouard 13743: modell[k].maintype= VTYPE;
13744: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
1.349 brouard 13745: ncova++; /* Varying variables without age */
13746: TvarV[ncova]=Tvar[k];
13747: TvarVind[ncova]=k;
13748: /* ncova++; /\* Varying variables without age *\/ */
13749: /* TvarV[ncova]=Tvar[k]; */
13750: /* TvarVind[ncova]=k; */
1.240 brouard 13751: }
1.339 brouard 13752: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){ /* Vn is time varying dummy */
1.240 brouard 13753: if(Tvard[k1][2] <=ncovcol){
1.349 brouard 13754: Fixed[k]= 2;
13755: Dummy[k]= 2;
1.240 brouard 13756: modell[k].maintype= VTYPE;
13757: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
1.349 brouard 13758: /* ncova++; /\* Varying variables with age *\/ */
13759: /* TvarV[ncova]=Tvar[k]; */
13760: /* TvarVind[ncova]=k; */
1.240 brouard 13761: }else if(Tvard[k1][2] <=ncovcol+nqv){
1.349 brouard 13762: Fixed[k]= 2;
13763: Dummy[k]= 3;
1.240 brouard 13764: modell[k].maintype= VTYPE;
13765: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
1.349 brouard 13766: /* ncova++; /\* Varying variables with age *\/ */
13767: /* TvarV[ncova]=Tvar[k]; */
13768: /* TvarVind[ncova]=k; */
1.240 brouard 13769: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
1.349 brouard 13770: Fixed[k]= 3;
13771: Dummy[k]= 2;
1.240 brouard 13772: modell[k].maintype= VTYPE;
13773: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
1.349 brouard 13774: /* ncova++; /\* Varying variables with age *\/ */
13775: /* TvarV[ncova]=Tvar[k]; */
13776: /* TvarVind[ncova]=k; */
1.240 brouard 13777: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
1.349 brouard 13778: Fixed[k]= 3;
13779: Dummy[k]= 3;
1.240 brouard 13780: modell[k].maintype= VTYPE;
13781: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
1.349 brouard 13782: /* ncova++; /\* Varying variables with age *\/ */
13783: /* TvarV[ncova]=Tvar[k]; */
13784: /* TvarVind[ncova]=k; */
1.240 brouard 13785: }
1.339 brouard 13786: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){ /* Vn is time varying quanti */
1.240 brouard 13787: if(Tvard[k1][2] <=ncovcol){
1.349 brouard 13788: Fixed[k]= 2;
13789: Dummy[k]= 2;
1.240 brouard 13790: modell[k].maintype= VTYPE;
13791: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
1.349 brouard 13792: /* ncova++; /\* Varying variables with age *\/ */
13793: /* TvarV[ncova]=Tvar[k]; */
13794: /* TvarVind[ncova]=k; */
1.240 brouard 13795: }else if(Tvard[k1][2] <=ncovcol+nqv){
1.349 brouard 13796: Fixed[k]= 2;
13797: Dummy[k]= 3;
1.240 brouard 13798: modell[k].maintype= VTYPE;
13799: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
1.349 brouard 13800: /* ncova++; /\* Varying variables with age *\/ */
13801: /* TvarV[ncova]=Tvar[k]; */
13802: /* TvarVind[ncova]=k; */
1.240 brouard 13803: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
1.349 brouard 13804: Fixed[k]= 3;
13805: Dummy[k]= 2;
1.240 brouard 13806: modell[k].maintype= VTYPE;
13807: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
1.349 brouard 13808: /* ncova++; /\* Varying variables with age *\/ */
13809: /* TvarV[ncova]=Tvar[k]; */
13810: /* TvarVind[ncova]=k; */
1.240 brouard 13811: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
1.349 brouard 13812: Fixed[k]= 3;
13813: Dummy[k]= 3;
1.240 brouard 13814: modell[k].maintype= VTYPE;
13815: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
1.349 brouard 13816: /* ncova++; /\* Varying variables with age *\/ */
13817: /* TvarV[ncova]=Tvar[k]; */
13818: /* TvarVind[ncova]=k; */
1.240 brouard 13819: }
1.227 brouard 13820: }else{
1.240 brouard 13821: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
13822: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
13823: } /*end k1*/
1.349 brouard 13824: } else{
1.226 brouard 13825: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
13826: 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 13827: }
1.342 brouard 13828: /* 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]); */
13829: /* printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype); */
1.227 brouard 13830: 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]);
13831: }
1.349 brouard 13832: ncovvta=ncovva;
1.227 brouard 13833: /* Searching for doublons in the model */
13834: for(k1=1; k1<= cptcovt;k1++){
13835: for(k2=1; k2 <k1;k2++){
1.285 brouard 13836: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
13837: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 13838: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
13839: if(Tvar[k1]==Tvar[k2]){
1.338 brouard 13840: 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]);
13841: 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 13842: return(1);
13843: }
13844: }else if (Typevar[k1] ==2){
13845: k3=Tposprod[k1];
13846: k4=Tposprod[k2];
13847: 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 13848: 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]]);
13849: 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 13850: return(1);
13851: }
13852: }
1.227 brouard 13853: }
13854: }
1.225 brouard 13855: }
13856: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
13857: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 13858: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
13859: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.349 brouard 13860:
13861: free_imatrix(existcomb,1,NCOVMAX,1,NCOVMAX);
1.137 brouard 13862: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 13863: /*endread:*/
1.225 brouard 13864: printf("Exiting decodemodel: ");
13865: return (1);
1.136 brouard 13866: }
13867:
1.169 brouard 13868: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 13869: {/* Check ages at death */
1.136 brouard 13870: int i, m;
1.218 brouard 13871: int firstone=0;
13872:
1.136 brouard 13873: for (i=1; i<=imx; i++) {
13874: for(m=2; (m<= maxwav); m++) {
13875: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
13876: anint[m][i]=9999;
1.216 brouard 13877: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
13878: s[m][i]=-1;
1.136 brouard 13879: }
13880: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 13881: *nberr = *nberr + 1;
1.218 brouard 13882: if(firstone == 0){
13883: firstone=1;
1.260 brouard 13884: 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 13885: }
1.262 brouard 13886: 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 13887: s[m][i]=-1; /* Droping the death status */
1.136 brouard 13888: }
13889: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 13890: (*nberr)++;
1.259 brouard 13891: 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 13892: 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 13893: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 13894: }
13895: }
13896: }
13897:
13898: for (i=1; i<=imx; i++) {
13899: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
13900: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 13901: 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 13902: if (s[m][i] >= nlstate+1) {
1.169 brouard 13903: if(agedc[i]>0){
13904: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 13905: agev[m][i]=agedc[i];
1.214 brouard 13906: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 13907: }else {
1.136 brouard 13908: if ((int)andc[i]!=9999){
13909: nbwarn++;
13910: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
13911: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
13912: agev[m][i]=-1;
13913: }
13914: }
1.169 brouard 13915: } /* agedc > 0 */
1.214 brouard 13916: } /* end if */
1.136 brouard 13917: else if(s[m][i] !=9){ /* Standard case, age in fractional
13918: years but with the precision of a month */
13919: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
13920: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
13921: agev[m][i]=1;
13922: else if(agev[m][i] < *agemin){
13923: *agemin=agev[m][i];
13924: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
13925: }
13926: else if(agev[m][i] >*agemax){
13927: *agemax=agev[m][i];
1.156 brouard 13928: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 13929: }
13930: /*agev[m][i]=anint[m][i]-annais[i];*/
13931: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 13932: } /* en if 9*/
1.136 brouard 13933: else { /* =9 */
1.214 brouard 13934: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 13935: agev[m][i]=1;
13936: s[m][i]=-1;
13937: }
13938: }
1.214 brouard 13939: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 13940: agev[m][i]=1;
1.214 brouard 13941: else{
13942: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
13943: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
13944: agev[m][i]=0;
13945: }
13946: } /* End for lastpass */
13947: }
1.136 brouard 13948:
13949: for (i=1; i<=imx; i++) {
13950: for(m=firstpass; (m<=lastpass); m++){
13951: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 13952: (*nberr)++;
1.136 brouard 13953: 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);
13954: 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);
13955: return 1;
13956: }
13957: }
13958: }
13959:
13960: /*for (i=1; i<=imx; i++){
13961: for (m=firstpass; (m<lastpass); m++){
13962: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
13963: }
13964:
13965: }*/
13966:
13967:
1.139 brouard 13968: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
13969: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 13970:
13971: return (0);
1.164 brouard 13972: /* endread:*/
1.136 brouard 13973: printf("Exiting calandcheckages: ");
13974: return (1);
13975: }
13976:
1.172 brouard 13977: #if defined(_MSC_VER)
13978: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
13979: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
13980: //#include "stdafx.h"
13981: //#include <stdio.h>
13982: //#include <tchar.h>
13983: //#include <windows.h>
13984: //#include <iostream>
13985: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
13986:
13987: LPFN_ISWOW64PROCESS fnIsWow64Process;
13988:
13989: BOOL IsWow64()
13990: {
13991: BOOL bIsWow64 = FALSE;
13992:
13993: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
13994: // (HANDLE, PBOOL);
13995:
13996: //LPFN_ISWOW64PROCESS fnIsWow64Process;
13997:
13998: HMODULE module = GetModuleHandle(_T("kernel32"));
13999: const char funcName[] = "IsWow64Process";
14000: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
14001: GetProcAddress(module, funcName);
14002:
14003: if (NULL != fnIsWow64Process)
14004: {
14005: if (!fnIsWow64Process(GetCurrentProcess(),
14006: &bIsWow64))
14007: //throw std::exception("Unknown error");
14008: printf("Unknown error\n");
14009: }
14010: return bIsWow64 != FALSE;
14011: }
14012: #endif
1.177 brouard 14013:
1.191 brouard 14014: void syscompilerinfo(int logged)
1.292 brouard 14015: {
14016: #include <stdint.h>
14017:
14018: /* #include "syscompilerinfo.h"*/
1.185 brouard 14019: /* command line Intel compiler 32bit windows, XP compatible:*/
14020: /* /GS /W3 /Gy
14021: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
14022: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
14023: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 14024: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
14025: */
14026: /* 64 bits */
1.185 brouard 14027: /*
14028: /GS /W3 /Gy
14029: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
14030: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
14031: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
14032: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
14033: /* Optimization are useless and O3 is slower than O2 */
14034: /*
14035: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
14036: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
14037: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
14038: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
14039: */
1.186 brouard 14040: /* Link is */ /* /OUT:"visual studio
1.185 brouard 14041: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
14042: /PDB:"visual studio
14043: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
14044: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
14045: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
14046: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
14047: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
14048: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
14049: uiAccess='false'"
14050: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
14051: /NOLOGO /TLBID:1
14052: */
1.292 brouard 14053:
14054:
1.177 brouard 14055: #if defined __INTEL_COMPILER
1.178 brouard 14056: #if defined(__GNUC__)
14057: struct utsname sysInfo; /* For Intel on Linux and OS/X */
14058: #endif
1.177 brouard 14059: #elif defined(__GNUC__)
1.179 brouard 14060: #ifndef __APPLE__
1.174 brouard 14061: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 14062: #endif
1.177 brouard 14063: struct utsname sysInfo;
1.178 brouard 14064: int cross = CROSS;
14065: if (cross){
14066: printf("Cross-");
1.191 brouard 14067: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 14068: }
1.174 brouard 14069: #endif
14070:
1.191 brouard 14071: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 14072: #if defined(__clang__)
1.191 brouard 14073: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 14074: #endif
14075: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 14076: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 14077: #endif
14078: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 14079: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 14080: #endif
14081: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 14082: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 14083: #endif
14084: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 14085: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 14086: #endif
14087: #if defined(_MSC_VER)
1.191 brouard 14088: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 14089: #endif
14090: #if defined(__PGI)
1.191 brouard 14091: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 14092: #endif
14093: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 14094: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 14095: #endif
1.191 brouard 14096: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 14097:
1.167 brouard 14098: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
14099: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
14100: // Windows (x64 and x86)
1.191 brouard 14101: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 14102: #elif __unix__ // all unices, not all compilers
14103: // Unix
1.191 brouard 14104: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 14105: #elif __linux__
14106: // linux
1.191 brouard 14107: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 14108: #elif __APPLE__
1.174 brouard 14109: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 14110: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 14111: #endif
14112:
14113: /* __MINGW32__ */
14114: /* __CYGWIN__ */
14115: /* __MINGW64__ */
14116: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
14117: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
14118: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
14119: /* _WIN64 // Defined for applications for Win64. */
14120: /* _M_X64 // Defined for compilations that target x64 processors. */
14121: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 14122:
1.167 brouard 14123: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 14124: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 14125: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 14126: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 14127: #else
1.191 brouard 14128: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 14129: #endif
14130:
1.169 brouard 14131: #if defined(__GNUC__)
14132: # if defined(__GNUC_PATCHLEVEL__)
14133: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
14134: + __GNUC_MINOR__ * 100 \
14135: + __GNUC_PATCHLEVEL__)
14136: # else
14137: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
14138: + __GNUC_MINOR__ * 100)
14139: # endif
1.174 brouard 14140: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 14141: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 14142:
14143: if (uname(&sysInfo) != -1) {
14144: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 14145: 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 14146: }
14147: else
14148: perror("uname() error");
1.179 brouard 14149: //#ifndef __INTEL_COMPILER
14150: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 14151: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 14152: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 14153: #endif
1.169 brouard 14154: #endif
1.172 brouard 14155:
1.286 brouard 14156: // void main ()
1.172 brouard 14157: // {
1.169 brouard 14158: #if defined(_MSC_VER)
1.174 brouard 14159: if (IsWow64()){
1.191 brouard 14160: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
14161: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 14162: }
14163: else{
1.191 brouard 14164: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
14165: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 14166: }
1.172 brouard 14167: // printf("\nPress Enter to continue...");
14168: // getchar();
14169: // }
14170:
