Annotation of imach/src/imach.c, revision 1.348
1.348 ! brouard 1: /* $Id: imach.c,v 1.347 2022/09/18 14:36:44 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.348 ! brouard 4: Revision 1.347 2022/09/18 14:36:44 brouard
! 5: Summary: version 0.99r42
! 6:
1.347 brouard 7: Revision 1.346 2022/09/16 13:52:36 brouard
8: * src/imach.c (Module): 0.99r41 Was an error when product of timevarying and fixed. Using FixedV[of name] now. Thank you Feinuo
9:
1.346 brouard 10: Revision 1.345 2022/09/16 13:40:11 brouard
11: Summary: Version 0.99r41
12:
13: * imach.c (Module): 0.99r41 Was an error when product of timevarying and fixed. Using FixedV[of name] now. Thank you Feinuo
14:
1.345 brouard 15: Revision 1.344 2022/09/14 19:33:30 brouard
16: Summary: version 0.99r40
17:
18: * imach.c (Module): Fixing names of variables in T_ (thanks to Feinuo)
19:
1.344 brouard 20: Revision 1.343 2022/09/14 14:22:16 brouard
21: Summary: version 0.99r39
22:
23: * imach.c (Module): Version 0.99r39 with colored dummy covariates
24: (fixed or time varying), using new last columns of
25: ILK_parameter.txt file.
26:
1.343 brouard 27: Revision 1.342 2022/09/11 19:54:09 brouard
28: Summary: 0.99r38
29:
30: * imach.c (Module): Adding timevarying products of any kinds,
31: should work before shifting cotvar from ncovcol+nqv columns in
32: order to have a correspondance between the column of cotvar and
33: the id of column.
34: (Module): Some cleaning and adding covariates in ILK.txt
35:
1.342 brouard 36: Revision 1.341 2022/09/11 07:58:42 brouard
37: Summary: Version 0.99r38
38:
39: After adding change in cotvar.
40:
1.341 brouard 41: Revision 1.340 2022/09/11 07:53:11 brouard
42: Summary: Version imach 0.99r37
43:
44: * imach.c (Module): Adding timevarying products of any kinds,
45: should work before shifting cotvar from ncovcol+nqv columns in
46: order to have a correspondance between the column of cotvar and
47: the id of column.
48:
1.340 brouard 49: Revision 1.339 2022/09/09 17:55:22 brouard
50: Summary: version 0.99r37
51:
52: * imach.c (Module): Many improvements for fixing products of fixed
53: timevarying as well as fixed * fixed, and test with quantitative
54: covariate.
55:
1.339 brouard 56: Revision 1.338 2022/09/04 17:40:33 brouard
57: Summary: 0.99r36
58:
59: * imach.c (Module): Now the easy runs i.e. without result or
60: model=1+age only did not work. The defautl combination should be 1
61: and not 0 because everything hasn't been tranformed yet.
62:
1.338 brouard 63: Revision 1.337 2022/09/02 14:26:02 brouard
64: Summary: version 0.99r35
65:
66: * src/imach.c: Version 0.99r35 because it outputs same results with
67: 1+age+V1+V1*age for females and 1+age for females only
68: (education=1 noweight)
69:
1.337 brouard 70: Revision 1.336 2022/08/31 09:52:36 brouard
71: *** empty log message ***
72:
1.336 brouard 73: Revision 1.335 2022/08/31 08:23:16 brouard
74: Summary: improvements...
75:
1.335 brouard 76: Revision 1.334 2022/08/25 09:08:41 brouard
77: Summary: In progress for quantitative
78:
1.334 brouard 79: Revision 1.333 2022/08/21 09:10:30 brouard
80: * src/imach.c (Module): Version 0.99r33 A lot of changes in
81: reassigning covariates: my first idea was that people will always
82: use the first covariate V1 into the model but in fact they are
83: producing data with many covariates and can use an equation model
84: with some of the covariate; it means that in a model V2+V3 instead
85: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
86: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
87: the equation model is restricted to two variables only (V2, V3)
88: and the combination for V2 should be codtabm(k,1) instead of
89: (codtabm(k,2), and the code should be
90: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
91: made. All of these should be simplified once a day like we did in
92: hpxij() for example by using precov[nres] which is computed in
93: decoderesult for each nres of each resultline. Loop should be done
94: on the equation model globally by distinguishing only product with
95: age (which are changing with age) and no more on type of
96: covariates, single dummies, single covariates.
97:
1.333 brouard 98: Revision 1.332 2022/08/21 09:06:25 brouard
99: Summary: Version 0.99r33
100:
101: * src/imach.c (Module): Version 0.99r33 A lot of changes in
102: reassigning covariates: my first idea was that people will always
103: use the first covariate V1 into the model but in fact they are
104: producing data with many covariates and can use an equation model
105: with some of the covariate; it means that in a model V2+V3 instead
106: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
107: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
108: the equation model is restricted to two variables only (V2, V3)
109: and the combination for V2 should be codtabm(k,1) instead of
110: (codtabm(k,2), and the code should be
111: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
112: made. All of these should be simplified once a day like we did in
113: hpxij() for example by using precov[nres] which is computed in
114: decoderesult for each nres of each resultline. Loop should be done
115: on the equation model globally by distinguishing only product with
116: age (which are changing with age) and no more on type of
117: covariates, single dummies, single covariates.
118:
1.332 brouard 119: Revision 1.331 2022/08/07 05:40:09 brouard
120: *** empty log message ***
121:
1.331 brouard 122: Revision 1.330 2022/08/06 07:18:25 brouard
123: Summary: last 0.99r31
124:
125: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function
126:
1.330 brouard 127: Revision 1.329 2022/08/03 17:29:54 brouard
128: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
129:
1.329 brouard 130: Revision 1.328 2022/07/27 17:40:48 brouard
131: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
132:
1.328 brouard 133: Revision 1.327 2022/07/27 14:47:35 brouard
134: Summary: Still a problem for one-step probabilities in case of quantitative variables
135:
1.327 brouard 136: Revision 1.326 2022/07/26 17:33:55 brouard
137: Summary: some test with nres=1
138:
1.326 brouard 139: Revision 1.325 2022/07/25 14:27:23 brouard
140: Summary: r30
141:
142: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
143: coredumped, revealed by Feiuno, thank you.
144:
1.325 brouard 145: Revision 1.324 2022/07/23 17:44:26 brouard
146: *** empty log message ***
147:
1.324 brouard 148: Revision 1.323 2022/07/22 12:30:08 brouard
149: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
150:
1.323 brouard 151: Revision 1.322 2022/07/22 12:27:48 brouard
152: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
153:
1.322 brouard 154: Revision 1.321 2022/07/22 12:04:24 brouard
155: Summary: r28
156:
157: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
158:
1.321 brouard 159: Revision 1.320 2022/06/02 05:10:11 brouard
160: *** empty log message ***
161:
1.320 brouard 162: Revision 1.319 2022/06/02 04:45:11 brouard
163: * imach.c (Module): Adding the Wald tests from the log to the main
164: htm for better display of the maximum likelihood estimators.