1.169 brouard 14171: #endif
14172:
1.167 brouard 14173:
1.219 brouard 14174: }
1.136 brouard 14175:
1.219 brouard 14176: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 14177: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.332 brouard 14178: /* Computes the prevalence limit for each combination of the dummy covariates */
1.235 brouard 14179: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 14180: /* double ftolpl = 1.e-10; */
1.180 brouard 14181: double age, agebase, agelim;
1.203 brouard 14182: double tot;
1.180 brouard 14183:
1.202 brouard 14184: strcpy(filerespl,"PL_");
14185: strcat(filerespl,fileresu);
14186: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 14187: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
14188: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 14189: }
1.288 brouard 14190: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
14191: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 14192: pstamp(ficrespl);
1.288 brouard 14193: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 14194: fprintf(ficrespl,"#Age ");
14195: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
14196: fprintf(ficrespl,"\n");
1.180 brouard 14197:
1.219 brouard 14198: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 14199:
1.219 brouard 14200: agebase=ageminpar;
14201: agelim=agemaxpar;
1.180 brouard 14202:
1.227 brouard 14203: /* i1=pow(2,ncoveff); */
1.234 brouard 14204: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 14205: if (cptcovn < 1){i1=1;}
1.180 brouard 14206:
1.337 brouard 14207: /* for(k=1; k<=i1;k++){ /\* For each combination k of dummy covariates in the model *\/ */
1.238 brouard 14208: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 14209: k=TKresult[nres];
1.338 brouard 14210: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 14211: /* if(i1 != 1 && TKresult[nres]!= k) /\* We found the combination k corresponding to the resultline value of dummies *\/ */
14212: /* continue; */
1.235 brouard 14213:
1.238 brouard 14214: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
14215: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
14216: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
14217: /* k=k+1; */
14218: /* to clean */
1.332 brouard 14219: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 14220: fprintf(ficrespl,"#******");
14221: printf("#******");
14222: fprintf(ficlog,"#******");
1.337 brouard 14223: 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 14224: /* fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Here problem for varying dummy*\/ */
1.337 brouard 14225: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14226: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14227: fprintf(ficrespl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14228: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14229: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14230: }
14231: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
14232: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
14233: /* fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
14234: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
14235: /* } */
1.238 brouard 14236: fprintf(ficrespl,"******\n");
14237: printf("******\n");
14238: fprintf(ficlog,"******\n");
14239: if(invalidvarcomb[k]){
14240: printf("\nCombination (%d) ignored because no case \n",k);
14241: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
14242: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
14243: continue;
14244: }
1.219 brouard 14245:
1.238 brouard 14246: fprintf(ficrespl,"#Age ");
1.337 brouard 14247: /* for(j=1;j<=cptcoveff;j++) { */
14248: /* fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14249: /* } */
14250: for(j=1;j<=cptcovs;j++) { /* New the quanti variable is added */
14251: fprintf(ficrespl,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 14252: }
14253: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
14254: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 14255:
1.238 brouard 14256: for (age=agebase; age<=agelim; age++){
14257: /* for (age=agebase; age<=agebase; age++){ */
1.337 brouard 14258: /**< Computes the prevalence limit in each live state at age x and for covariate combination (k and) nres */
14259: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres); /* Nicely done */
1.238 brouard 14260: fprintf(ficrespl,"%.0f ",age );
1.337 brouard 14261: /* for(j=1;j<=cptcoveff;j++) */
14262: /* fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14263: for(j=1;j<=cptcovs;j++)
14264: fprintf(ficrespl,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 14265: tot=0.;
14266: for(i=1; i<=nlstate;i++){
14267: tot += prlim[i][i];
14268: fprintf(ficrespl," %.5f", prlim[i][i]);
14269: }
14270: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
14271: } /* Age */
14272: /* was end of cptcod */
1.337 brouard 14273: } /* nres */
14274: /* } /\* for each combination *\/ */
1.219 brouard 14275: return 0;
1.180 brouard 14276: }
14277:
1.218 brouard 14278: 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 14279: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 14280:
14281: /* Computes the back prevalence limit for any combination of covariate values
14282: * at any age between ageminpar and agemaxpar
14283: */
1.235 brouard 14284: int i, j, k, i1, nres=0 ;
1.217 brouard 14285: /* double ftolpl = 1.e-10; */
14286: double age, agebase, agelim;
14287: double tot;
1.218 brouard 14288: /* double ***mobaverage; */
14289: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 14290:
14291: strcpy(fileresplb,"PLB_");
14292: strcat(fileresplb,fileresu);
14293: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 14294: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
14295: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 14296: }
1.288 brouard 14297: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
14298: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 14299: pstamp(ficresplb);
1.288 brouard 14300: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 14301: fprintf(ficresplb,"#Age ");
14302: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
14303: fprintf(ficresplb,"\n");
14304:
1.218 brouard 14305:
14306: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
14307:
14308: agebase=ageminpar;
14309: agelim=agemaxpar;
14310:
14311:
1.227 brouard 14312: i1=pow(2,cptcoveff);
1.218 brouard 14313: if (cptcovn < 1){i1=1;}
1.227 brouard 14314:
1.238 brouard 14315: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.338 brouard 14316: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
14317: k=TKresult[nres];
14318: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
14319: /* if(i1 != 1 && TKresult[nres]!= k) */
14320: /* continue; */
14321: /* /\*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*\/ */
1.238 brouard 14322: fprintf(ficresplb,"#******");
14323: printf("#******");
14324: fprintf(ficlog,"#******");
1.338 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) */
14326: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14327: fprintf(ficresplb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14328: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 14329: }
1.338 brouard 14330: /* for(j=1;j<=cptcoveff ;j++) {/\* all covariates *\/ */
14331: /* fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14332: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14333: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14334: /* } */
14335: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
14336: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14337: /* fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14338: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14339: /* } */
1.238 brouard 14340: fprintf(ficresplb,"******\n");
14341: printf("******\n");
14342: fprintf(ficlog,"******\n");
14343: if(invalidvarcomb[k]){
14344: printf("\nCombination (%d) ignored because no cases \n",k);
14345: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
14346: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
14347: continue;
14348: }
1.218 brouard 14349:
1.238 brouard 14350: fprintf(ficresplb,"#Age ");
1.338 brouard 14351: for(j=1;j<=cptcovs;j++) {
14352: fprintf(ficresplb,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 14353: }
14354: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
14355: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 14356:
14357:
1.238 brouard 14358: for (age=agebase; age<=agelim; age++){
14359: /* for (age=agebase; age<=agebase; age++){ */
14360: if(mobilavproj > 0){
14361: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
14362: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 14363: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 14364: }else if (mobilavproj == 0){
14365: 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);
14366: 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);
14367: exit(1);
14368: }else{
14369: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 14370: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 14371: /* printf("TOTOT\n"); */
14372: /* exit(1); */
1.238 brouard 14373: }
14374: fprintf(ficresplb,"%.0f ",age );
1.338 brouard 14375: for(j=1;j<=cptcovs;j++)
14376: fprintf(ficresplb,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 14377: tot=0.;
14378: for(i=1; i<=nlstate;i++){
14379: tot += bprlim[i][i];
14380: fprintf(ficresplb," %.5f", bprlim[i][i]);
14381: }
14382: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
14383: } /* Age */
14384: /* was end of cptcod */
1.255 brouard 14385: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.338 brouard 14386: /* } /\* end of any combination *\/ */
1.238 brouard 14387: } /* end of nres */
1.218 brouard 14388: /* hBijx(p, bage, fage); */
14389: /* fclose(ficrespijb); */
14390:
14391: return 0;
1.217 brouard 14392: }
1.218 brouard 14393:
1.180 brouard 14394: int hPijx(double *p, int bage, int fage){
14395: /*------------- h Pij x at various ages ------------*/
1.336 brouard 14396: /* to be optimized with precov */
1.180 brouard 14397: int stepsize;
14398: int agelim;
14399: int hstepm;
14400: int nhstepm;
1.359 brouard 14401: int h, i, i1, j, k, nres=0;
1.180 brouard 14402:
14403: double agedeb;
14404: double ***p3mat;
14405:
1.337 brouard 14406: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
14407: if((ficrespij=fopen(filerespij,"w"))==NULL) {
14408: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
14409: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
14410: }
14411: printf("Computing pij: result on file '%s' \n", filerespij);
14412: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
14413:
14414: stepsize=(int) (stepm+YEARM-1)/YEARM;
14415: /*if (stepm<=24) stepsize=2;*/
14416:
14417: agelim=AGESUP;
14418: hstepm=stepsize*YEARM; /* Every year of age */
14419: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
14420:
14421: /* hstepm=1; aff par mois*/
14422: pstamp(ficrespij);
14423: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
14424: i1= pow(2,cptcoveff);
14425: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
14426: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
14427: /* k=k+1; */
14428: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
14429: k=TKresult[nres];
1.338 brouard 14430: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 14431: /* for(k=1; k<=i1;k++){ */
14432: /* if(i1 != 1 && TKresult[nres]!= k) */
14433: /* continue; */
14434: fprintf(ficrespij,"\n#****** ");
14435: for(j=1;j<=cptcovs;j++){
14436: fprintf(ficrespij," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
14437: /* fprintf(ficrespij,"@wV%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14438: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
14439: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
14440: /* fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
14441: }
14442: fprintf(ficrespij,"******\n");
14443:
14444: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
14445: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
14446: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
14447:
14448: /* nhstepm=nhstepm*YEARM; aff par mois*/
14449:
14450: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
14451: oldm=oldms;savm=savms;
14452: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
14453: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
14454: for(i=1; i<=nlstate;i++)
14455: for(j=1; j<=nlstate+ndeath;j++)
14456: fprintf(ficrespij," %1d-%1d",i,j);
14457: fprintf(ficrespij,"\n");
14458: for (h=0; h<=nhstepm; h++){
14459: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
14460: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.183 brouard 14461: for(i=1; i<=nlstate;i++)
14462: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 14463: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.183 brouard 14464: fprintf(ficrespij,"\n");
14465: }
1.337 brouard 14466: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
14467: fprintf(ficrespij,"\n");
1.180 brouard 14468: }
1.337 brouard 14469: }
14470: /*}*/
14471: return 0;
1.180 brouard 14472: }
1.218 brouard 14473:
14474: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 14475: /*------------- h Bij x at various ages ------------*/
1.336 brouard 14476: /* To be optimized with precov */
1.217 brouard 14477: int stepsize;
1.218 brouard 14478: /* int agelim; */
14479: int ageminl;
1.217 brouard 14480: int hstepm;
14481: int nhstepm;
1.238 brouard 14482: int h, i, i1, j, k, nres;
1.218 brouard 14483:
1.217 brouard 14484: double agedeb;
14485: double ***p3mat;
1.218 brouard 14486:
14487: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
14488: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
14489: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
14490: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
14491: }
14492: printf("Computing pij back: result on file '%s' \n", filerespijb);
14493: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
14494:
14495: stepsize=(int) (stepm+YEARM-1)/YEARM;
14496: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 14497:
1.218 brouard 14498: /* agelim=AGESUP; */
1.289 brouard 14499: ageminl=AGEINF; /* was 30 */
1.218 brouard 14500: hstepm=stepsize*YEARM; /* Every year of age */
14501: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
14502:
14503: /* hstepm=1; aff par mois*/
14504: pstamp(ficrespijb);
1.255 brouard 14505: 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 14506: i1= pow(2,cptcoveff);
1.218 brouard 14507: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
14508: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
14509: /* k=k+1; */
1.238 brouard 14510: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 14511: k=TKresult[nres];
1.338 brouard 14512: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 14513: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
14514: /* if(i1 != 1 && TKresult[nres]!= k) */
14515: /* continue; */
14516: fprintf(ficrespijb,"\n#****** ");
14517: for(j=1;j<=cptcovs;j++){
1.338 brouard 14518: fprintf(ficrespijb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.337 brouard 14519: /* for(j=1;j<=cptcoveff;j++) */
14520: /* fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14521: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
14522: /* fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14523: }
14524: fprintf(ficrespijb,"******\n");
14525: if(invalidvarcomb[k]){ /* Is it necessary here? */
14526: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
14527: continue;
14528: }
14529:
14530: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
14531: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
14532: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
14533: 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 */
14534: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
14535:
14536: /* nhstepm=nhstepm*YEARM; aff par mois*/
14537:
14538: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
14539: /* and memory limitations if stepm is small */
14540:
14541: /* oldm=oldms;savm=savms; */
14542: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
14543: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
14544: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
14545: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
14546: for(i=1; i<=nlstate;i++)
14547: for(j=1; j<=nlstate+ndeath;j++)
14548: fprintf(ficrespijb," %1d-%1d",i,j);
14549: fprintf(ficrespijb,"\n");
14550: for (h=0; h<=nhstepm; h++){
14551: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
14552: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
14553: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
1.217 brouard 14554: for(i=1; i<=nlstate;i++)
14555: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 14556: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.217 brouard 14557: fprintf(ficrespijb,"\n");
1.337 brouard 14558: }
14559: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
14560: fprintf(ficrespijb,"\n");
14561: } /* end age deb */
14562: /* } /\* end combination *\/ */
1.238 brouard 14563: } /* end nres */
1.218 brouard 14564: return 0;
14565: } /* hBijx */
1.217 brouard 14566:
1.180 brouard 14567:
1.136 brouard 14568: /***********************************************/
14569: /**************** Main Program *****************/
14570: /***********************************************/
14571:
14572: int main(int argc, char *argv[])
14573: {
14574: #ifdef GSL
14575: const gsl_multimin_fminimizer_type *T;
14576: size_t iteri = 0, it;
14577: int rval = GSL_CONTINUE;
14578: int status = GSL_SUCCESS;
14579: double ssval;
14580: #endif
14581: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 14582: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
14583: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 14584: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 14585: int jj, ll, li, lj, lk;
1.136 brouard 14586: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 14587: int num_filled;
1.136 brouard 14588: int itimes;
14589: int NDIM=2;
14590: int vpopbased=0;
1.235 brouard 14591: int nres=0;
1.258 brouard 14592: int endishere=0;
1.277 brouard 14593: int noffset=0;
1.274 brouard 14594: int ncurrv=0; /* Temporary variable */
14595:
1.164 brouard 14596: char ca[32], cb[32];
1.136 brouard 14597: /* FILE *fichtm; *//* Html File */
14598: /* FILE *ficgp;*/ /*Gnuplot File */
14599: struct stat info;
1.191 brouard 14600: double agedeb=0.;
1.194 brouard 14601:
14602: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 14603: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 14604:
1.165 brouard 14605: double fret;
1.191 brouard 14606: double dum=0.; /* Dummy variable */
1.359 brouard 14607: /* double*** p3mat;*/
1.218 brouard 14608: /* double ***mobaverage; */
1.319 brouard 14609: double wald;
1.164 brouard 14610:
1.351 brouard 14611: char line[MAXLINE], linetmp[MAXLINE];
1.197 brouard 14612: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
14613:
1.234 brouard 14614: char modeltemp[MAXLINE];
1.332 brouard 14615: char resultline[MAXLINE], resultlineori[MAXLINE];
1.230 brouard 14616:
1.136 brouard 14617: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 14618: char *tok, *val; /* pathtot */
1.334 brouard 14619: /* int firstobs=1, lastobs=10; /\* nobs = lastobs-firstobs declared globally ;*\/ */
1.359 brouard 14620: int c, h; /* c2; */
1.191 brouard 14621: int jl=0;
14622: int i1, j1, jk, stepsize=0;
1.194 brouard 14623: int count=0;
14624:
1.164 brouard 14625: int *tab;
1.136 brouard 14626: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 14627: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
14628: /* double anprojf, mprojf, jprojf; */
14629: /* double jintmean,mintmean,aintmean; */
14630: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
14631: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
14632: double yrfproj= 10.0; /* Number of years of forward projections */
14633: double yrbproj= 10.0; /* Number of years of backward projections */
14634: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 14635: int mobilav=0,popforecast=0;
1.191 brouard 14636: int hstepm=0, nhstepm=0;
1.136 brouard 14637: int agemortsup;
14638: float sumlpop=0.;
14639: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
14640: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
14641:
1.191 brouard 14642: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 14643: double ftolpl=FTOL;
14644: double **prlim;
1.217 brouard 14645: double **bprlim;
1.317 brouard 14646: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
14647: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 14648: double ***paramstart; /* Matrix of starting parameter values */
14649: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 14650: double **matcov; /* Matrix of covariance */
1.203 brouard 14651: double **hess; /* Hessian matrix */
1.136 brouard 14652: double ***delti3; /* Scale */
14653: double *delti; /* Scale */
14654: double ***eij, ***vareij;
1.359 brouard 14655: //double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 14656:
1.136 brouard 14657: double *epj, vepp;
1.164 brouard 14658:
1.273 brouard 14659: double dateprev1, dateprev2;
1.296 brouard 14660: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
14661: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
14662:
1.217 brouard 14663:
1.136 brouard 14664: double **ximort;
1.145 brouard 14665: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 14666: int *dcwave;
14667:
1.164 brouard 14668: char z[1]="c";
1.136 brouard 14669:
14670: /*char *strt;*/
14671: char strtend[80];
1.126 brouard 14672:
1.164 brouard 14673:
1.126 brouard 14674: /* setlocale (LC_ALL, ""); */
14675: /* bindtextdomain (PACKAGE, LOCALEDIR); */
14676: /* textdomain (PACKAGE); */
14677: /* setlocale (LC_CTYPE, ""); */
14678: /* setlocale (LC_MESSAGES, ""); */
14679:
14680: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 14681: rstart_time = time(NULL);
14682: /* (void) gettimeofday(&start_time,&tzp);*/
14683: start_time = *localtime(&rstart_time);
1.126 brouard 14684: curr_time=start_time;
1.157 brouard 14685: /*tml = *localtime(&start_time.tm_sec);*/
14686: /* strcpy(strstart,asctime(&tml)); */
14687: strcpy(strstart,asctime(&start_time));
1.126 brouard 14688:
14689: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 14690: /* tp.tm_sec = tp.tm_sec +86400; */
14691: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 14692: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
14693: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
14694: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 14695: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 14696: /* strt=asctime(&tmg); */
14697: /* printf("Time(after) =%s",strstart); */
14698: /* (void) time (&time_value);
14699: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
14700: * tm = *localtime(&time_value);
14701: * strstart=asctime(&tm);
14702: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
14703: */
14704:
14705: nberr=0; /* Number of errors and warnings */
14706: nbwarn=0;
1.184 brouard 14707: #ifdef WIN32
14708: _getcwd(pathcd, size);
14709: #else
1.126 brouard 14710: getcwd(pathcd, size);
1.184 brouard 14711: #endif
1.191 brouard 14712: syscompilerinfo(0);
1.359 brouard 14713: printf("\nIMaCh prax version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 14714: if(argc <=1){
14715: printf("\nEnter the parameter file name: ");
1.205 brouard 14716: if(!fgets(pathr,FILENAMELENGTH,stdin)){
14717: printf("ERROR Empty parameter file name\n");
14718: goto end;
14719: }
1.126 brouard 14720: i=strlen(pathr);
14721: if(pathr[i-1]=='\n')
14722: pathr[i-1]='\0';
1.156 brouard 14723: i=strlen(pathr);
1.205 brouard 14724: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 14725: pathr[i-1]='\0';
1.205 brouard 14726: }
14727: i=strlen(pathr);
14728: if( i==0 ){
14729: printf("ERROR Empty parameter file name\n");
14730: goto end;
14731: }
14732: for (tok = pathr; tok != NULL; ){
1.126 brouard 14733: printf("Pathr |%s|\n",pathr);
14734: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
14735: printf("val= |%s| pathr=%s\n",val,pathr);
14736: strcpy (pathtot, val);
14737: if(pathr[0] == '\0') break; /* Dirty */
14738: }
14739: }
1.281 brouard 14740: else if (argc<=2){
14741: strcpy(pathtot,argv[1]);
14742: }
1.126 brouard 14743: else{
14744: strcpy(pathtot,argv[1]);
1.281 brouard 14745: strcpy(z,argv[2]);
14746: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 14747: }
14748: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
14749: /*cygwin_split_path(pathtot,path,optionfile);
14750: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
14751: /* cutv(path,optionfile,pathtot,'\\');*/
14752:
14753: /* Split argv[0], imach program to get pathimach */
14754: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
14755: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
14756: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
14757: /* strcpy(pathimach,argv[0]); */
14758: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
14759: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
14760: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 14761: #ifdef WIN32
14762: _chdir(path); /* Can be a relative path */
14763: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
14764: #else
1.126 brouard 14765: chdir(path); /* Can be a relative path */
1.184 brouard 14766: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
14767: #endif
14768: printf("Current directory %s!\n",pathcd);
1.126 brouard 14769: strcpy(command,"mkdir ");
14770: strcat(command,optionfilefiname);
14771: if((outcmd=system(command)) != 0){
1.169 brouard 14772: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 14773: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
14774: /* fclose(ficlog); */
14775: /* exit(1); */
14776: }
14777: /* if((imk=mkdir(optionfilefiname))<0){ */
14778: /* perror("mkdir"); */
14779: /* } */
14780:
14781: /*-------- arguments in the command line --------*/
14782:
1.186 brouard 14783: /* Main Log file */
1.126 brouard 14784: strcat(filelog, optionfilefiname);
14785: strcat(filelog,".log"); /* */
14786: if((ficlog=fopen(filelog,"w"))==NULL) {
14787: printf("Problem with logfile %s\n",filelog);
14788: goto end;
14789: }
14790: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 14791: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 14792: fprintf(ficlog,"\nEnter the parameter file name: \n");
14793: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
14794: path=%s \n\
14795: optionfile=%s\n\
14796: optionfilext=%s\n\
1.156 brouard 14797: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 14798:
1.197 brouard 14799: syscompilerinfo(1);
1.167 brouard 14800:
1.126 brouard 14801: printf("Local time (at start):%s",strstart);
14802: fprintf(ficlog,"Local time (at start): %s",strstart);
14803: fflush(ficlog);
14804: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 14805: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 14806:
14807: /* */
14808: strcpy(fileres,"r");
14809: strcat(fileres, optionfilefiname);
1.201 brouard 14810: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 14811: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 14812: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 14813:
1.186 brouard 14814: /* Main ---------arguments file --------*/
1.126 brouard 14815:
14816: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 14817: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
14818: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 14819: fflush(ficlog);
1.149 brouard 14820: /* goto end; */
14821: exit(70);
1.126 brouard 14822: }
14823:
14824: strcpy(filereso,"o");
1.201 brouard 14825: strcat(filereso,fileresu);
1.126 brouard 14826: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
14827: printf("Problem with Output resultfile: %s\n", filereso);
14828: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
14829: fflush(ficlog);
14830: goto end;
14831: }
1.278 brouard 14832: /*-------- Rewriting parameter file ----------*/
14833: strcpy(rfileres,"r"); /* "Rparameterfile */
14834: strcat(rfileres,optionfilefiname); /* Parameter file first name */
14835: strcat(rfileres,"."); /* */
14836: strcat(rfileres,optionfilext); /* Other files have txt extension */
14837: if((ficres =fopen(rfileres,"w"))==NULL) {
14838: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
14839: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
14840: fflush(ficlog);
14841: goto end;
14842: }
14843: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 14844:
1.278 brouard 14845:
1.126 brouard 14846: /* Reads comments: lines beginning with '#' */
14847: numlinepar=0;
1.277 brouard 14848: /* Is it a BOM UTF-8 Windows file? */
14849: /* First parameter line */
1.197 brouard 14850: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 14851: noffset=0;
14852: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
14853: {
14854: noffset=noffset+3;
14855: printf("# File is an UTF8 Bom.\n"); // 0xBF
14856: }
1.302 brouard 14857: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
14858: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 14859: {
14860: noffset=noffset+2;
14861: printf("# File is an UTF16BE BOM file\n");
14862: }
14863: else if( line[0] == 0 && line[1] == 0)
14864: {
14865: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
14866: noffset=noffset+4;
14867: printf("# File is an UTF16BE BOM file\n");
14868: }
14869: } else{
14870: ;/*printf(" Not a BOM file\n");*/
14871: }
14872:
1.197 brouard 14873: /* If line starts with a # it is a comment */
1.277 brouard 14874: if (line[noffset] == '#') {
1.197 brouard 14875: numlinepar++;
14876: fputs(line,stdout);
14877: fputs(line,ficparo);
1.278 brouard 14878: fputs(line,ficres);
1.197 brouard 14879: fputs(line,ficlog);
14880: continue;
14881: }else
14882: break;
14883: }
14884: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
14885: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
14886: if (num_filled != 5) {
14887: printf("Should be 5 parameters\n");
1.283 brouard 14888: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 14889: }
1.126 brouard 14890: numlinepar++;
1.197 brouard 14891: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 14892: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
14893: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
14894: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 14895: }
14896: /* Second parameter line */
14897: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 14898: /* while(fscanf(ficpar,"%[^\n]", line)) { */
14899: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 14900: if (line[0] == '#') {
14901: numlinepar++;
1.283 brouard 14902: printf("%s",line);
14903: fprintf(ficres,"%s",line);
14904: fprintf(ficparo,"%s",line);
14905: fprintf(ficlog,"%s",line);
1.197 brouard 14906: continue;
14907: }else
14908: break;
14909: }
1.223 brouard 14910: 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", \
14911: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
14912: if (num_filled != 11) {
14913: 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 14914: printf("but line=%s\n",line);
1.283 brouard 14915: 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");
14916: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 14917: }
1.286 brouard 14918: if( lastpass > maxwav){
14919: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
14920: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
14921: fflush(ficlog);
14922: goto end;
14923: }
14924: 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 14925: 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 14926: 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 14927: 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 14928: }
1.203 brouard 14929: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 14930: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 14931: /* Third parameter line */
14932: while(fgets(line, MAXLINE, ficpar)) {
14933: /* If line starts with a # it is a comment */
14934: if (line[0] == '#') {
14935: numlinepar++;
1.283 brouard 14936: printf("%s",line);
14937: fprintf(ficres,"%s",line);
14938: fprintf(ficparo,"%s",line);
14939: fprintf(ficlog,"%s",line);
1.197 brouard 14940: continue;
14941: }else
14942: break;
14943: }
1.351 brouard 14944: if((num_filled=sscanf(line,"model=%[^.\n]", model)) !=EOF){ /* Every character after model but dot and return */
14945: if (num_filled != 1){
14946: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
14947: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
14948: model[0]='\0';
14949: goto end;
14950: }else{
14951: trimbtab(linetmp,line); /* Trims multiple blanks in line */
14952: strcpy(line, linetmp);
14953: }
14954: }
14955: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){ /* Every character after 1+age but dot and return */
1.279 brouard 14956: if (num_filled != 1){
1.302 brouard 14957: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
14958: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 14959: model[0]='\0';
14960: goto end;
14961: }
14962: else{
14963: if (model[0]=='+'){
14964: for(i=1; i<=strlen(model);i++)
14965: modeltemp[i-1]=model[i];
1.201 brouard 14966: strcpy(model,modeltemp);
1.197 brouard 14967: }
14968: }
1.338 brouard 14969: /* printf(" model=1+age%s modeltemp= %s, model=1+age+%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 14970: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 14971: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
14972: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
14973: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 14974: }
14975: /* 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); */
14976: /* numlinepar=numlinepar+3; /\* In general *\/ */
14977: /* 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 14978: /* 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); */
14979: /* 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 14980: fflush(ficlog);
1.190 brouard 14981: /* if(model[0]=='#'|| model[0]== '\0'){ */
14982: if(model[0]=='#'){
1.279 brouard 14983: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
14984: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
14985: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 14986: if(mle != -1){
1.279 brouard 14987: 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 14988: exit(1);