165:
1.319 brouard 166: Revision 1.318 2022/05/24 08:10:59 brouard
167: * imach.c (Module): Some attempts to find a bug of wrong estimates
168: of confidencce intervals with product in the equation modelC
169:
1.318 brouard 170: Revision 1.317 2022/05/15 15:06:23 brouard
171: * imach.c (Module): Some minor improvements
172:
1.317 brouard 173: Revision 1.316 2022/05/11 15:11:31 brouard
174: Summary: r27
175:
1.316 brouard 176: Revision 1.315 2022/05/11 15:06:32 brouard
177: *** empty log message ***
178:
1.315 brouard 179: Revision 1.314 2022/04/13 17:43:09 brouard
180: * imach.c (Module): Adding link to text data files
181:
1.314 brouard 182: Revision 1.313 2022/04/11 15:57:42 brouard
183: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
184:
1.313 brouard 185: Revision 1.312 2022/04/05 21:24:39 brouard
186: *** empty log message ***
187:
1.312 brouard 188: Revision 1.311 2022/04/05 21:03:51 brouard
189: Summary: Fixed quantitative covariates
190:
191: Fixed covariates (dummy or quantitative)
192: with missing values have never been allowed but are ERRORS and
193: program quits. Standard deviations of fixed covariates were
194: wrongly computed. Mean and standard deviations of time varying
195: covariates are still not computed.
196:
1.311 brouard 197: Revision 1.310 2022/03/17 08:45:53 brouard
198: Summary: 99r25
199:
200: Improving detection of errors: result lines should be compatible with
201: the model.
202:
1.310 brouard 203: Revision 1.309 2021/05/20 12:39:14 brouard
204: Summary: Version 0.99r24
205:
1.309 brouard 206: Revision 1.308 2021/03/31 13:11:57 brouard
207: Summary: Version 0.99r23
208:
209:
210: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
211:
1.308 brouard 212: Revision 1.307 2021/03/08 18:11:32 brouard
213: Summary: 0.99r22 fixed bug on result:
214:
1.307 brouard 215: Revision 1.306 2021/02/20 15:44:02 brouard
216: Summary: Version 0.99r21
217:
218: * imach.c (Module): Fix bug on quitting after result lines!
219: (Module): Version 0.99r21
220:
1.306 brouard 221: Revision 1.305 2021/02/20 15:28:30 brouard
222: * imach.c (Module): Fix bug on quitting after result lines!
223:
1.305 brouard 224: Revision 1.304 2021/02/12 11:34:20 brouard
225: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
226:
1.304 brouard 227: Revision 1.303 2021/02/11 19:50:15 brouard
228: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
229:
1.303 brouard 230: Revision 1.302 2020/02/22 21:00:05 brouard
231: * (Module): imach.c Update mle=-3 (for computing Life expectancy
232: and life table from the data without any state)
233:
1.302 brouard 234: Revision 1.301 2019/06/04 13:51:20 brouard
235: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
236:
1.301 brouard 237: Revision 1.300 2019/05/22 19:09:45 brouard
238: Summary: version 0.99r19 of May 2019
239:
1.300 brouard 240: Revision 1.299 2019/05/22 18:37:08 brouard
241: Summary: Cleaned 0.99r19
242:
1.299 brouard 243: Revision 1.298 2019/05/22 18:19:56 brouard
244: *** empty log message ***
245:
1.298 brouard 246: Revision 1.297 2019/05/22 17:56:10 brouard
247: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
248:
1.297 brouard 249: Revision 1.296 2019/05/20 13:03:18 brouard
250: Summary: Projection syntax simplified
251:
252:
253: We can now start projections, forward or backward, from the mean date
254: of inteviews up to or down to a number of years of projection:
255: prevforecast=1 yearsfproj=15.3 mobil_average=0
256: or
257: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
258: or
259: prevbackcast=1 yearsbproj=12.3 mobil_average=1
260: or
261: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
262:
1.296 brouard 263: Revision 1.295 2019/05/18 09:52:50 brouard
264: Summary: doxygen tex bug
265:
1.295 brouard 266: Revision 1.294 2019/05/16 14:54:33 brouard
267: Summary: There was some wrong lines added
268:
1.294 brouard 269: Revision 1.293 2019/05/09 15:17:34 brouard
270: *** empty log message ***
271:
1.293 brouard 272: Revision 1.292 2019/05/09 14:17:20 brouard
273: Summary: Some updates
274:
1.292 brouard 275: Revision 1.291 2019/05/09 13:44:18 brouard
276: Summary: Before ncovmax
277:
1.291 brouard 278: Revision 1.290 2019/05/09 13:39:37 brouard
279: Summary: 0.99r18 unlimited number of individuals
280:
281: 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.
282:
1.290 brouard 283: Revision 1.289 2018/12/13 09:16:26 brouard
284: Summary: Bug for young ages (<-30) will be in r17
285:
1.289 brouard 286: Revision 1.288 2018/05/02 20:58:27 brouard
287: Summary: Some bugs fixed
288:
1.288 brouard 289: Revision 1.287 2018/05/01 17:57:25 brouard
290: Summary: Bug fixed by providing frequencies only for non missing covariates
291:
1.287 brouard 292: Revision 1.286 2018/04/27 14:27:04 brouard
293: Summary: some minor bugs
294:
1.286 brouard 295: Revision 1.285 2018/04/21 21:02:16 brouard
296: Summary: Some bugs fixed, valgrind tested
297:
1.285 brouard 298: Revision 1.284 2018/04/20 05:22:13 brouard
299: Summary: Computing mean and stdeviation of fixed quantitative variables
300:
1.284 brouard 301: Revision 1.283 2018/04/19 14:49:16 brouard
302: Summary: Some minor bugs fixed
303:
1.283 brouard 304: Revision 1.282 2018/02/27 22:50:02 brouard
305: *** empty log message ***
306:
1.282 brouard 307: Revision 1.281 2018/02/27 19:25:23 brouard
308: Summary: Adding second argument for quitting
309:
1.281 brouard 310: Revision 1.280 2018/02/21 07:58:13 brouard
311: Summary: 0.99r15
312:
313: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
314:
1.280 brouard 315: Revision 1.279 2017/07/20 13:35:01 brouard
316: Summary: temporary working
317:
1.279 brouard 318: Revision 1.278 2017/07/19 14:09:02 brouard
319: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
320:
1.278 brouard 321: Revision 1.277 2017/07/17 08:53:49 brouard
322: Summary: BOM files can be read now
323:
1.277 brouard 324: Revision 1.276 2017/06/30 15:48:31 brouard
325: Summary: Graphs improvements
326:
1.276 brouard 327: Revision 1.275 2017/06/30 13:39:33 brouard
328: Summary: Saito's color
329:
1.275 brouard 330: Revision 1.274 2017/06/29 09:47:08 brouard
331: Summary: Version 0.99r14
332:
1.274 brouard 333: Revision 1.273 2017/06/27 11:06:02 brouard
334: Summary: More documentation on projections
335:
1.273 brouard 336: Revision 1.272 2017/06/27 10:22:40 brouard
337: Summary: Color of backprojection changed from 6 to 5(yellow)
338:
1.272 brouard 339: Revision 1.271 2017/06/27 10:17:50 brouard
340: Summary: Some bug with rint
341:
1.271 brouard 342: Revision 1.270 2017/05/24 05:45:29 brouard
343: *** empty log message ***
344:
1.270 brouard 345: Revision 1.269 2017/05/23 08:39:25 brouard
346: Summary: Code into subroutine, cleanings
347:
1.269 brouard 348: Revision 1.268 2017/05/18 20:09:32 brouard
349: Summary: backprojection and confidence intervals of backprevalence
350:
1.268 brouard 351: Revision 1.267 2017/05/13 10:25:05 brouard
352: Summary: temporary save for backprojection
353:
1.267 brouard 354: Revision 1.266 2017/05/13 07:26:12 brouard
355: Summary: Version 0.99r13 (improvements and bugs fixed)
356:
1.266 brouard 357: Revision 1.265 2017/04/26 16:22:11 brouard
358: Summary: imach 0.99r13 Some bugs fixed
359:
1.265 brouard 360: Revision 1.264 2017/04/26 06:01:29 brouard
361: Summary: Labels in graphs
362:
1.264 brouard 363: Revision 1.263 2017/04/24 15:23:15 brouard
364: Summary: to save
365:
1.263 brouard 366: Revision 1.262 2017/04/18 16:48:12 brouard
367: *** empty log message ***
368:
1.262 brouard 369: Revision 1.261 2017/04/05 10:14:09 brouard
370: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
371:
1.261 brouard 372: Revision 1.260 2017/04/04 17:46:59 brouard
373: Summary: Gnuplot indexations fixed (humm)
374:
1.260 brouard 375: Revision 1.259 2017/04/04 13:01:16 brouard
376: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
377:
1.259 brouard 378: Revision 1.258 2017/04/03 10:17:47 brouard
379: Summary: Version 0.99r12
380:
381: Some cleanings, conformed with updated documentation.