14989: }
14990: }
1.126 brouard 14991: while((c=getc(ficpar))=='#' && c!= EOF){
14992: ungetc(c,ficpar);
14993: fgets(line, MAXLINE, ficpar);
14994: numlinepar++;
1.195 brouard 14995: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
14996: z[0]=line[1];
1.342 brouard 14997: }else if(line[1]=='d'){ /* For debugging individual values of covariates in ficresilk */
1.343 brouard 14998: debugILK=1;printf("DebugILK\n");
1.195 brouard 14999: }
15000: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 15001: fputs(line, stdout);
15002: //puts(line);
1.126 brouard 15003: fputs(line,ficparo);
15004: fputs(line,ficlog);
15005: }
15006: ungetc(c,ficpar);
15007:
15008:
1.290 brouard 15009: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
15010: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
15011: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
1.341 brouard 15012: /* if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /\**< Time varying covariate (dummy and quantitative)*\/ */
15013: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs); /**< Might be better */
1.136 brouard 15014: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
15015: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
15016: v1+v2*age+v2*v3 makes cptcovn = 3
15017: */
15018: if (strlen(model)>1)
1.187 brouard 15019: 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 15020: else
1.187 brouard 15021: ncovmodel=2; /* Constant and age */
1.133 brouard 15022: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
15023: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 15024: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
15025: 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);
15026: 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);
15027: fflush(stdout);
15028: fclose (ficlog);
15029: goto end;
15030: }
1.126 brouard 15031: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
15032: delti=delti3[1][1];
15033: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
15034: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 15035: /* We could also provide initial parameters values giving by simple logistic regression
15036: * only one way, that is without matrix product. We will have nlstate maximizations */
15037: /* for(i=1;i<nlstate;i++){ */
15038: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
15039: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
15040: /* } */
1.126 brouard 15041: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 15042: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
15043: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 15044: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
15045: fclose (ficparo);
15046: fclose (ficlog);
15047: goto end;
15048: exit(0);
1.220 brouard 15049: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 15050: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 15051: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
15052: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 15053: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
15054: matcov=matrix(1,npar,1,npar);
1.203 brouard 15055: hess=matrix(1,npar,1,npar);
1.220 brouard 15056: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 15057: /* Read guessed parameters */
1.126 brouard 15058: /* Reads comments: lines beginning with '#' */
15059: while((c=getc(ficpar))=='#' && c!= EOF){
15060: ungetc(c,ficpar);
15061: fgets(line, MAXLINE, ficpar);
15062: numlinepar++;
1.141 brouard 15063: fputs(line,stdout);
1.126 brouard 15064: fputs(line,ficparo);
15065: fputs(line,ficlog);
15066: }
15067: ungetc(c,ficpar);
15068:
15069: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 15070: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 15071: for(i=1; i <=nlstate; i++){
1.234 brouard 15072: j=0;
1.126 brouard 15073: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 15074: if(jj==i) continue;
15075: j++;
1.292 brouard 15076: while((c=getc(ficpar))=='#' && c!= EOF){
15077: ungetc(c,ficpar);
15078: fgets(line, MAXLINE, ficpar);
15079: numlinepar++;
15080: fputs(line,stdout);
15081: fputs(line,ficparo);
15082: fputs(line,ficlog);
15083: }
15084: ungetc(c,ficpar);
1.234 brouard 15085: fscanf(ficpar,"%1d%1d",&i1,&j1);
15086: if ((i1 != i) || (j1 != jj)){
15087: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 15088: It might be a problem of design; if ncovcol and the model are correct\n \
15089: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 15090: exit(1);
15091: }
15092: fprintf(ficparo,"%1d%1d",i1,j1);
15093: if(mle==1)
15094: printf("%1d%1d",i,jj);
15095: fprintf(ficlog,"%1d%1d",i,jj);
15096: for(k=1; k<=ncovmodel;k++){
15097: fscanf(ficpar," %lf",¶m[i][j][k]);
15098: if(mle==1){
15099: printf(" %lf",param[i][j][k]);
15100: fprintf(ficlog," %lf",param[i][j][k]);
15101: }
15102: else
15103: fprintf(ficlog," %lf",param[i][j][k]);
15104: fprintf(ficparo," %lf",param[i][j][k]);
15105: }
15106: fscanf(ficpar,"\n");
15107: numlinepar++;
15108: if(mle==1)
15109: printf("\n");
15110: fprintf(ficlog,"\n");
15111: fprintf(ficparo,"\n");
1.126 brouard 15112: }
15113: }
15114: fflush(ficlog);
1.234 brouard 15115:
1.251 brouard 15116: /* Reads parameters values */
1.126 brouard 15117: p=param[1][1];
1.251 brouard 15118: pstart=paramstart[1][1];
1.126 brouard 15119:
15120: /* Reads comments: lines beginning with '#' */
15121: while((c=getc(ficpar))=='#' && c!= EOF){
15122: ungetc(c,ficpar);
15123: fgets(line, MAXLINE, ficpar);
15124: numlinepar++;
1.141 brouard 15125: fputs(line,stdout);
1.126 brouard 15126: fputs(line,ficparo);
15127: fputs(line,ficlog);
15128: }
15129: ungetc(c,ficpar);
15130:
15131: for(i=1; i <=nlstate; i++){
15132: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 15133: fscanf(ficpar,"%1d%1d",&i1,&j1);
15134: if ( (i1-i) * (j1-j) != 0){
15135: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
15136: exit(1);
15137: }
15138: printf("%1d%1d",i,j);
15139: fprintf(ficparo,"%1d%1d",i1,j1);
15140: fprintf(ficlog,"%1d%1d",i1,j1);
15141: for(k=1; k<=ncovmodel;k++){
15142: fscanf(ficpar,"%le",&delti3[i][j][k]);
15143: printf(" %le",delti3[i][j][k]);
15144: fprintf(ficparo," %le",delti3[i][j][k]);
15145: fprintf(ficlog," %le",delti3[i][j][k]);
15146: }
15147: fscanf(ficpar,"\n");
15148: numlinepar++;
15149: printf("\n");
15150: fprintf(ficparo,"\n");
15151: fprintf(ficlog,"\n");
1.126 brouard 15152: }
15153: }
15154: fflush(ficlog);
1.234 brouard 15155:
1.145 brouard 15156: /* Reads covariance matrix */
1.126 brouard 15157: delti=delti3[1][1];
1.220 brouard 15158:
15159:
1.126 brouard 15160: /* 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 15161:
1.126 brouard 15162: /* Reads comments: lines beginning with '#' */
15163: while((c=getc(ficpar))=='#' && c!= EOF){
15164: ungetc(c,ficpar);
15165: fgets(line, MAXLINE, ficpar);
15166: numlinepar++;
1.141 brouard 15167: fputs(line,stdout);
1.126 brouard 15168: fputs(line,ficparo);
15169: fputs(line,ficlog);
15170: }
15171: ungetc(c,ficpar);
1.220 brouard 15172:
1.126 brouard 15173: matcov=matrix(1,npar,1,npar);
1.203 brouard 15174: hess=matrix(1,npar,1,npar);
1.131 brouard 15175: for(i=1; i <=npar; i++)
15176: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 15177:
1.194 brouard 15178: /* Scans npar lines */
1.126 brouard 15179: for(i=1; i <=npar; i++){
1.226 brouard 15180: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 15181: if(count != 3){
1.226 brouard 15182: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 15183: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
15184: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 15185: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 15186: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
15187: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 15188: exit(1);
1.220 brouard 15189: }else{
1.226 brouard 15190: if(mle==1)
15191: printf("%1d%1d%d",i1,j1,jk);
15192: }
15193: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
15194: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 15195: for(j=1; j <=i; j++){
1.226 brouard 15196: fscanf(ficpar," %le",&matcov[i][j]);
15197: if(mle==1){
15198: printf(" %.5le",matcov[i][j]);
15199: }
15200: fprintf(ficlog," %.5le",matcov[i][j]);
15201: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 15202: }
15203: fscanf(ficpar,"\n");
15204: numlinepar++;
15205: if(mle==1)
1.220 brouard 15206: printf("\n");
1.126 brouard 15207: fprintf(ficlog,"\n");
15208: fprintf(ficparo,"\n");
15209: }
1.194 brouard 15210: /* End of read covariance matrix npar lines */
1.126 brouard 15211: for(i=1; i <=npar; i++)
15212: for(j=i+1;j<=npar;j++)
1.226 brouard 15213: matcov[i][j]=matcov[j][i];
1.126 brouard 15214:
15215: if(mle==1)
15216: printf("\n");
15217: fprintf(ficlog,"\n");
15218:
15219: fflush(ficlog);
15220:
15221: } /* End of mle != -3 */
1.218 brouard 15222:
1.186 brouard 15223: /* Main data
15224: */
1.290 brouard 15225: nobs=lastobs-firstobs+1; /* was = lastobs;*/
15226: /* num=lvector(1,n); */
15227: /* moisnais=vector(1,n); */
15228: /* annais=vector(1,n); */
15229: /* moisdc=vector(1,n); */
15230: /* andc=vector(1,n); */
15231: /* weight=vector(1,n); */
15232: /* agedc=vector(1,n); */
15233: /* cod=ivector(1,n); */
15234: /* for(i=1;i<=n;i++){ */
15235: num=lvector(firstobs,lastobs);
15236: moisnais=vector(firstobs,lastobs);
15237: annais=vector(firstobs,lastobs);
15238: moisdc=vector(firstobs,lastobs);
15239: andc=vector(firstobs,lastobs);
15240: weight=vector(firstobs,lastobs);
15241: agedc=vector(firstobs,lastobs);
15242: cod=ivector(firstobs,lastobs);
15243: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 15244: num[i]=0;
15245: moisnais[i]=0;
15246: annais[i]=0;
15247: moisdc[i]=0;
15248: andc[i]=0;
15249: agedc[i]=0;
15250: cod[i]=0;
15251: weight[i]=1.0; /* Equal weights, 1 by default */
15252: }
1.290 brouard 15253: mint=matrix(1,maxwav,firstobs,lastobs);
15254: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 15255: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.336 brouard 15256: /* printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel)); */
1.126 brouard 15257: tab=ivector(1,NCOVMAX);
1.144 brouard 15258: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 15259: 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 15260:
1.136 brouard 15261: /* Reads data from file datafile */
15262: if (readdata(datafile, firstobs, lastobs, &imx)==1)
15263: goto end;
15264:
15265: /* Calculation of the number of parameters from char model */
1.234 brouard 15266: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 15267: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
15268: k=3 V4 Tvar[k=3]= 4 (from V4)
15269: k=2 V1 Tvar[k=2]= 1 (from V1)
15270: k=1 Tvar[1]=2 (from V2)
1.234 brouard 15271: */
15272:
15273: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
15274: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 brouard 15275: TnsdVar=ivector(1,NCOVMAX); /* */
1.335 brouard 15276: /* for(i=1; i<=NCOVMAX;i++) TnsdVar[i]=3; */
1.234 brouard 15277: TvarsD=ivector(1,NCOVMAX); /* */
15278: TvarsQind=ivector(1,NCOVMAX); /* */
15279: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 15280: TvarF=ivector(1,NCOVMAX); /* */
15281: TvarFind=ivector(1,NCOVMAX); /* */
15282: TvarV=ivector(1,NCOVMAX); /* */
15283: TvarVind=ivector(1,NCOVMAX); /* */
15284: TvarA=ivector(1,NCOVMAX); /* */
15285: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 15286: TvarFD=ivector(1,NCOVMAX); /* */
15287: TvarFDind=ivector(1,NCOVMAX); /* */
15288: TvarFQ=ivector(1,NCOVMAX); /* */
15289: TvarFQind=ivector(1,NCOVMAX); /* */
15290: TvarVD=ivector(1,NCOVMAX); /* */
15291: TvarVDind=ivector(1,NCOVMAX); /* */
15292: TvarVQ=ivector(1,NCOVMAX); /* */
15293: TvarVQind=ivector(1,NCOVMAX); /* */
1.339 brouard 15294: TvarVV=ivector(1,NCOVMAX); /* */
15295: TvarVVind=ivector(1,NCOVMAX); /* */
1.349 brouard 15296: TvarVVA=ivector(1,NCOVMAX); /* */
15297: TvarVVAind=ivector(1,NCOVMAX); /* */
15298: TvarAVVA=ivector(1,NCOVMAX); /* */
15299: TvarAVVAind=ivector(1,NCOVMAX); /* */
1.231 brouard 15300:
1.230 brouard 15301: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 15302: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 15303: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
15304: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
15305: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.349 brouard 15306: DummyV=ivector(-1,NCOVMAX); /* 1 to 3 */
15307: FixedV=ivector(-1,NCOVMAX); /* 1 to 3 */
15308:
1.137 brouard 15309: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
15310: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
15311: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
15312: */
15313: /* For model-covariate k tells which data-covariate to use but
15314: because this model-covariate is a construction we invent a new column
15315: ncovcol + k1
15316: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
15317: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 15318: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
15319: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 15320: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
15321: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 15322: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 15323: */
1.145 brouard 15324: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
15325: 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 15326: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
15327: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.351 brouard 15328: Tvardk=imatrix(0,NCOVMAX,1,2);
1.145 brouard 15329: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 15330: 4 covariates (3 plus signs)
15331: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 15332: */
15333: for(i=1;i<NCOVMAX;i++)
15334: Tage[i]=0;
1.230 brouard 15335: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 15336: * individual dummy, fixed or varying:
15337: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
15338: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 15339: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
15340: * V1 df, V2 qf, V3 & V4 dv, V5 qv
15341: * Tmodelind[1]@9={9,0,3,2,}*/
15342: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
15343: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 15344: * individual quantitative, fixed or varying:
15345: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
15346: * 3, 1, 0, 0, 0, 0, 0, 0},
15347: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.349 brouard 15348:
15349: /* Probably useless zeroes */
15350: for(i=1;i<NCOVMAX;i++){
15351: DummyV[i]=0;
15352: FixedV[i]=0;
15353: }
15354:
15355: for(i=1; i <=ncovcol;i++){
15356: DummyV[i]=0;
15357: FixedV[i]=0;
15358: }
15359: for(i=ncovcol+1; i <=ncovcol+nqv;i++){
15360: DummyV[i]=1;
15361: FixedV[i]=0;
15362: }
15363: for(i=ncovcol+nqv+1; i <=ncovcol+nqv+ntv;i++){
15364: DummyV[i]=0;
15365: FixedV[i]=1;
15366: }
15367: for(i=ncovcol+nqv+ntv+1; i <=ncovcol+nqv+ntv+nqtv;i++){
15368: DummyV[i]=1;
15369: FixedV[i]=1;
15370: }
15371: for(i=1; i <=ncovcol+nqv+ntv+nqtv;i++){
15372: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",i,i,DummyV[i],i,FixedV[i]);
15373: 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]);
15374: }
15375:
15376:
15377:
1.186 brouard 15378: /* Main decodemodel */
15379:
1.187 brouard 15380:
1.223 brouard 15381: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 15382: goto end;
15383:
1.137 brouard 15384: if((double)(lastobs-imx)/(double)imx > 1.10){
15385: nbwarn++;
15386: 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);
15387: 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);
15388: }
1.136 brouard 15389: /* if(mle==1){*/
1.137 brouard 15390: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
15391: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 15392: }
15393:
15394: /*-calculation of age at interview from date of interview and age at death -*/
15395: agev=matrix(1,maxwav,1,imx);
15396:
15397: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
15398: goto end;
15399:
1.126 brouard 15400:
1.136 brouard 15401: agegomp=(int)agemin;
1.290 brouard 15402: free_vector(moisnais,firstobs,lastobs);
15403: free_vector(annais,firstobs,lastobs);
1.126 brouard 15404: /* free_matrix(mint,1,maxwav,1,n);
15405: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 15406: /* free_vector(moisdc,1,n); */
15407: /* free_vector(andc,1,n); */
1.145 brouard 15408: /* */
15409:
1.126 brouard 15410: wav=ivector(1,imx);
1.214 brouard 15411: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
15412: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
15413: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
15414: 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.*/
15415: bh=imatrix(1,lastpass-firstpass+2,1,imx);
15416: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 15417:
15418: /* Concatenates waves */
1.214 brouard 15419: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
15420: Death is a valid wave (if date is known).
15421: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
15422: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
15423: and mw[mi+1][i]. dh depends on stepm.