382:
1.258 brouard 383: Revision 1.257 2017/03/29 16:53:30 brouard
384: Summary: Temp
385:
1.257 brouard 386: Revision 1.256 2017/03/27 05:50:23 brouard
387: Summary: Temporary
388:
1.256 brouard 389: Revision 1.255 2017/03/08 16:02:28 brouard
390: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
391:
1.255 brouard 392: Revision 1.254 2017/03/08 07:13:00 brouard
393: Summary: Fixing data parameter line
394:
1.254 brouard 395: Revision 1.253 2016/12/15 11:59:41 brouard
396: Summary: 0.99 in progress
397:
1.253 brouard 398: Revision 1.252 2016/09/15 21:15:37 brouard
399: *** empty log message ***
400:
1.252 brouard 401: Revision 1.251 2016/09/15 15:01:13 brouard
402: Summary: not working
403:
1.251 brouard 404: Revision 1.250 2016/09/08 16:07:27 brouard
405: Summary: continue
406:
1.250 brouard 407: Revision 1.249 2016/09/07 17:14:18 brouard
408: Summary: Starting values from frequencies
409:
1.249 brouard 410: Revision 1.248 2016/09/07 14:10:18 brouard
411: *** empty log message ***
412:
1.248 brouard 413: Revision 1.247 2016/09/02 11:11:21 brouard
414: *** empty log message ***
415:
1.247 brouard 416: Revision 1.246 2016/09/02 08:49:22 brouard
417: *** empty log message ***
418:
1.246 brouard 419: Revision 1.245 2016/09/02 07:25:01 brouard
420: *** empty log message ***
421:
1.245 brouard 422: Revision 1.244 2016/09/02 07:17:34 brouard
423: *** empty log message ***
424:
1.244 brouard 425: Revision 1.243 2016/09/02 06:45:35 brouard
426: *** empty log message ***
427:
1.243 brouard 428: Revision 1.242 2016/08/30 15:01:20 brouard
429: Summary: Fixing a lots
430:
1.242 brouard 431: Revision 1.241 2016/08/29 17:17:25 brouard
432: Summary: gnuplot problem in Back projection to fix
433:
1.241 brouard 434: Revision 1.240 2016/08/29 07:53:18 brouard
435: Summary: Better
436:
1.240 brouard 437: Revision 1.239 2016/08/26 15:51:03 brouard
438: Summary: Improvement in Powell output in order to copy and paste
439:
440: Author:
441:
1.239 brouard 442: Revision 1.238 2016/08/26 14:23:35 brouard
443: Summary: Starting tests of 0.99
444:
1.238 brouard 445: Revision 1.237 2016/08/26 09:20:19 brouard
446: Summary: to valgrind
447:
1.237 brouard 448: Revision 1.236 2016/08/25 10:50:18 brouard
449: *** empty log message ***
450:
1.236 brouard 451: Revision 1.235 2016/08/25 06:59:23 brouard
452: *** empty log message ***
453:
1.235 brouard 454: Revision 1.234 2016/08/23 16:51:20 brouard
455: *** empty log message ***
456:
1.234 brouard 457: Revision 1.233 2016/08/23 07:40:50 brouard
458: Summary: not working
459:
1.233 brouard 460: Revision 1.232 2016/08/22 14:20:21 brouard
461: Summary: not working
462:
1.232 brouard 463: Revision 1.231 2016/08/22 07:17:15 brouard
464: Summary: not working
465:
1.231 brouard 466: Revision 1.230 2016/08/22 06:55:53 brouard
467: Summary: Not working
468:
1.230 brouard 469: Revision 1.229 2016/07/23 09:45:53 brouard
470: Summary: Completing for func too
471:
1.229 brouard 472: Revision 1.228 2016/07/22 17:45:30 brouard
473: Summary: Fixing some arrays, still debugging
474:
1.227 brouard 475: Revision 1.226 2016/07/12 18:42:34 brouard
476: Summary: temp
477:
1.226 brouard 478: Revision 1.225 2016/07/12 08:40:03 brouard
479: Summary: saving but not running
480:
1.225 brouard 481: Revision 1.224 2016/07/01 13:16:01 brouard
482: Summary: Fixes
483:
1.224 brouard 484: Revision 1.223 2016/02/19 09:23:35 brouard
485: Summary: temporary
486:
1.223 brouard 487: Revision 1.222 2016/02/17 08:14:50 brouard
488: Summary: Probably last 0.98 stable version 0.98r6
489:
1.222 brouard 490: Revision 1.221 2016/02/15 23:35:36 brouard
491: Summary: minor bug
492:
1.220 brouard 493: Revision 1.219 2016/02/15 00:48:12 brouard
494: *** empty log message ***
495:
1.219 brouard 496: Revision 1.218 2016/02/12 11:29:23 brouard
497: Summary: 0.99 Back projections
498:
1.218 brouard 499: Revision 1.217 2015/12/23 17:18:31 brouard
500: Summary: Experimental backcast
501:
1.217 brouard 502: Revision 1.216 2015/12/18 17:32:11 brouard
503: Summary: 0.98r4 Warning and status=-2
504:
505: Version 0.98r4 is now:
506: - displaying an error when status is -1, date of interview unknown and date of death known;
507: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
508: Older changes concerning s=-2, dating from 2005 have been supersed.