15424: */
15425:
1.126 brouard 15426: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 15427: /* Concatenates waves */
1.145 brouard 15428:
1.290 brouard 15429: free_vector(moisdc,firstobs,lastobs);
15430: free_vector(andc,firstobs,lastobs);
1.215 brouard 15431:
1.126 brouard 15432: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
15433: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
15434: ncodemax[1]=1;
1.145 brouard 15435: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 15436: cptcoveff=0;
1.220 brouard 15437: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
1.335 brouard 15438: 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 15439: }
15440:
15441: ncovcombmax=pow(2,cptcoveff);
1.338 brouard 15442: invalidvarcomb=ivector(0, ncovcombmax);
15443: for(i=0;i<ncovcombmax;i++)
1.227 brouard 15444: invalidvarcomb[i]=0;
15445:
1.211 brouard 15446: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 15447: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 15448: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 15449:
1.200 brouard 15450: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 15451: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 15452: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 15453: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
15454: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
15455: * (currently 0 or 1) in the data.
15456: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
15457: * corresponding modality (h,j).
15458: */
15459:
1.145 brouard 15460: h=0;
15461: /*if (cptcovn > 0) */
1.126 brouard 15462: m=pow(2,cptcoveff);
15463:
1.144 brouard 15464: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 15465: * For k=4 covariates, h goes from 1 to m=2**k
15466: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
15467: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 15468: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
15469: *______________________________ *______________________
15470: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
15471: * 2 2 1 1 1 * 1 0 0 0 1
15472: * 3 i=2 1 2 1 1 * 2 0 0 1 0
15473: * 4 2 2 1 1 * 3 0 0 1 1
15474: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
15475: * 6 2 1 2 1 * 5 0 1 0 1
15476: * 7 i=4 1 2 2 1 * 6 0 1 1 0
15477: * 8 2 2 2 1 * 7 0 1 1 1
15478: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
15479: * 10 2 1 1 2 * 9 1 0 0 1
15480: * 11 i=6 1 2 1 2 * 10 1 0 1 0
15481: * 12 2 2 1 2 * 11 1 0 1 1
15482: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
15483: * 14 2 1 2 2 * 13 1 1 0 1
15484: * 15 i=8 1 2 2 2 * 14 1 1 1 0
15485: * 16 2 2 2 2 * 15 1 1 1 1
15486: */
1.212 brouard 15487: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 15488: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
15489: * and the value of each covariate?
15490: * V1=1, V2=1, V3=2, V4=1 ?
15491: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
15492: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
15493: * In order to get the real value in the data, we use nbcode
15494: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
15495: * We are keeping this crazy system in order to be able (in the future?)
15496: * to have more than 2 values (0 or 1) for a covariate.
15497: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
15498: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
15499: * bbbbbbbb
15500: * 76543210
15501: * h-1 00000101 (6-1=5)
1.219 brouard 15502: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 15503: * &
15504: * 1 00000001 (1)
1.219 brouard 15505: * 00000000 = 1 & ((h-1) >> (k-1))
15506: * +1= 00000001 =1
1.211 brouard 15507: *
15508: * h=14, k=3 => h'=h-1=13, k'=k-1=2
15509: * h' 1101 =2^3+2^2+0x2^1+2^0
15510: * >>k' 11
15511: * & 00000001
15512: * = 00000001
15513: * +1 = 00000010=2 = codtabm(14,3)
15514: * Reverse h=6 and m=16?
15515: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
15516: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
15517: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
15518: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
15519: * V3=decodtabm(14,3,2**4)=2
15520: * h'=13 1101 =2^3+2^2+0x2^1+2^0
15521: *(h-1) >> (j-1) 0011 =13 >> 2
15522: * &1 000000001
15523: * = 000000001
15524: * +1= 000000010 =2
15525: * 2211
15526: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
15527: * V3=2
1.220 brouard 15528: * codtabm and decodtabm are identical
1.211 brouard 15529: */
15530:
1.145 brouard 15531:
15532: free_ivector(Ndum,-1,NCOVMAX);
15533:
15534:
1.126 brouard 15535:
1.186 brouard 15536: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 15537: strcpy(optionfilegnuplot,optionfilefiname);
15538: if(mle==-3)
1.201 brouard 15539: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 15540: strcat(optionfilegnuplot,".gp");
15541:
15542: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
15543: printf("Problem with file %s",optionfilegnuplot);
15544: }
15545: else{
1.204 brouard 15546: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 15547: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 15548: //fprintf(ficgp,"set missing 'NaNq'\n");
15549: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 15550: }
15551: /* fclose(ficgp);*/
1.186 brouard 15552:
15553:
15554: /* Initialisation of --------- index.htm --------*/
1.126 brouard 15555:
15556: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
15557: if(mle==-3)
1.201 brouard 15558: strcat(optionfilehtm,"-MORT_");
1.126 brouard 15559: strcat(optionfilehtm,".htm");
15560: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 15561: printf("Problem with %s \n",optionfilehtm);
15562: exit(0);
1.126 brouard 15563: }
15564:
15565: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
15566: strcat(optionfilehtmcov,"-cov.htm");
15567: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
15568: printf("Problem with %s \n",optionfilehtmcov), exit(0);
15569: }
15570: else{
15571: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
15572: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 15573: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 15574: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
15575: }
15576:
1.335 brouard 15577: fprintf(fichtm,"<html><head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n\
15578: <title>IMaCh %s</title></head>\n\
15579: <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n\
15580: <font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>\
15581: -EUROREVES-Institut de longévité-2013-2022-Japan Society for the Promotion of Sciences 日本学術振興会 \
15582: (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - \
15583: <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \n", optionfilehtm);
15584:
15585: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 15586: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 15587: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.337 brouard 15588: 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 15589: \n\
15590: <hr size=\"2\" color=\"#EC5E5E\">\
15591: <ul><li><h4>Parameter files</h4>\n\
15592: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
15593: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
15594: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
15595: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
15596: - Date and time at start: %s</ul>\n",\
1.335 brouard 15597: version,fullversion,optionfilehtm,optionfilehtm,title,datafile,datafile,firstpass,lastpass,stepm, weightopt, model, \
1.126 brouard 15598: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
15599: fileres,fileres,\
15600: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
15601: fflush(fichtm);
15602:
15603: strcpy(pathr,path);
15604: strcat(pathr,optionfilefiname);
1.184 brouard 15605: #ifdef WIN32
15606: _chdir(optionfilefiname); /* Move to directory named optionfile */
15607: #else
1.126 brouard 15608: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 15609: #endif
15610:
1.126 brouard 15611:
1.220 brouard 15612: /* Calculates basic frequencies. Computes observed prevalence at single age
15613: and for any valid combination of covariates
1.126 brouard 15614: and prints on file fileres'p'. */
1.359 brouard 15615: freqsummary(fileres, p, pstart, (double)agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 15616: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 15617:
15618: fprintf(fichtm,"\n");
1.286 brouard 15619: 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 15620: ftol, stepm);
15621: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
15622: ncurrv=1;
15623: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
15624: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
15625: ncurrv=i;
15626: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 15627: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 15628: ncurrv=i;
15629: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 15630: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 15631: ncurrv=i;
15632: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
15633: 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", \
15634: nlstate, ndeath, maxwav, mle, weightopt);
15635:
15636: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
15637: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
15638:
15639:
1.317 brouard 15640: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 15641: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
15642: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 15643: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 15644: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 15645: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
15646: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
15647: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
15648: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 15649:
1.126 brouard 15650: /* For Powell, parameters are in a vector p[] starting at p[1]
15651: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
15652: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
15653:
15654: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 15655: /* For mortality only */
1.126 brouard 15656: if (mle==-3){
1.136 brouard 15657: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 15658: for(i=1;i<=NDIM;i++)
15659: for(j=1;j<=NDIM;j++)
15660: ximort[i][j]=0.;
1.186 brouard 15661: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 15662: cens=ivector(firstobs,lastobs);
15663: ageexmed=vector(firstobs,lastobs);
15664: agecens=vector(firstobs,lastobs);
15665: dcwave=ivector(firstobs,lastobs);
1.223 brouard 15666:
1.126 brouard 15667: for (i=1; i<=imx; i++){
15668: dcwave[i]=-1;
15669: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 15670: if (s[m][i]>nlstate) {
15671: dcwave[i]=m;
15672: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
15673: break;
15674: }
1.126 brouard 15675: }
1.226 brouard 15676:
1.126 brouard 15677: for (i=1; i<=imx; i++) {
15678: if (wav[i]>0){
1.226 brouard 15679: ageexmed[i]=agev[mw[1][i]][i];
15680: j=wav[i];
15681: agecens[i]=1.;
15682:
15683: if (ageexmed[i]> 1 && wav[i] > 0){
15684: agecens[i]=agev[mw[j][i]][i];
15685: cens[i]= 1;
15686: }else if (ageexmed[i]< 1)
15687: cens[i]= -1;
15688: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
15689: cens[i]=0 ;
1.126 brouard 15690: }
15691: else cens[i]=-1;
15692: }
15693:
15694: for (i=1;i<=NDIM;i++) {
15695: for (j=1;j<=NDIM;j++)
1.226 brouard 15696: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 15697: }
15698:
1.302 brouard 15699: p[1]=0.0268; p[NDIM]=0.083;
15700: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 15701:
15702:
1.136 brouard 15703: #ifdef GSL
15704: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 15705: #else
1.359 brouard 15706: printf("Powell-mort\n"); fprintf(ficlog,"Powell-mort\n");
1.136 brouard 15707: #endif
1.201 brouard 15708: strcpy(filerespow,"POW-MORT_");
15709: strcat(filerespow,fileresu);
1.126 brouard 15710: if((ficrespow=fopen(filerespow,"w"))==NULL) {
15711: printf("Problem with resultfile: %s\n", filerespow);
15712: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
15713: }
1.136 brouard 15714: #ifdef GSL
15715: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 15716: #else
1.126 brouard 15717: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 15718: #endif
1.126 brouard 15719: /* for (i=1;i<=nlstate;i++)
15720: for(j=1;j<=nlstate+ndeath;j++)
15721: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
15722: */
15723: fprintf(ficrespow,"\n");
1.136 brouard 15724: #ifdef GSL
15725: /* gsl starts here */
15726: T = gsl_multimin_fminimizer_nmsimplex;
15727: gsl_multimin_fminimizer *sfm = NULL;
15728: gsl_vector *ss, *x;
15729: gsl_multimin_function minex_func;
15730:
15731: /* Initial vertex size vector */
15732: ss = gsl_vector_alloc (NDIM);
15733:
15734: if (ss == NULL){
15735: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
15736: }
15737: /* Set all step sizes to 1 */
15738: gsl_vector_set_all (ss, 0.001);
15739:
15740: /* Starting point */
1.126 brouard 15741:
1.136 brouard 15742: x = gsl_vector_alloc (NDIM);
15743:
15744: if (x == NULL){
15745: gsl_vector_free(ss);
15746: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
15747: }
15748:
15749: /* Initialize method and iterate */
15750: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 15751: /* gsl_vector_set(x, 0, 0.0268); */
15752: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 15753: gsl_vector_set(x, 0, p[1]);
15754: gsl_vector_set(x, 1, p[2]);
15755:
15756: minex_func.f = &gompertz_f;
15757: minex_func.n = NDIM;
15758: minex_func.params = (void *)&p; /* ??? */
15759:
15760: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
15761: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
15762:
15763: printf("Iterations beginning .....\n\n");
15764: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
15765:
15766: iteri=0;
15767: while (rval == GSL_CONTINUE){
15768: iteri++;
15769: status = gsl_multimin_fminimizer_iterate(sfm);
15770:
15771: if (status) printf("error: %s\n", gsl_strerror (status));
15772: fflush(0);
15773:
15774: if (status)
15775: break;
15776:
15777: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
15778: ssval = gsl_multimin_fminimizer_size (sfm);
15779:
15780: if (rval == GSL_SUCCESS)
15781: printf ("converged to a local maximum at\n");
15782:
15783: printf("%5d ", iteri);
15784: for (it = 0; it < NDIM; it++){
15785: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
15786: }
15787: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
15788: }
15789:
15790: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
15791:
15792: gsl_vector_free(x); /* initial values */
15793: gsl_vector_free(ss); /* inital step size */
15794: for (it=0; it<NDIM; it++){
15795: p[it+1]=gsl_vector_get(sfm->x,it);
15796: fprintf(ficrespow," %.12lf", p[it]);
15797: }
15798: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
15799: #endif
15800: #ifdef POWELL
15801: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
15802: #endif
1.126 brouard 15803: fclose(ficrespow);
15804:
1.203 brouard 15805: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 15806:
15807: for(i=1; i <=NDIM; i++)
15808: for(j=i+1;j<=NDIM;j++)
1.359 brouard 15809: matcov[i][j]=matcov[j][i];
1.126 brouard 15810:
15811: printf("\nCovariance matrix\n ");
1.203 brouard 15812: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 15813: for(i=1; i <=NDIM; i++) {
15814: for(j=1;j<=NDIM;j++){
1.220 brouard 15815: printf("%f ",matcov[i][j]);
15816: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 15817: }
1.203 brouard 15818: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 15819: }
15820:
15821: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 15822: for (i=1;i<=NDIM;i++) {
1.126 brouard 15823: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 15824: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
15825: }
1.302 brouard 15826: lsurv=vector(agegomp,AGESUP);
15827: lpop=vector(agegomp,AGESUP);
15828: tpop=vector(agegomp,AGESUP);
1.126 brouard 15829: lsurv[agegomp]=100000;
15830:
15831: for (k=agegomp;k<=AGESUP;k++) {
15832: agemortsup=k;
15833: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
15834: }
15835:
15836: for (k=agegomp;k<agemortsup;k++)
15837: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
15838:
15839: for (k=agegomp;k<agemortsup;k++){
15840: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
15841: sumlpop=sumlpop+lpop[k];
15842: }
15843:
15844: tpop[agegomp]=sumlpop;
15845: for (k=agegomp;k<(agemortsup-3);k++){
15846: /* tpop[k+1]=2;*/
15847: tpop[k+1]=tpop[k]-lpop[k];
15848: }
15849:
15850:
15851: printf("\nAge lx qx dx Lx Tx e(x)\n");
15852: for (k=agegomp;k<(agemortsup-2);k++)
15853: 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]);
15854:
15855:
15856: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 15857: ageminpar=50;
15858: agemaxpar=100;
1.194 brouard 15859: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
15860: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
15861: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
15862: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
15863: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
15864: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
15865: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 15866: }else{
15867: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