509:
1.216 brouard 510: Revision 1.215 2015/12/16 08:52:24 brouard
511: Summary: 0.98r4 working
512:
1.215 brouard 513: Revision 1.214 2015/12/16 06:57:54 brouard
514: Summary: temporary not working
515:
1.214 brouard 516: Revision 1.213 2015/12/11 18:22:17 brouard
517: Summary: 0.98r4
518:
1.213 brouard 519: Revision 1.212 2015/11/21 12:47:24 brouard
520: Summary: minor typo
521:
1.212 brouard 522: Revision 1.211 2015/11/21 12:41:11 brouard
523: Summary: 0.98r3 with some graph of projected cross-sectional
524:
525: Author: Nicolas Brouard
526:
1.211 brouard 527: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 528: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 529: Summary: Adding ftolpl parameter
530: Author: N Brouard
531:
532: We had difficulties to get smoothed confidence intervals. It was due
533: to the period prevalence which wasn't computed accurately. The inner
534: parameter ftolpl is now an outer parameter of the .imach parameter
535: file after estepm. If ftolpl is small 1.e-4 and estepm too,
536: computation are long.
537:
1.209 brouard 538: Revision 1.208 2015/11/17 14:31:57 brouard
539: Summary: temporary
540:
1.208 brouard 541: Revision 1.207 2015/10/27 17:36:57 brouard
542: *** empty log message ***
543:
1.207 brouard 544: Revision 1.206 2015/10/24 07:14:11 brouard
545: *** empty log message ***
546:
1.206 brouard 547: Revision 1.205 2015/10/23 15:50:53 brouard
548: Summary: 0.98r3 some clarification for graphs on likelihood contributions
549:
1.205 brouard 550: Revision 1.204 2015/10/01 16:20:26 brouard
551: Summary: Some new graphs of contribution to likelihood
552:
1.204 brouard 553: Revision 1.203 2015/09/30 17:45:14 brouard
554: Summary: looking at better estimation of the hessian
555:
556: Also a better criteria for convergence to the period prevalence And
557: therefore adding the number of years needed to converge. (The
558: prevalence in any alive state shold sum to one
559:
1.203 brouard 560: Revision 1.202 2015/09/22 19:45:16 brouard
561: Summary: Adding some overall graph on contribution to likelihood. Might change
562:
1.202 brouard 563: Revision 1.201 2015/09/15 17:34:58 brouard
564: Summary: 0.98r0
565:
566: - Some new graphs like suvival functions
567: - Some bugs fixed like model=1+age+V2.
568:
1.201 brouard 569: Revision 1.200 2015/09/09 16:53:55 brouard
570: Summary: Big bug thanks to Flavia
571:
572: Even model=1+age+V2. did not work anymore
573:
1.200 brouard 574: Revision 1.199 2015/09/07 14:09:23 brouard
575: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
576:
1.199 brouard 577: Revision 1.198 2015/09/03 07:14:39 brouard
578: Summary: 0.98q5 Flavia
579:
1.198 brouard 580: Revision 1.197 2015/09/01 18:24:39 brouard
581: *** empty log message ***
582:
1.197 brouard 583: Revision 1.196 2015/08/18 23:17:52 brouard
584: Summary: 0.98q5
585:
1.196 brouard 586: Revision 1.195 2015/08/18 16:28:39 brouard
587: Summary: Adding a hack for testing purpose
588:
589: After reading the title, ftol and model lines, if the comment line has
590: a q, starting with #q, the answer at the end of the run is quit. It
591: permits to run test files in batch with ctest. The former workaround was
592: $ echo q | imach foo.imach
593:
1.195 brouard 594: Revision 1.194 2015/08/18 13:32:00 brouard
595: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
596:
1.194 brouard 597: Revision 1.193 2015/08/04 07:17:42 brouard
598: Summary: 0.98q4
599:
1.193 brouard 600: Revision 1.192 2015/07/16 16:49:02 brouard
601: Summary: Fixing some outputs
602:
1.192 brouard 603: Revision 1.191 2015/07/14 10:00:33 brouard
604: Summary: Some fixes
605:
1.191 brouard 606: Revision 1.190 2015/05/05 08:51:13 brouard
607: Summary: Adding digits in output parameters (7 digits instead of 6)
608:
609: Fix 1+age+.
610:
1.190 brouard 611: Revision 1.189 2015/04/30 14:45:16 brouard
612: Summary: 0.98q2
613:
1.189 brouard 614: Revision 1.188 2015/04/30 08:27:53 brouard
615: *** empty log message ***
616:
1.188 brouard 617: Revision 1.187 2015/04/29 09:11:15 brouard
618: *** empty log message ***
619:
1.187 brouard 620: Revision 1.186 2015/04/23 12:01:52 brouard
621: Summary: V1*age is working now, version 0.98q1
622:
623: Some codes had been disabled in order to simplify and Vn*age was
624: working in the optimization phase, ie, giving correct MLE parameters,
625: but, as usual, outputs were not correct and program core dumped.
626:
1.186 brouard 627: Revision 1.185 2015/03/11 13:26:42 brouard
628: Summary: Inclusion of compile and links command line for Intel Compiler
629:
1.185 brouard 630: Revision 1.184 2015/03/11 11:52:39 brouard
631: Summary: Back from Windows 8. Intel Compiler
632:
1.184 brouard 633: Revision 1.183 2015/03/10 20:34:32 brouard
634: Summary: 0.98q0, trying with directest, mnbrak fixed
635:
636: We use directest instead of original Powell test; probably no
637: incidence on the results, but better justifications;
638: We fixed Numerical Recipes mnbrak routine which was wrong and gave
639: wrong results.
640:
1.183 brouard 641: Revision 1.182 2015/02/12 08:19:57 brouard
642: Summary: Trying to keep directest which seems simpler and more general
643: Author: Nicolas Brouard
644:
1.182 brouard 645: Revision 1.181 2015/02/11 23:22:24 brouard
646: Summary: Comments on Powell added
647:
648: Author:
649:
1.181 brouard 650: Revision 1.180 2015/02/11 17:33:45 brouard
651: Summary: Finishing move from main to function (hpijx and prevalence_limit)
652:
1.180 brouard 653: Revision 1.179 2015/01/04 09:57:06 brouard
654: Summary: back to OS/X
655:
1.179 brouard 656: Revision 1.178 2015/01/04 09:35:48 brouard
657: *** empty log message ***
658:
1.178 brouard 659: Revision 1.177 2015/01/03 18:40:56 brouard
660: Summary: Still testing ilc32 on OSX
661:
1.177 brouard 662: Revision 1.176 2015/01/03 16:45:04 brouard
663: *** empty log message ***
664:
1.176 brouard 665: Revision 1.175 2015/01/03 16:33:42 brouard
666: *** empty log message ***
667:
1.175 brouard 668: Revision 1.174 2015/01/03 16:15:49 brouard
669: Summary: Still in cross-compilation
670:
1.174 brouard 671: Revision 1.173 2015/01/03 12:06:26 brouard
672: Summary: trying to detect cross-compilation
673:
1.173 brouard 674: Revision 1.172 2014/12/27 12:07:47 brouard
675: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
676:
1.172 brouard 677: Revision 1.171 2014/12/23 13:26:59 brouard
678: Summary: Back from Visual C
679:
680: Still problem with utsname.h on Windows
681:
1.171 brouard 682: Revision 1.170 2014/12/23 11:17:12 brouard
683: Summary: Cleaning some \%% back to %%
684:
685: The escape was mandatory for a specific compiler (which one?), but too many warnings.