15868: 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 15869: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 15870: }
1.201 brouard 15871: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 15872: stepm, weightopt,\
15873: model,imx,p,matcov,agemortsup);
15874:
1.302 brouard 15875: free_vector(lsurv,agegomp,AGESUP);
15876: free_vector(lpop,agegomp,AGESUP);
15877: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 15878: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 15879: free_ivector(dcwave,firstobs,lastobs);
15880: free_vector(agecens,firstobs,lastobs);
15881: free_vector(ageexmed,firstobs,lastobs);
15882: free_ivector(cens,firstobs,lastobs);
1.220 brouard 15883: #ifdef GSL
1.136 brouard 15884: #endif
1.186 brouard 15885: } /* Endof if mle==-3 mortality only */
1.205 brouard 15886: /* Standard */
15887: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
15888: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
15889: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 15890: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 15891: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
15892: for (k=1; k<=npar;k++)
15893: printf(" %d %8.5f",k,p[k]);
15894: printf("\n");
1.205 brouard 15895: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
15896: /* mlikeli uses func not funcone */
1.247 brouard 15897: /* for(i=1;i<nlstate;i++){ */
15898: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
15899: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
15900: /* } */
1.205 brouard 15901: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
15902: }
15903: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
15904: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
15905: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
15906: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
15907: }
15908: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 15909: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
15910: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
1.335 brouard 15911: /* exit(0); */
1.126 brouard 15912: for (k=1; k<=npar;k++)
15913: printf(" %d %8.5f",k,p[k]);
15914: printf("\n");
15915:
15916: /*--------- results files --------------*/
1.283 brouard 15917: /* 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 15918:
15919:
15920: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 15921: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 15922: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 15923:
15924: printf("#model= 1 + age ");
15925: fprintf(ficres,"#model= 1 + age ");
15926: fprintf(ficlog,"#model= 1 + age ");
15927: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
15928: </ul>", model);
15929:
15930: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
15931: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
15932: if(nagesqr==1){
15933: printf(" + age*age ");
15934: fprintf(ficres," + age*age ");
15935: fprintf(ficlog," + age*age ");
15936: fprintf(fichtm, "<th>+ age*age</th>");
15937: }
15938: for(j=1;j <=ncovmodel-2;j++){
15939: if(Typevar[j]==0) {
15940: printf(" + V%d ",Tvar[j]);
15941: fprintf(ficres," + V%d ",Tvar[j]);
15942: fprintf(ficlog," + V%d ",Tvar[j]);
15943: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
15944: }else if(Typevar[j]==1) {
15945: printf(" + V%d*age ",Tvar[j]);
15946: fprintf(ficres," + V%d*age ",Tvar[j]);
15947: fprintf(ficlog," + V%d*age ",Tvar[j]);
15948: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
15949: }else if(Typevar[j]==2) {
15950: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
15951: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
15952: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
15953: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.349 brouard 15954: }else if(Typevar[j]==3) { /* TO VERIFY */
15955: printf(" + V%d*V%d*age ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
15956: fprintf(ficres," + V%d*V%d*age ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
15957: fprintf(ficlog," + V%d*V%d*age ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
15958: fprintf(fichtm, "<th>+ V%d*V%d*age</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.319 brouard 15959: }
15960: }
15961: printf("\n");
15962: fprintf(ficres,"\n");
15963: fprintf(ficlog,"\n");
15964: fprintf(fichtm, "</tr>");
15965: fprintf(fichtm, "\n");
15966:
15967:
1.126 brouard 15968: for(i=1,jk=1; i <=nlstate; i++){
15969: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 15970: if (k != i) {
1.319 brouard 15971: fprintf(fichtm, "<tr>");
1.225 brouard 15972: printf("%d%d ",i,k);
15973: fprintf(ficlog,"%d%d ",i,k);
15974: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 15975: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 15976: for(j=1; j <=ncovmodel; j++){
15977: printf("%12.7f ",p[jk]);
15978: fprintf(ficlog,"%12.7f ",p[jk]);
15979: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 15980: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 15981: jk++;
15982: }
15983: printf("\n");
15984: fprintf(ficlog,"\n");
15985: fprintf(ficres,"\n");
1.319 brouard 15986: fprintf(fichtm, "</tr>\n");
1.225 brouard 15987: }
1.126 brouard 15988: }
15989: }
1.319 brouard 15990: /* fprintf(fichtm,"</tr>\n"); */
15991: fprintf(fichtm,"</table>\n");
15992: fprintf(fichtm, "\n");
15993:
1.203 brouard 15994: if(mle != 0){
15995: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 15996: ftolhess=ftol; /* Usually correct */
1.203 brouard 15997: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
15998: 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");
15999: 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 16000: 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 16001: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
16002: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
16003: if(nagesqr==1){
16004: printf(" + age*age ");
16005: fprintf(ficres," + age*age ");
16006: fprintf(ficlog," + age*age ");
16007: fprintf(fichtm, "<th>+ age*age</th>");
16008: }
16009: for(j=1;j <=ncovmodel-2;j++){
16010: if(Typevar[j]==0) {
16011: printf(" + V%d ",Tvar[j]);
16012: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
16013: }else if(Typevar[j]==1) {
16014: printf(" + V%d*age ",Tvar[j]);
16015: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
16016: }else if(Typevar[j]==2) {
16017: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.349 brouard 16018: }else if(Typevar[j]==3) { /* TO VERIFY */
16019: fprintf(fichtm, "<th>+ V%d*V%d*age</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
1.319 brouard 16020: }
16021: }
16022: fprintf(fichtm, "</tr>\n");
16023:
1.203 brouard 16024: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 16025: for(k=1; k <=(nlstate+ndeath); k++){
16026: if (k != i) {
1.319 brouard 16027: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 16028: printf("%d%d ",i,k);
16029: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 16030: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 16031: for(j=1; j <=ncovmodel; j++){
1.319 brouard 16032: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 16033: 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]));
16034: 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 16035: if(fabs(wald) > 1.96){
1.321 brouard 16036: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 16037: }else{
16038: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
16039: }
1.324 brouard 16040: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 16041: 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 16042: jk++;
16043: }
16044: printf("\n");
16045: fprintf(ficlog,"\n");
1.319 brouard 16046: fprintf(fichtm, "</tr>\n");
1.225 brouard 16047: }
16048: }
1.193 brouard 16049: }
1.203 brouard 16050: } /* end of hesscov and Wald tests */
1.319 brouard 16051: fprintf(fichtm,"</table>\n");
1.225 brouard 16052:
1.203 brouard 16053: /* */
1.126 brouard 16054: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
16055: printf("# Scales (for hessian or gradient estimation)\n");
16056: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
16057: for(i=1,jk=1; i <=nlstate; i++){
16058: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 16059: if (j!=i) {
16060: fprintf(ficres,"%1d%1d",i,j);
16061: printf("%1d%1d",i,j);
16062: fprintf(ficlog,"%1d%1d",i,j);
16063: for(k=1; k<=ncovmodel;k++){
16064: printf(" %.5e",delti[jk]);
16065: fprintf(ficlog," %.5e",delti[jk]);
16066: fprintf(ficres," %.5e",delti[jk]);
16067: jk++;
16068: }
16069: printf("\n");
16070: fprintf(ficlog,"\n");
16071: fprintf(ficres,"\n");
16072: }
1.126 brouard 16073: }
16074: }
16075:
16076: 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 16077: if(mle >= 1) /* Too big for the screen */
1.126 brouard 16078: 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");
16079: 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");
16080: /* # 121 Var(a12)\n\ */
16081: /* # 122 Cov(b12,a12) Var(b12)\n\ */
16082: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
16083: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
16084: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
16085: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
16086: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
16087: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
16088:
16089:
16090: /* Just to have a covariance matrix which will be more understandable
16091: even is we still don't want to manage dictionary of variables
16092: */
16093: for(itimes=1;itimes<=2;itimes++){
16094: jj=0;
16095: for(i=1; i <=nlstate; i++){
1.225 brouard 16096: for(j=1; j <=nlstate+ndeath; j++){
16097: if(j==i) continue;
16098: for(k=1; k<=ncovmodel;k++){
16099: jj++;
16100: ca[0]= k+'a'-1;ca[1]='\0';
16101: if(itimes==1){
16102: if(mle>=1)
16103: printf("#%1d%1d%d",i,j,k);
16104: fprintf(ficlog,"#%1d%1d%d",i,j,k);
16105: fprintf(ficres,"#%1d%1d%d",i,j,k);
16106: }else{
16107: if(mle>=1)
16108: printf("%1d%1d%d",i,j,k);
16109: fprintf(ficlog,"%1d%1d%d",i,j,k);
16110: fprintf(ficres,"%1d%1d%d",i,j,k);
16111: }
16112: ll=0;
16113: for(li=1;li <=nlstate; li++){
16114: for(lj=1;lj <=nlstate+ndeath; lj++){
16115: if(lj==li) continue;
16116: for(lk=1;lk<=ncovmodel;lk++){
16117: ll++;
16118: if(ll<=jj){
16119: cb[0]= lk +'a'-1;cb[1]='\0';
16120: if(ll<jj){
16121: if(itimes==1){
16122: if(mle>=1)
16123: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
16124: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
16125: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
16126: }else{
16127: if(mle>=1)
16128: printf(" %.5e",matcov[jj][ll]);
16129: fprintf(ficlog," %.5e",matcov[jj][ll]);
16130: fprintf(ficres," %.5e",matcov[jj][ll]);
16131: }
16132: }else{
16133: if(itimes==1){
16134: if(mle>=1)
16135: printf(" Var(%s%1d%1d)",ca,i,j);
16136: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
16137: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
16138: }else{
16139: if(mle>=1)
16140: printf(" %.7e",matcov[jj][ll]);
16141: fprintf(ficlog," %.7e",matcov[jj][ll]);
16142: fprintf(ficres," %.7e",matcov[jj][ll]);
16143: }
16144: }
16145: }
16146: } /* end lk */
16147: } /* end lj */
16148: } /* end li */
16149: if(mle>=1)
16150: printf("\n");
16151: fprintf(ficlog,"\n");
16152: fprintf(ficres,"\n");
16153: numlinepar++;
16154: } /* end k*/
16155: } /*end j */
1.126 brouard 16156: } /* end i */
16157: } /* end itimes */
16158:
16159: fflush(ficlog);
16160: fflush(ficres);
1.225 brouard 16161: while(fgets(line, MAXLINE, ficpar)) {
16162: /* If line starts with a # it is a comment */
16163: if (line[0] == '#') {
16164: numlinepar++;
16165: fputs(line,stdout);
16166: fputs(line,ficparo);
16167: fputs(line,ficlog);
1.299 brouard 16168: fputs(line,ficres);
1.225 brouard 16169: continue;
16170: }else
16171: break;
16172: }
16173:
1.209 brouard 16174: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
16175: /* ungetc(c,ficpar); */
16176: /* fgets(line, MAXLINE, ficpar); */
16177: /* fputs(line,stdout); */
16178: /* fputs(line,ficparo); */
16179: /* } */
16180: /* ungetc(c,ficpar); */
1.126 brouard 16181:
16182: estepm=0;
1.209 brouard 16183: 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 16184:
16185: if (num_filled != 6) {
16186: 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);
16187: 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);
16188: goto end;
16189: }
16190: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
16191: }
16192: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
16193: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
16194:
1.209 brouard 16195: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 16196: if (estepm==0 || estepm < stepm) estepm=stepm;
16197: if (fage <= 2) {
16198: bage = ageminpar;
16199: fage = agemaxpar;
16200: }
16201:
16202: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 16203: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
16204: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 16205:
1.186 brouard 16206: /* Other stuffs, more or less useful */
1.254 brouard 16207: while(fgets(line, MAXLINE, ficpar)) {
16208: /* If line starts with a # it is a comment */
16209: if (line[0] == '#') {
16210: numlinepar++;
16211: fputs(line,stdout);
16212: fputs(line,ficparo);
16213: fputs(line,ficlog);
1.299 brouard 16214: fputs(line,ficres);
1.254 brouard 16215: continue;
16216: }else
16217: break;
16218: }
16219:
16220: 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){
16221:
16222: if (num_filled != 7) {
16223: 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);
16224: 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);
16225: goto end;
16226: }
16227: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
16228: 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);
16229: 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);
16230: 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 16231: }
1.254 brouard 16232:
16233: while(fgets(line, MAXLINE, ficpar)) {
16234: /* If line starts with a # it is a comment */
16235: if (line[0] == '#') {
16236: numlinepar++;
16237: fputs(line,stdout);
16238: fputs(line,ficparo);
16239: fputs(line,ficlog);
1.299 brouard 16240: fputs(line,ficres);
1.254 brouard 16241: continue;
16242: }else
16243: break;
1.126 brouard 16244: }
16245:
16246:
16247: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
16248: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
16249:
1.254 brouard 16250: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
16251: if (num_filled != 1) {
16252: 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);
16253: 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);
16254: goto end;
16255: }
16256: printf("pop_based=%d\n",popbased);
16257: fprintf(ficlog,"pop_based=%d\n",popbased);
16258: fprintf(ficparo,"pop_based=%d\n",popbased);
16259: fprintf(ficres,"pop_based=%d\n",popbased);
16260: }
16261:
1.258 brouard 16262: /* Results */
1.359 brouard 16263: /* Value of covariate in each resultine will be computed (if product) and sorted according to model rank */
1.332 brouard 16264: /* It is precov[] because we need the varying age in order to compute the real cov[] of the model equation */
16265: precov=matrix(1,MAXRESULTLINESPONE,1,NCOVMAX+1);
1.307 brouard 16266: endishere=0;
1.258 brouard 16267: nresult=0;
1.308 brouard 16268: parameterline=0;
1.258 brouard 16269: do{
16270: if(!fgets(line, MAXLINE, ficpar)){
16271: endishere=1;
1.308 brouard 16272: parameterline=15;
1.258 brouard 16273: }else if (line[0] == '#') {
16274: /* If line starts with a # it is a comment */
1.254 brouard 16275: numlinepar++;
16276: fputs(line,stdout);
16277: fputs(line,ficparo);
16278: fputs(line,ficlog);
1.299 brouard 16279: fputs(line,ficres);
1.254 brouard 16280: continue;
1.258 brouard 16281: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
16282: parameterline=11;
1.296 brouard 16283: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 16284: parameterline=12;
1.307 brouard 16285: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 16286: parameterline=13;
1.307 brouard 16287: }
1.258 brouard 16288: else{
16289: parameterline=14;
1.254 brouard 16290: }
1.308 brouard 16291: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 16292: case 11:
1.296 brouard 16293: 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)){
16294: 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 16295: 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);
16296: 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);
16297: 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);
16298: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 16299: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
16300: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 16301: prvforecast = 1;
16302: }
16303: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 16304: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
16305: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
16306: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 16307: prvforecast = 2;
16308: }
16309: else {
16310: 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);
16311: 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);
16312: goto end;
1.258 brouard 16313: }
1.254 brouard 16314: break;
1.258 brouard 16315: case 12:
1.296 brouard 16316: 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)){
16317: 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);
16318: 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);