686:
1.170 brouard 687: Revision 1.169 2014/12/22 23:08:31 brouard
688: Summary: 0.98p
689:
690: Outputs some informations on compiler used, OS etc. Testing on different platforms.
691:
1.169 brouard 692: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 693: Summary: update
1.169 brouard 694:
1.168 brouard 695: Revision 1.167 2014/12/22 13:50:56 brouard
696: Summary: Testing uname and compiler version and if compiled 32 or 64
697:
698: Testing on Linux 64
699:
1.167 brouard 700: Revision 1.166 2014/12/22 11:40:47 brouard
701: *** empty log message ***
702:
1.166 brouard 703: Revision 1.165 2014/12/16 11:20:36 brouard
704: Summary: After compiling on Visual C
705:
706: * imach.c (Module): Merging 1.61 to 1.162
707:
1.165 brouard 708: Revision 1.164 2014/12/16 10:52:11 brouard
709: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
710:
711: * imach.c (Module): Merging 1.61 to 1.162
712:
1.164 brouard 713: Revision 1.163 2014/12/16 10:30:11 brouard
714: * imach.c (Module): Merging 1.61 to 1.162
715:
1.163 brouard 716: Revision 1.162 2014/09/25 11:43:39 brouard
717: Summary: temporary backup 0.99!
718:
1.162 brouard 719: Revision 1.1 2014/09/16 11:06:58 brouard
720: Summary: With some code (wrong) for nlopt
721:
722: Author:
723:
724: Revision 1.161 2014/09/15 20:41:41 brouard
725: Summary: Problem with macro SQR on Intel compiler
726:
1.161 brouard 727: Revision 1.160 2014/09/02 09:24:05 brouard
728: *** empty log message ***
729:
1.160 brouard 730: Revision 1.159 2014/09/01 10:34:10 brouard
731: Summary: WIN32
732: Author: Brouard
733:
1.159 brouard 734: Revision 1.158 2014/08/27 17:11:51 brouard
735: *** empty log message ***
736:
1.158 brouard 737: Revision 1.157 2014/08/27 16:26:55 brouard
738: Summary: Preparing windows Visual studio version
739: Author: Brouard
740:
741: In order to compile on Visual studio, time.h is now correct and time_t
742: and tm struct should be used. difftime should be used but sometimes I
743: just make the differences in raw time format (time(&now).
744: Trying to suppress #ifdef LINUX
745: Add xdg-open for __linux in order to open default browser.
746:
1.157 brouard 747: Revision 1.156 2014/08/25 20:10:10 brouard
748: *** empty log message ***
749:
1.156 brouard 750: Revision 1.155 2014/08/25 18:32:34 brouard
751: Summary: New compile, minor changes
752: Author: Brouard
753:
1.155 brouard 754: Revision 1.154 2014/06/20 17:32:08 brouard
755: Summary: Outputs now all graphs of convergence to period prevalence
756:
1.154 brouard 757: Revision 1.153 2014/06/20 16:45:46 brouard
758: Summary: If 3 live state, convergence to period prevalence on same graph
759: Author: Brouard
760:
1.153 brouard 761: Revision 1.152 2014/06/18 17:54:09 brouard
762: Summary: open browser, use gnuplot on same dir than imach if not found in the path
763:
1.152 brouard 764: Revision 1.151 2014/06/18 16:43:30 brouard
765: *** empty log message ***
766:
1.151 brouard 767: Revision 1.150 2014/06/18 16:42:35 brouard
768: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
769: Author: brouard
770:
1.150 brouard 771: Revision 1.149 2014/06/18 15:51:14 brouard
772: Summary: Some fixes in parameter files errors
773: Author: Nicolas Brouard
774:
1.149 brouard 775: Revision 1.148 2014/06/17 17:38:48 brouard
776: Summary: Nothing new
777: Author: Brouard
778:
779: Just a new packaging for OS/X version 0.98nS
780:
1.148 brouard 781: Revision 1.147 2014/06/16 10:33:11 brouard
782: *** empty log message ***
783:
1.147 brouard 784: Revision 1.146 2014/06/16 10:20:28 brouard
785: Summary: Merge
786: Author: Brouard
787:
788: Merge, before building revised version.
789:
1.146 brouard 790: Revision 1.145 2014/06/10 21:23:15 brouard
791: Summary: Debugging with valgrind
792: Author: Nicolas Brouard
793:
794: Lot of changes in order to output the results with some covariates
795: After the Edimburgh REVES conference 2014, it seems mandatory to
796: improve the code.
797: No more memory valgrind error but a lot has to be done in order to
798: continue the work of splitting the code into subroutines.
799: Also, decodemodel has been improved. Tricode is still not
800: optimal. nbcode should be improved. Documentation has been added in
801: the source code.
802:
1.144 brouard 803: Revision 1.143 2014/01/26 09:45:38 brouard
804: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
805:
806: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
807: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
808:
1.143 brouard 809: Revision 1.142 2014/01/26 03:57:36 brouard
810: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
811:
812: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
813:
1.142 brouard 814: Revision 1.141 2014/01/26 02:42:01 brouard
815: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
816:
1.141 brouard 817: Revision 1.140 2011/09/02 10:37:54 brouard
818: Summary: times.h is ok with mingw32 now.
819:
1.140 brouard 820: Revision 1.139 2010/06/14 07:50:17 brouard
821: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
822: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
823:
1.139 brouard 824: Revision 1.138 2010/04/30 18:19:40 brouard
825: *** empty log message ***
826:
1.138 brouard 827: Revision 1.137 2010/04/29 18:11:38 brouard
828: (Module): Checking covariates for more complex models
829: than V1+V2. A lot of change to be done. Unstable.
830:
1.137 brouard 831: Revision 1.136 2010/04/26 20:30:53 brouard
832: (Module): merging some libgsl code. Fixing computation
833: of likelione (using inter/intrapolation if mle = 0) in order to
834: get same likelihood as if mle=1.
835: Some cleaning of code and comments added.
836:
1.136 brouard 837: Revision 1.135 2009/10/29 15:33:14 brouard
838: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
839:
1.135 brouard 840: Revision 1.134 2009/10/29 13:18:53 brouard
841: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
842:
1.134 brouard 843: Revision 1.133 2009/07/06 10:21:25 brouard
844: just nforces
845:
1.133 brouard 846: Revision 1.132 2009/07/06 08:22:05 brouard
847: Many tings
848:
1.132 brouard 849: Revision 1.131 2009/06/20 16:22:47 brouard
850: Some dimensions resccaled
851:
1.131 brouard 852: Revision 1.130 2009/05/26 06:44:34 brouard
853: (Module): Max Covariate is now set to 20 instead of 8. A
854: lot of cleaning with variables initialized to 0. Trying to make
855: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
856:
1.130 brouard 857: Revision 1.129 2007/08/31 13:49:27 lievre
858: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
859:
1.129 lievre 860: Revision 1.128 2006/06/30 13:02:05 brouard
861: (Module): Clarifications on computing e.j
862:
1.128 brouard 863: Revision 1.127 2006/04/28 18:11:50 brouard
864: (Module): Yes the sum of survivors was wrong since
865: imach-114 because nhstepm was no more computed in the age
866: loop. Now we define nhstepma in the age loop.
867: (Module): In order to speed up (in case of numerous covariates) we
868: compute health expectancies (without variances) in a first step
869: and then all the health expectancies with variances or standard
870: deviation (needs data from the Hessian matrices) which slows the
871: computation.
872: In the future we should be able to stop the program is only health
873: expectancies and graph are needed without standard deviations.
874:
1.127 brouard 875: Revision 1.126 2006/04/28 17:23:28 brouard
876: (Module): Yes the sum of survivors was wrong since
877: imach-114 because nhstepm was no more computed in the age
878: loop. Now we define nhstepma in the age loop.
879: Version 0.98h
880:
1.126 brouard 881: Revision 1.125 2006/04/04 15:20:31 lievre
882: Errors in calculation of health expectancies. Age was not initialized.
883: Forecasting file added.
884:
885: Revision 1.124 2006/03/22 17:13:53 lievre
886: Parameters are printed with %lf instead of %f (more numbers after the comma).
887: The log-likelihood is printed in the log file
888:
889: Revision 1.123 2006/03/20 10:52:43 brouard
890: * imach.c (Module): <title> changed, corresponds to .htm file
891: name. <head> headers where missing.
892:
893: * imach.c (Module): Weights can have a decimal point as for
894: English (a comma might work with a correct LC_NUMERIC environment,
895: otherwise the weight is truncated).
896: Modification of warning when the covariates values are not 0 or
897: 1.
898: Version 0.98g
899:
900: Revision 1.122 2006/03/20 09:45:41 brouard
901: (Module): Weights can have a decimal point as for
902: English (a comma might work with a correct LC_NUMERIC environment,
903: otherwise the weight is truncated).
904: Modification of warning when the covariates values are not 0 or
905: 1.
906: Version 0.98g
907:
908: Revision 1.121 2006/03/16 17:45:01 lievre
909: * imach.c (Module): Comments concerning covariates added
910:
911: * imach.c (Module): refinements in the computation of lli if
912: status=-2 in order to have more reliable computation if stepm is
913: not 1 month. Version 0.98f
914:
915: Revision 1.120 2006/03/16 15:10:38 lievre
916: (Module): refinements in the computation of lli if
917: status=-2 in order to have more reliable computation if stepm is
918: not 1 month. Version 0.98f
919:
920: Revision 1.119 2006/03/15 17:42:26 brouard
921: (Module): Bug if status = -2, the loglikelihood was
922: computed as likelihood omitting the logarithm. Version O.98e
923:
924: Revision 1.118 2006/03/14 18:20:07 brouard
925: (Module): varevsij Comments added explaining the second
926: table of variances if popbased=1 .
927: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
928: (Module): Function pstamp added
929: (Module): Version 0.98d
930:
931: Revision 1.117 2006/03/14 17:16:22 brouard
932: (Module): varevsij Comments added explaining the second
933: table of variances if popbased=1 .
934: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
935: (Module): Function pstamp added
936: (Module): Version 0.98d
937:
938: Revision 1.116 2006/03/06 10:29:27 brouard
939: (Module): Variance-covariance wrong links and
940: varian-covariance of ej. is needed (Saito).
941:
942: Revision 1.115 2006/02/27 12:17:45 brouard
943: (Module): One freematrix added in mlikeli! 0.98c
944:
945: Revision 1.114 2006/02/26 12:57:58 brouard
946: (Module): Some improvements in processing parameter
947: filename with strsep.
948:
949: Revision 1.113 2006/02/24 14:20:24 brouard
950: (Module): Memory leaks checks with valgrind and:
951: datafile was not closed, some imatrix were not freed and on matrix
952: allocation too.
953:
954: Revision 1.112 2006/01/30 09:55:26 brouard
955: (Module): Back to gnuplot.exe instead of wgnuplot.exe
956:
957: Revision 1.111 2006/01/25 20:38:18 brouard
958: (Module): Lots of cleaning and bugs added (Gompertz)
959: (Module): Comments can be added in data file. Missing date values
960: can be a simple dot '.'.
961:
962: Revision 1.110 2006/01/25 00:51:50 brouard
963: (Module): Lots of cleaning and bugs added (Gompertz)
964:
965: Revision 1.109 2006/01/24 19:37:15 brouard
966: (Module): Comments (lines starting with a #) are allowed in data.
967:
968: Revision 1.108 2006/01/19 18:05:42 lievre
969: Gnuplot problem appeared...
970: To be fixed
971:
972: Revision 1.107 2006/01/19 16:20:37 brouard
973: Test existence of gnuplot in imach path
974:
975: Revision 1.106 2006/01/19 13:24:36 brouard
976: Some cleaning and links added in html output
977:
978: Revision 1.105 2006/01/05 20:23:19 lievre
979: *** empty log message ***
980:
981: Revision 1.104 2005/09/30 16:11:43 lievre
982: (Module): sump fixed, loop imx fixed, and simplifications.
983: (Module): If the status is missing at the last wave but we know
984: that the person is alive, then we can code his/her status as -2
985: (instead of missing=-1 in earlier versions) and his/her
986: contributions to the likelihood is 1 - Prob of dying from last
987: health status (= 1-p13= p11+p12 in the easiest case of somebody in
988: the healthy state at last known wave). Version is 0.98
989:
990: Revision 1.103 2005/09/30 15:54:49 lievre
991: (Module): sump fixed, loop imx fixed, and simplifications.
992:
993: Revision 1.102 2004/09/15 17:31:30 brouard
994: Add the possibility to read data file including tab characters.
995:
996: Revision 1.101 2004/09/15 10:38:38 brouard
997: Fix on curr_time
998:
999: Revision 1.100 2004/07/12 18:29:06 brouard
1000: Add version for Mac OS X. Just define UNIX in Makefile
1001:
1002: Revision 1.99 2004/06/05 08:57:40 brouard
1003: *** empty log message ***
1004:
1005: Revision 1.98 2004/05/16 15:05:56 brouard
1006: New version 0.97 . First attempt to estimate force of mortality
1007: directly from the data i.e. without the need of knowing the health
1008: state at each age, but using a Gompertz model: log u =a + b*age .
1009: This is the basic analysis of mortality and should be done before any
1010: other analysis, in order to test if the mortality estimated from the
1011: cross-longitudinal survey is different from the mortality estimated
1012: from other sources like vital statistic data.
1013:
1014: The same imach parameter file can be used but the option for mle should be -3.
1015:
1.324 brouard 1016: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 1017: former routines in order to include the new code within the former code.
1018:
1019: The output is very simple: only an estimate of the intercept and of
1020: the slope with 95% confident intervals.