16319: 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);
16320: 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);
16321: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 16322: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
16323: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 16324: prvbackcast = 1;
16325: }
16326: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 16327: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
16328: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
16329: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 16330: prvbackcast = 2;
16331: }
16332: else {
16333: 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);
16334: 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);
16335: goto end;
1.258 brouard 16336: }
1.230 brouard 16337: break;
1.258 brouard 16338: case 13:
1.332 brouard 16339: num_filled=sscanf(line,"result:%[^\n]\n",resultlineori);
1.307 brouard 16340: nresult++; /* Sum of resultlines */
1.342 brouard 16341: /* printf("Result %d: result:%s\n",nresult, resultlineori); */
1.332 brouard 16342: /* removefirstspace(&resultlineori); */
16343:
16344: if(strstr(resultlineori,"v") !=0){
16345: printf("Error. 'v' must be in upper case 'V' result: %s ",resultlineori);
16346: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultlineori);fflush(ficlog);
16347: return 1;
16348: }
16349: trimbb(resultline, resultlineori); /* Suppressing double blank in the resultline */
1.342 brouard 16350: /* printf("Decoderesult resultline=\"%s\" resultlineori=\"%s\"\n", resultline, resultlineori); */
1.318 brouard 16351: if(nresult > MAXRESULTLINESPONE-1){
16352: 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);
16353: 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 16354: goto end;
16355: }
1.332 brouard 16356:
1.310 brouard 16357: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 16358: fprintf(ficparo,"result: %s\n",resultline);
16359: fprintf(ficres,"result: %s\n",resultline);
16360: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 16361: } else
16362: goto end;
1.307 brouard 16363: break;
16364: case 14:
16365: printf("Error: Unknown command '%s'\n",line);
16366: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 16367: if(line[0] == ' ' || line[0] == '\n'){
16368: printf("It should not be an empty line '%s'\n",line);
16369: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
16370: }
1.307 brouard 16371: if(ncovmodel >=2 && nresult==0 ){
16372: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
16373: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 16374: }
1.307 brouard 16375: /* goto end; */
16376: break;
1.308 brouard 16377: case 15:
16378: printf("End of resultlines.\n");
16379: fprintf(ficlog,"End of resultlines.\n");
16380: break;
16381: default: /* parameterline =0 */
1.307 brouard 16382: nresult=1;
16383: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 16384: } /* End switch parameterline */
16385: }while(endishere==0); /* End do */
1.126 brouard 16386:
1.230 brouard 16387: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 16388: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 16389:
16390: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 16391: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 16392: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 16393: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
16394: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 16395: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 16396: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
16397: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 16398: }else{
1.270 brouard 16399: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 16400: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
16401: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
16402: if(prvforecast==1){
16403: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
16404: jprojd=jproj1;
16405: mprojd=mproj1;
16406: anprojd=anproj1;
16407: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
16408: jprojf=jproj2;
16409: mprojf=mproj2;
16410: anprojf=anproj2;
16411: } else if(prvforecast == 2){
16412: dateprojd=dateintmean;
16413: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
16414: dateprojf=dateintmean+yrfproj;
16415: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
16416: }
16417: if(prvbackcast==1){
16418: datebackd=(jback1+12*mback1+365*anback1)/365;
16419: jbackd=jback1;
16420: mbackd=mback1;
16421: anbackd=anback1;
16422: datebackf=(jback2+12*mback2+365*anback2)/365;
16423: jbackf=jback2;
16424: mbackf=mback2;
16425: anbackf=anback2;
16426: } else if(prvbackcast == 2){
16427: datebackd=dateintmean;
16428: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
16429: datebackf=dateintmean-yrbproj;
16430: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
16431: }
16432:
1.350 brouard 16433: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);/* HERE valgrind Tvard*/
1.220 brouard 16434: }
16435: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 16436: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
16437: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 16438:
1.225 brouard 16439: /*------------ free_vector -------------*/
16440: /* chdir(path); */
1.220 brouard 16441:
1.215 brouard 16442: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
16443: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
16444: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
16445: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 16446: free_lvector(num,firstobs,lastobs);
16447: free_vector(agedc,firstobs,lastobs);
1.126 brouard 16448: /*free_matrix(covar,0,NCOVMAX,1,n);*/
16449: /*free_matrix(covar,1,NCOVMAX,1,n);*/
16450: fclose(ficparo);
16451: fclose(ficres);
1.220 brouard 16452:
16453:
1.186 brouard 16454: /* Other results (useful)*/
1.220 brouard 16455:
16456:
1.126 brouard 16457: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 16458: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
16459: prlim=matrix(1,nlstate,1,nlstate);
1.332 brouard 16460: /* Computes the prevalence limit for each combination k of the dummy covariates by calling prevalim(k) */
1.209 brouard 16461: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 16462: fclose(ficrespl);
16463:
16464: /*------------- h Pij x at various ages ------------*/
1.180 brouard 16465: /*#include "hpijx.h"*/
1.332 brouard 16466: /** 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?*/
16467: /* calls hpxij with combination k */
1.180 brouard 16468: hPijx(p, bage, fage);
1.145 brouard 16469: fclose(ficrespij);
1.227 brouard 16470:
1.220 brouard 16471: /* ncovcombmax= pow(2,cptcoveff); */
1.332 brouard 16472: /*-------------- Variance of one-step probabilities for a combination ij or for nres ?---*/
1.145 brouard 16473: k=1;
1.126 brouard 16474: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 16475:
1.269 brouard 16476: /* Prevalence for each covariate combination in probs[age][status][cov] */
16477: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
16478: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 16479: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 16480: for(k=1;k<=ncovcombmax;k++)
16481: probs[i][j][k]=0.;
1.269 brouard 16482: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
16483: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 16484: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 16485: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
16486: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 16487: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 16488: for(k=1;k<=ncovcombmax;k++)
16489: mobaverages[i][j][k]=0.;
1.219 brouard 16490: mobaverage=mobaverages;
16491: if (mobilav!=0) {
1.235 brouard 16492: printf("Movingaveraging observed prevalence\n");
1.258 brouard 16493: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 16494: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
16495: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
16496: printf(" Error in movingaverage mobilav=%d\n",mobilav);
16497: }
1.269 brouard 16498: } else if (mobilavproj !=0) {
1.235 brouard 16499: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 16500: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 16501: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
16502: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
16503: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
16504: }
1.269 brouard 16505: }else{
16506: printf("Internal error moving average\n");
16507: fflush(stdout);
16508: exit(1);
1.219 brouard 16509: }
16510: }/* end if moving average */
1.227 brouard 16511:
1.126 brouard 16512: /*---------- Forecasting ------------------*/
1.296 brouard 16513: if(prevfcast==1){
16514: /* /\* if(stepm ==1){*\/ */
16515: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
16516: /*This done previously after freqsummary.*/
16517: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
16518: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
16519:
16520: /* } else if (prvforecast==2){ */
16521: /* /\* if(stepm ==1){*\/ */
16522: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
16523: /* } */
16524: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
16525: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 16526: }
1.269 brouard 16527:
1.296 brouard 16528: /* Prevbcasting */
16529: if(prevbcast==1){
1.219 brouard 16530: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
16531: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
16532: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
16533:
16534: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
16535:
16536: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 16537:
1.219 brouard 16538: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
16539: fclose(ficresplb);
16540:
1.222 brouard 16541: hBijx(p, bage, fage, mobaverage);
16542: fclose(ficrespijb);
1.219 brouard 16543:
1.296 brouard 16544: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
16545: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
16546: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
16547: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
16548: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
16549: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
16550:
16551:
1.269 brouard 16552: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 16553:
16554:
1.269 brouard 16555: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 16556: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
16557: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
16558: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 16559: } /* end Prevbcasting */
1.268 brouard 16560:
1.186 brouard 16561:
16562: /* ------ Other prevalence ratios------------ */
1.126 brouard 16563:
1.215 brouard 16564: free_ivector(wav,1,imx);
16565: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
16566: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
16567: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 16568:
16569:
1.127 brouard 16570: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 16571:
1.201 brouard 16572: strcpy(filerese,"E_");
16573: strcat(filerese,fileresu);
1.126 brouard 16574: if((ficreseij=fopen(filerese,"w"))==NULL) {
16575: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
16576: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
16577: }
1.208 brouard 16578: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
16579: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 16580:
16581: pstamp(ficreseij);
1.219 brouard 16582:
1.351 brouard 16583: /* i1=pow(2,cptcoveff); /\* Number of combination of dummy covariates *\/ */
16584: /* if (cptcovn < 1){i1=1;} */
1.235 brouard 16585:
1.351 brouard 16586: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
16587: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
16588: /* if(i1 != 1 && TKresult[nres]!= k) */
16589: /* continue; */
1.219 brouard 16590: fprintf(ficreseij,"\n#****** ");
1.235 brouard 16591: printf("\n#****** ");
1.351 brouard 16592: for(j=1;j<=cptcovs;j++){
16593: /* for(j=1;j<=cptcoveff;j++) { */
16594: /* fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
16595: fprintf(ficreseij," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
16596: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
16597: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.235 brouard 16598: }
16599: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.337 brouard 16600: printf(" V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]); /* TvarsQ[j] gives the name of the jth quantitative (fixed or time v) */
16601: fprintf(ficreseij,"V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]);
1.219 brouard 16602: }
16603: fprintf(ficreseij,"******\n");
1.235 brouard 16604: printf("******\n");
1.219 brouard 16605:
16606: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
16607: oldm=oldms;savm=savms;
1.330 brouard 16608: /* 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 16609: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 16610:
1.219 brouard 16611: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 16612: }
16613: fclose(ficreseij);
1.208 brouard 16614: printf("done evsij\n");fflush(stdout);
16615: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 16616:
1.218 brouard 16617:
1.227 brouard 16618: /*---------- State-specific expectancies and variances ------------*/
1.336 brouard 16619: /* Should be moved in a function */
1.201 brouard 16620: strcpy(filerest,"T_");
16621: strcat(filerest,fileresu);
1.127 brouard 16622: if((ficrest=fopen(filerest,"w"))==NULL) {
16623: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
16624: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
16625: }
1.208 brouard 16626: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
16627: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 16628: strcpy(fileresstde,"STDE_");
16629: strcat(fileresstde,fileresu);
1.126 brouard 16630: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 16631: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
16632: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 16633: }
1.227 brouard 16634: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
16635: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 16636:
1.201 brouard 16637: strcpy(filerescve,"CVE_");
16638: strcat(filerescve,fileresu);
1.126 brouard 16639: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 16640: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
16641: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 16642: }
1.227 brouard 16643: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
16644: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 16645:
1.201 brouard 16646: strcpy(fileresv,"V_");
16647: strcat(fileresv,fileresu);
1.126 brouard 16648: if((ficresvij=fopen(fileresv,"w"))==NULL) {
16649: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
16650: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
16651: }
1.227 brouard 16652: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
16653: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 16654:
1.235 brouard 16655: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
16656: if (cptcovn < 1){i1=1;}
16657:
1.334 brouard 16658: for(nres=1; nres <= nresult; nres++) /* For each resultline, find the combination and output results according to the values of dummies and then quanti. */
16659: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying. For each nres and each value at position k
16660: * we know Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline
16661: * Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline
16662: * and Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
16663: /* */
16664: 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 16665: continue;
1.359 brouard 16666: printf("\n# model=1+age+%s \n#****** Result for:", model); /* HERE model is empty */
16667: fprintf(ficrest,"\n# model=1+age+%s \n#****** Result for:", model);
16668: fprintf(ficlog,"\n# model=1+age+%s \n#****** Result for:", model);
1.334 brouard 16669: /* It might not be a good idea to mix dummies and quantitative */
16670: /* 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 *\/ */
16671: 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 */
16672: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* Output by variables in the resultline *\/ */
16673: /* Tvaraff[j] is the name of the dummy variable in position j in the equation model:
16674: * Tvaraff[1]@9={4, 3, 0, 0, 0, 0, 0, 0, 0}, in model=V5+V4+V3+V4*V3+V5*age
16675: * (V5 is quanti) V4 and V3 are dummies
16676: * TnsdVar[4] is the position 1 and TnsdVar[3]=2 in codtabm(k,l)(V4 V3)=V4 V3
16677: * l=1 l=2
16678: * k=1 1 1 0 0
16679: * k=2 2 1 1 0
16680: * k=3 [1] [2] 0 1
16681: * k=4 2 2 1 1
16682: * If nres=1 result: V3=1 V4=0 then k=3 and outputs
16683: * If nres=2 result: V4=1 V3=0 then k=2 and outputs
16684: * nres=1 =>k=3 j=1 V4= nbcode[4][codtabm(3,1)=1)=0; j=2 V3= nbcode[3][codtabm(3,2)=2]=1
16685: * nres=2 =>k=2 j=1 V4= nbcode[4][codtabm(2,1)=2)=1; j=2 V3= nbcode[3][codtabm(2,2)=1]=0
16686: */
16687: /* Tvresult[nres][j] Name of the variable at position j in this resultline */
16688: /* Tresult[nres][j] Value of this variable at position j could be a float if quantitative */
16689: /* We give up with the combinations!! */
1.342 brouard 16690: /* if(debugILK) */
16691: /* 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 16692:
16693: 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 16694: /* 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] */
16695: 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 */
16696: 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 */
16697: 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 16698: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
16699: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
16700: }else{
16701: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
16702: }
16703: /* fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
16704: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
16705: }else if(Dummy[modelresult[nres][j]]==1){ /* Quanti variable */
16706: /* For each selected (single) quantitative value */
1.337 brouard 16707: printf(" V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
16708: fprintf(ficlog," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
16709: fprintf(ficrest," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
1.334 brouard 16710: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
16711: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
16712: }else{
16713: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
16714: }
16715: }else{
16716: 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 */
16717: 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 */
16718: exit(1);
16719: }
1.335 brouard 16720: } /* End loop for each variable in the resultline */
1.334 brouard 16721: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
16722: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /\* Wrong j is not in the equation model *\/ */
16723: /* fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
16724: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
16725: /* } */
1.208 brouard 16726: fprintf(ficrest,"******\n");
1.227 brouard 16727: fprintf(ficlog,"******\n");
16728: printf("******\n");
1.208 brouard 16729:
16730: fprintf(ficresstdeij,"\n#****** ");
16731: fprintf(ficrescveij,"\n#****** ");
1.337 brouard 16732: /* It could have been: for(j=1;j<=cptcoveff;j++) {printf("V=%d=%lg",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);} */
16733: /* But it won't be sorted and depends on how the resultline is ordered */
1.225 brouard 16734: for(j=1;j<=cptcoveff;j++) {
1.334 brouard 16735: fprintf(ficresstdeij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
16736: /* fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
16737: /* fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
16738: }
16739: 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 16740: fprintf(ficresstdeij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
16741: fprintf(ficrescveij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
1.235 brouard 16742: }
1.208 brouard 16743: fprintf(ficresstdeij,"******\n");
16744: fprintf(ficrescveij,"******\n");
16745:
16746: fprintf(ficresvij,"\n#****** ");
1.238 brouard 16747: /* pstamp(ficresvij); */
1.225 brouard 16748: for(j=1;j<=cptcoveff;j++)
1.335 brouard 16749: fprintf(ficresvij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
16750: /* fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[TnsdVar[Tvaraff[j]]])]); */
1.235 brouard 16751: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 16752: /* fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]); /\* To solve *\/ */
1.337 brouard 16753: fprintf(ficresvij," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /* Solved */
1.235 brouard 16754: }
1.208 brouard 16755: fprintf(ficresvij,"******\n");
16756:
16757: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
16758: oldm=oldms;savm=savms;
1.235 brouard 16759: printf(" cvevsij ");
16760: fprintf(ficlog, " cvevsij ");
16761: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 16762: printf(" end cvevsij \n ");
16763: fprintf(ficlog, " end cvevsij \n ");
16764:
16765: /*
16766: */
16767: /* goto endfree; */
16768:
16769: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
16770: pstamp(ficrest);
16771:
1.269 brouard 16772: epj=vector(1,nlstate+1);
1.208 brouard 16773: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 16774: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
16775: cptcod= 0; /* To be deleted */
1.360 ! brouard 16776: printf("varevsij vpopbased=%d popbased=%d \n",vpopbased,popbased);
! 16777: fprintf(ficlog, "varevsij vpopbased=%d popbased=%d \n",vpopbased,popbased);
1.235 brouard 16778: 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 16779: fprintf(ficrest,"# Total life expectancy with std error and decomposition into time to be expected in each state\n\
! 16780: # (these are weighted average of eij where weights are ");
1.227 brouard 16781: if(vpopbased==1)
1.360 ! brouard 16782: 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 16783: else
1.360 ! brouard 16784: fprintf(ficrest,"the age specific forward period (stable) prevalences in each state) \n");
! 16785: fprintf(ficrest,"# with proportions of time spent in each state with standard error (on the right of the table.\n ");
1.335 brouard 16786: fprintf(ficrest,"# Age popbased mobilav e.. (std) "); /* Adding covariate values? */
1.227 brouard 16787: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
1.360 ! brouard 16788: for (i=1;i<=nlstate;i++) fprintf(ficrest," %% e.%d/e.. (std) ",i);
1.227 brouard 16789: fprintf(ficrest,"\n");
16790: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 16791: printf("Computing age specific forward period (stable) prevalences in each health state \n");
16792: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 16793: for(age=bage; age <=fage ;age++){
1.235 brouard 16794: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 16795: if (vpopbased==1) {
16796: if(mobilav ==0){
16797: for(i=1; i<=nlstate;i++)
16798: prlim[i][i]=probs[(int)age][i][k];
16799: }else{ /* mobilav */
16800: for(i=1; i<=nlstate;i++)
16801: prlim[i][i]=mobaverage[(int)age][i][k];
16802: }
16803: }
1.219 brouard 16804:
1.227 brouard 16805: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
16806: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
16807: /* printf(" age %4.0f ",age); */
16808: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
16809: for(i=1, epj[j]=0.;i <=nlstate;i++) {
16810: epj[j] += prlim[i][i]*eij[i][j][(int)age];
16811: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
16812: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
16813: }
16814: epj[nlstate+1] +=epj[j];
16815: }
16816: /* printf(" age %4.0f \n",age); */
1.219 brouard 16817:
1.227 brouard 16818: for(i=1, vepp=0.;i <=nlstate;i++)
16819: for(j=1;j <=nlstate;j++)
16820: vepp += vareij[i][j][(int)age];
16821: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
1.360 ! brouard 16822: /* vareij[j][i] is the variance of epj */
1.227 brouard 16823: for(j=1;j <=nlstate;j++){
16824: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
16825: }
1.360 ! brouard 16826: /* And proportion of time spent in state j */
! 16827: /* $$ E[r(X,Y)-E(r(X,Y))]^2=[\frac{1}{\mu_y} -\frac{\mu_x}{{\mu_y}^2}]' Var(X,Y)[\frac{1}{\mu_y} -\frac{\mu_x}{{\mu_y}^2}]$$ */
! 16828: /* \sigma^2_x/\mu_y^2 +\sigma^2_y \mu^2x/\mu_y^4 */
! 16829: /*\mu_x = epj[j], \sigma^2_x = vareij[j][j][(int)age] and \mu_y=epj[nlstate+1], \sigma^2_y=vepp */
! 16830: /* vareij[j][j][(int)age]/epj[nlstate+1]^2 + vepp/epj[nlstata+1]^4 */
! 16831: for(j=1;j <=nlstate;j++){
! 16832: /* 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] )); */
! 16833: fprintf(ficrest," %7.3f (%7.3f)", epj[j]/epj[nlstate+1], sqrt( vareij[j][j][(int)age]/epj[nlstate+1]/epj[nlstate+1] + vepp/epj[nlstate+1]/epj[nlstate+1]/epj[nlstate+1]/epj[nlstate+1] ));
! 16834: }
1.227 brouard 16835: fprintf(ficrest,"\n");
16836: }
1.208 brouard 16837: } /* End vpopbased */
1.269 brouard 16838: free_vector(epj,1,nlstate+1);
1.208 brouard 16839: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
16840: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 16841: printf("done selection\n");fflush(stdout);
16842: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 16843:
1.335 brouard 16844: } /* End k selection or end covariate selection for nres */
1.227 brouard 16845:
16846: printf("done State-specific expectancies\n");fflush(stdout);
16847: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
16848:
1.335 brouard 16849: /* variance-covariance of forward period prevalence */
1.269 brouard 16850: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 16851:
1.227 brouard 16852:
1.290 brouard 16853: free_vector(weight,firstobs,lastobs);
1.351 brouard 16854: free_imatrix(Tvardk,0,NCOVMAX,1,2);
1.227 brouard 16855: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 16856: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
16857: free_matrix(anint,1,maxwav,firstobs,lastobs);
16858: free_matrix(mint,1,maxwav,firstobs,lastobs);
16859: free_ivector(cod,firstobs,lastobs);
1.227 brouard 16860: free_ivector(tab,1,NCOVMAX);
16861: fclose(ficresstdeij);
16862: fclose(ficrescveij);
16863: fclose(ficresvij);
16864: fclose(ficrest);
16865: fclose(ficpar);
16866:
16867:
1.126 brouard 16868: /*---------- End : free ----------------*/
1.219 brouard 16869: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 16870: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
16871: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 16872: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
16873: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 16874: } /* mle==-3 arrives here for freeing */
1.227 brouard 16875: /* endfree:*/
1.359 brouard 16876: if(mle!=-3) free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /* Could be elsewhere ?*/
1.227 brouard 16877: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
16878: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
16879: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.341 brouard 16880: /* if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs); */
16881: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs);
1.290 brouard 16882: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
16883: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
16884: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 16885: free_matrix(matcov,1,npar,1,npar);
16886: free_matrix(hess,1,npar,1,npar);
16887: /*free_vector(delti,1,npar);*/
16888: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
16889: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 16890: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 16891: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
16892:
16893: free_ivector(ncodemax,1,NCOVMAX);
16894: free_ivector(ncodemaxwundef,1,NCOVMAX);
16895: free_ivector(Dummy,-1,NCOVMAX);
16896: free_ivector(Fixed,-1,NCOVMAX);
1.349 brouard 16897: free_ivector(DummyV,-1,NCOVMAX);
16898: free_ivector(FixedV,-1,NCOVMAX);
1.227 brouard 16899: free_ivector(Typevar,-1,NCOVMAX);
16900: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 16901: free_ivector(TvarsQ,1,NCOVMAX);
16902: free_ivector(TvarsQind,1,NCOVMAX);
16903: free_ivector(TvarsD,1,NCOVMAX);
1.330 brouard 16904: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 16905: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 16906: free_ivector(TvarFD,1,NCOVMAX);
16907: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 16908: free_ivector(TvarF,1,NCOVMAX);
16909: free_ivector(TvarFind,1,NCOVMAX);
16910: free_ivector(TvarV,1,NCOVMAX);
16911: free_ivector(TvarVind,1,NCOVMAX);
16912: free_ivector(TvarA,1,NCOVMAX);
16913: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 16914: free_ivector(TvarFQ,1,NCOVMAX);
16915: free_ivector(TvarFQind,1,NCOVMAX);
16916: free_ivector(TvarVD,1,NCOVMAX);
16917: free_ivector(TvarVDind,1,NCOVMAX);
16918: free_ivector(TvarVQ,1,NCOVMAX);
16919: free_ivector(TvarVQind,1,NCOVMAX);
1.349 brouard 16920: free_ivector(TvarAVVA,1,NCOVMAX);
16921: free_ivector(TvarAVVAind,1,NCOVMAX);
16922: free_ivector(TvarVVA,1,NCOVMAX);
16923: free_ivector(TvarVVAind,1,NCOVMAX);
1.339 brouard 16924: free_ivector(TvarVV,1,NCOVMAX);
16925: free_ivector(TvarVVind,1,NCOVMAX);
16926:
1.230 brouard 16927: free_ivector(Tvarsel,1,NCOVMAX);
16928: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 16929: free_ivector(Tposprod,1,NCOVMAX);
16930: free_ivector(Tprod,1,NCOVMAX);
16931: free_ivector(Tvaraff,1,NCOVMAX);
1.338 brouard 16932: free_ivector(invalidvarcomb,0,ncovcombmax);
1.227 brouard 16933: free_ivector(Tage,1,NCOVMAX);
16934: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 16935: free_ivector(TmodelInvind,1,NCOVMAX);
16936: free_ivector(TmodelInvQind,1,NCOVMAX);
1.332 brouard 16937:
1.359 brouard 16938: /* free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /\* Could be elsewhere ?*\/ */
1.332 brouard 16939:
1.227 brouard 16940: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
16941: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 16942: fflush(fichtm);
16943: fflush(ficgp);
16944:
1.227 brouard 16945:
1.126 brouard 16946: if((nberr >0) || (nbwarn>0)){
1.216 brouard 16947: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
16948: 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 16949: }else{
16950: printf("End of Imach\n");
16951: fprintf(ficlog,"End of Imach\n");
16952: }
16953: printf("See log file on %s\n",filelog);
16954: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 16955: /*(void) gettimeofday(&end_time,&tzp);*/
16956: rend_time = time(NULL);
16957: end_time = *localtime(&rend_time);
16958: /* tml = *localtime(&end_time.tm_sec); */
16959: strcpy(strtend,asctime(&end_time));
1.126 brouard 16960: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
16961: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 16962: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 16963:
1.157 brouard 16964: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
16965: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
16966: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 16967: /* printf("Total time was %d uSec.\n", total_usecs);*/
16968: /* if(fileappend(fichtm,optionfilehtm)){ */
16969: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
16970: fclose(fichtm);
16971: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
16972: fclose(fichtmcov);
16973: fclose(ficgp);
16974: fclose(ficlog);
16975: /*------ End -----------*/
1.227 brouard 16976:
1.281 brouard 16977:
16978: /* Executes gnuplot */
1.227 brouard 16979:
16980: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 16981: #ifdef WIN32
1.227 brouard 16982: if (_chdir(pathcd) != 0)
16983: printf("Can't move to directory %s!\n",path);
16984: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 16985: #else
1.227 brouard 16986: if(chdir(pathcd) != 0)
16987: printf("Can't move to directory %s!\n", path);
16988: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 16989: #endif
1.126 brouard 16990: printf("Current directory %s!\n",pathcd);
16991: /*strcat(plotcmd,CHARSEPARATOR);*/
16992: sprintf(plotcmd,"gnuplot");
1.157 brouard 16993: #ifdef _WIN32
1.126 brouard 16994: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
16995: #endif
16996: if(!stat(plotcmd,&info)){
1.158 brouard 16997: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 16998: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 16999: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 17000: }else
17001: strcpy(pplotcmd,plotcmd);
1.157 brouard 17002: #ifdef __unix
1.126 brouard 17003: strcpy(plotcmd,GNUPLOTPROGRAM);
17004: if(!stat(plotcmd,&info)){
1.158 brouard 17005: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 17006: }else
17007: strcpy(pplotcmd,plotcmd);
17008: #endif
17009: }else
17010: strcpy(pplotcmd,plotcmd);
17011:
17012: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 17013: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 17014: strcpy(pplotcmd,plotcmd);
1.227 brouard 17015:
1.126 brouard 17016: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 17017: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 17018: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 17019: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 17020: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 17021: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 17022: strcpy(plotcmd,pplotcmd);
17023: }
1.126 brouard 17024: }
1.158 brouard 17025: printf(" Successful, please wait...");
1.126 brouard 17026: while (z[0] != 'q') {
17027: /* chdir(path); */
1.154 brouard 17028: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 17029: scanf("%s",z);
17030: /* if (z[0] == 'c') system("./imach"); */
17031: if (z[0] == 'e') {
1.158 brouard 17032: #ifdef __APPLE__
1.152 brouard 17033: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 17034: #elif __linux
17035: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 17036: #else
1.152 brouard 17037: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 17038: #endif
17039: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
17040: system(pplotcmd);
1.126 brouard 17041: }
17042: else if (z[0] == 'g') system(plotcmd);
17043: else if (z[0] == 'q') exit(0);
17044: }
1.227 brouard 17045: end:
1.126 brouard 17046: while (z[0] != 'q') {
1.195 brouard 17047: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 17048: scanf("%s",z);
17049: }
1.283 brouard 17050: printf("End\n");
1.282 brouard 17051: exit(0);
1.126 brouard 17052: }
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