1021:
1022: Current limitations:
1023: A) Even if you enter covariates, i.e. with the
1024: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
1025: B) There is no computation of Life Expectancy nor Life Table.
1026:
1027: Revision 1.97 2004/02/20 13:25:42 lievre
1028: Version 0.96d. Population forecasting command line is (temporarily)
1029: suppressed.
1030:
1031: Revision 1.96 2003/07/15 15:38:55 brouard
1032: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
1033: rewritten within the same printf. Workaround: many printfs.
1034:
1035: Revision 1.95 2003/07/08 07:54:34 brouard
1036: * imach.c (Repository):
1037: (Repository): Using imachwizard code to output a more meaningful covariance
1038: matrix (cov(a12,c31) instead of numbers.
1039:
1040: Revision 1.94 2003/06/27 13:00:02 brouard
1041: Just cleaning
1042:
1043: Revision 1.93 2003/06/25 16:33:55 brouard
1044: (Module): On windows (cygwin) function asctime_r doesn't
1045: exist so I changed back to asctime which exists.
1046: (Module): Version 0.96b
1047:
1048: Revision 1.92 2003/06/25 16:30:45 brouard
1049: (Module): On windows (cygwin) function asctime_r doesn't
1050: exist so I changed back to asctime which exists.
1051:
1052: Revision 1.91 2003/06/25 15:30:29 brouard
1053: * imach.c (Repository): Duplicated warning errors corrected.
1054: (Repository): Elapsed time after each iteration is now output. It
1055: helps to forecast when convergence will be reached. Elapsed time
1056: is stamped in powell. We created a new html file for the graphs
1057: concerning matrix of covariance. It has extension -cov.htm.
1058:
1059: Revision 1.90 2003/06/24 12:34:15 brouard
1060: (Module): Some bugs corrected for windows. Also, when
1061: mle=-1 a template is output in file "or"mypar.txt with the design
1062: of the covariance matrix to be input.
1063:
1064: Revision 1.89 2003/06/24 12:30:52 brouard
1065: (Module): Some bugs corrected for windows. Also, when
1066: mle=-1 a template is output in file "or"mypar.txt with the design
1067: of the covariance matrix to be input.
1068:
1069: Revision 1.88 2003/06/23 17:54:56 brouard
1070: * 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.
1071:
1072: Revision 1.87 2003/06/18 12:26:01 brouard
1073: Version 0.96
1074:
1075: Revision 1.86 2003/06/17 20:04:08 brouard
1076: (Module): Change position of html and gnuplot routines and added
1077: routine fileappend.
1078:
1079: Revision 1.85 2003/06/17 13:12:43 brouard
1080: * imach.c (Repository): Check when date of death was earlier that
1081: current date of interview. It may happen when the death was just
1082: prior to the death. In this case, dh was negative and likelihood
1083: was wrong (infinity). We still send an "Error" but patch by
1084: assuming that the date of death was just one stepm after the
1085: interview.
1086: (Repository): Because some people have very long ID (first column)
1087: we changed int to long in num[] and we added a new lvector for
1088: memory allocation. But we also truncated to 8 characters (left
1089: truncation)
1090: (Repository): No more line truncation errors.
1091:
1092: Revision 1.84 2003/06/13 21:44:43 brouard
1093: * imach.c (Repository): Replace "freqsummary" at a correct
1094: place. It differs from routine "prevalence" which may be called
1095: many times. Probs is memory consuming and must be used with
1096: parcimony.
1097: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
1098:
1099: Revision 1.83 2003/06/10 13:39:11 lievre
1100: *** empty log message ***
1101:
1102: Revision 1.82 2003/06/05 15:57:20 brouard
1103: Add log in imach.c and fullversion number is now printed.
1104:
1105: */
1106: /*
1107: Interpolated Markov Chain
1108:
1109: Short summary of the programme:
1110:
1.227 brouard 1111: This program computes Healthy Life Expectancies or State-specific
1112: (if states aren't health statuses) Expectancies from
1113: cross-longitudinal data. Cross-longitudinal data consist in:
1114:
1115: -1- a first survey ("cross") where individuals from different ages
1116: are interviewed on their health status or degree of disability (in
1117: the case of a health survey which is our main interest)
1118:
1119: -2- at least a second wave of interviews ("longitudinal") which
1120: measure each change (if any) in individual health status. Health
1121: expectancies are computed from the time spent in each health state
1122: according to a model. More health states you consider, more time is
1123: necessary to reach the Maximum Likelihood of the parameters involved
1124: in the model. The simplest model is the multinomial logistic model
1125: where pij is the probability to be observed in state j at the second
1126: wave conditional to be observed in state i at the first
1127: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1128: etc , where 'age' is age and 'sex' is a covariate. If you want to
1129: have a more complex model than "constant and age", you should modify
1130: the program where the markup *Covariates have to be included here
1131: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1132: convergence.
1133:
1134: The advantage of this computer programme, compared to a simple
1135: multinomial logistic model, is clear when the delay between waves is not
1136: identical for each individual. Also, if a individual missed an
1137: intermediate interview, the information is lost, but taken into
1138: account using an interpolation or extrapolation.
1139:
1140: hPijx is the probability to be observed in state i at age x+h
1141: conditional to the observed state i at age x. The delay 'h' can be
1142: split into an exact number (nh*stepm) of unobserved intermediate
1143: states. This elementary transition (by month, quarter,
1144: semester or year) is modelled as a multinomial logistic. The hPx
1145: matrix is simply the matrix product of nh*stepm elementary matrices
1146: and the contribution of each individual to the likelihood is simply
1147: hPijx.
1148:
1149: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1150: of the life expectancies. It also computes the period (stable) prevalence.
1151:
1152: Back prevalence and projections:
1.227 brouard 1153:
1154: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1155: double agemaxpar, double ftolpl, int *ncvyearp, double
1156: dateprev1,double dateprev2, int firstpass, int lastpass, int
1157: mobilavproj)
1158:
1159: Computes the back prevalence limit for any combination of
1160: covariate values k at any age between ageminpar and agemaxpar and
1161: returns it in **bprlim. In the loops,
1162:
1163: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1164: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1165:
1166: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1167: Computes for any combination of covariates k and any age between bage and fage
1168: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1169: oldm=oldms;savm=savms;
1.227 brouard 1170:
1.267 brouard 1171: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1172: Computes the transition matrix starting at age 'age' over
1173: 'nhstepm*hstepm*stepm' months (i.e. until
1174: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1175: nhstepm*hstepm matrices.
1176:
1177: Returns p3mat[i][j][h] after calling
1178: p3mat[i][j][h]=matprod2(newm,
1179: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1180: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1181: oldm);
1.226 brouard 1182:
1183: Important routines
1184:
1185: - func (or funcone), computes logit (pij) distinguishing
1186: o fixed variables (single or product dummies or quantitative);
1187: o varying variables by:
1188: (1) wave (single, product dummies, quantitative),
1189: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1190: % fixed dummy (treated) or quantitative (not done because time-consuming);
1191: % varying dummy (not done) or quantitative (not done);
1192: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1193: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1194: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1195: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1196: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1197:
1.226 brouard 1198:
1199:
1.324 brouard 1200: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1201: Institut national d'études démographiques, Paris.
1.126 brouard 1202: This software have been partly granted by Euro-REVES, a concerted action
1203: from the European Union.
1204: It is copyrighted identically to a GNU software product, ie programme and
1205: software can be distributed freely for non commercial use. Latest version
1206: can be accessed at http://euroreves.ined.fr/imach .
1207:
1208: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1209: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1210:
1211: **********************************************************************/
1212: /*
1213: main
1214: read parameterfile
1215: read datafile
1216: concatwav
1217: freqsummary
1218: if (mle >= 1)
1219: mlikeli
1220: print results files
1221: if mle==1
1222: computes hessian
1223: read end of parameter file: agemin, agemax, bage, fage, estepm
1224: begin-prev-date,...
1225: open gnuplot file
1226: open html file
1.145 brouard 1227: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1228: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1229: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1230: freexexit2 possible for memory heap.
1231:
1232: h Pij x | pij_nom ficrestpij
1233: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1234: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1235: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1236:
1237: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1238: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1239: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1240: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1241: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1242:
1.126 brouard 1243: forecasting if prevfcast==1 prevforecast call prevalence()
1244: health expectancies
1245: Variance-covariance of DFLE
1246: prevalence()
1247: movingaverage()
1248: varevsij()
1249: if popbased==1 varevsij(,popbased)
1250: total life expectancies
1251: Variance of period (stable) prevalence
1252: end
1253: */
1254:
1.187 brouard 1255: /* #define DEBUG */
1256: /* #define DEBUGBRENT */
1.203 brouard 1257: /* #define DEBUGLINMIN */
1258: /* #define DEBUGHESS */
1259: #define DEBUGHESSIJ
1.224 brouard 1260: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1261: #define POWELL /* Instead of NLOPT */
1.224 brouard 1262: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1263: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1264: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1265: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1266:
1267: #include <math.h>
1268: #include <stdio.h>
1269: #include <stdlib.h>
1270: #include <string.h>
1.226 brouard 1271: #include <ctype.h>
1.159 brouard 1272:
1273: #ifdef _WIN32
1274: #include <io.h>
1.172 brouard 1275: #include <windows.h>
1276: #include <tchar.h>
1.159 brouard 1277: #else
1.126 brouard 1278: #include <unistd.h>
1.159 brouard 1279: #endif
1.126 brouard 1280:
1281: #include <limits.h>
1282: #include <sys/types.h>
1.171 brouard 1283:
1284: #if defined(__GNUC__)
1285: #include <sys/utsname.h> /* Doesn't work on Windows */
1286: #endif
1287:
1.126 brouard 1288: #include <sys/stat.h>
1289: #include <errno.h>
1.159 brouard 1290: /* extern int errno; */
1.126 brouard 1291:
1.157 brouard 1292: /* #ifdef LINUX */
1293: /* #include <time.h> */
1294: /* #include "timeval.h" */
1295: /* #else */
1296: /* #include <sys/time.h> */
1297: /* #endif */
1298:
1.126 brouard 1299: #include <time.h>
1300:
1.136 brouard 1301: #ifdef GSL
1302: #include <gsl/gsl_errno.h>
1303: #include <gsl/gsl_multimin.h>
1304: #endif
1305:
1.167 brouard 1306:
1.162 brouard 1307: #ifdef NLOPT
1308: #include <nlopt.h>
1309: typedef struct {
1310: double (* function)(double [] );
1311: } myfunc_data ;
1312: #endif
1313:
1.126 brouard 1314: /* #include <libintl.h> */
1315: /* #define _(String) gettext (String) */
1316:
1.251 brouard 1317: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1318:
1319: #define GNUPLOTPROGRAM "gnuplot"
1.343 brouard 1320: #define GNUPLOTVERSION 5.1
1321: double gnuplotversion=GNUPLOTVERSION;
1.126 brouard 1322: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1323: #define FILENAMELENGTH 256
1.126 brouard 1324:
1325: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1326: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1327:
1.144 brouard 1328: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1329: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1330:
1331: #define NINTERVMAX 8
1.144 brouard 1332: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1333: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1334: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1335: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1336: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1337: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1338: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1339: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1340: /* #define AGESUP 130 */
1.288 brouard 1341: /* #define AGESUP 150 */
1342: #define AGESUP 200
1.268 brouard 1343: #define AGEINF 0
1.218 brouard 1344: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1345: #define AGEBASE 40
1.194 brouard 1346: #define AGEOVERFLOW 1.e20
1.164 brouard 1347: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1348: #ifdef _WIN32
1349: #define DIRSEPARATOR '\\'
1350: #define CHARSEPARATOR "\\"
1351: #define ODIRSEPARATOR '/'
1352: #else
1.126 brouard 1353: #define DIRSEPARATOR '/'
1354: #define CHARSEPARATOR "/"
1355: #define ODIRSEPARATOR '\\'
1356: #endif
1357:
1.348 ! brouard 1358: /* $Id: imach.c,v 1.347 2022/09/18 14:36:44 brouard Exp $ */
1.126 brouard 1359: /* $State: Exp $ */
1.196 brouard 1360: #include "version.h"
1361: char version[]=__IMACH_VERSION__;
1.337 brouard 1362: char copyright[]="September 2022,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 2000-2022";
1.348 ! brouard 1363: char fullversion[]="$Revision: 1.347 $ $Date: 2022/09/18 14:36:44 $";
1.126 brouard 1364: char strstart[80];
1365: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1366: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.342 brouard 1367: int debugILK=0; /* debugILK is set by a #d in a comment line */
1.187 brouard 1368: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 brouard 1369: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
1370: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
1.335 brouard 1371: 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 1372: 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 1373: int cptcovs=0; /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
1374: int cptcovsnq=0; /**< cptcovsnq number of SIMPLE covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1375: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1376: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.335 brouard 1377: 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 1378: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1379: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.339 brouard 1380: int ncovvt=0; /* Total number of effective (wave) varying covariates (dummy or quantitative or products [without age]) in the model */
1.232 brouard 1381: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1382: int nsd=0; /**< Total number of single dummy variables (output) */
1383: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1384: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1385: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1386: int ntveff=0; /**< ntveff number of effective time varying variables */
1387: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1388: int cptcov=0; /* Working variable */
1.334 brouard 1389: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs+1 declared globally ;*/
1.290 brouard 1390: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1391: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1392: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1393: int nlstate=2; /* Number of live states */
1394: int ndeath=1; /* Number of dead states */
1.130 brouard 1395: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.339 brouard 1396: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable*/
1397: int ncovcolt=0; /* ncovcolt=ncovcol+nqv+ntv+nqtv; total of covariates in the data, not in the model equation*/
1.126 brouard 1398: int popbased=0;
1399:
1400: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1401: int maxwav=0; /* Maxim number of waves */
1402: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1403: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1404: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1405: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1406: int mle=1, weightopt=0;
1.126 brouard 1407: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1408: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1409: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1410: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1411: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1412: int selected(int kvar); /* Is covariate kvar selected for printing results */
1413:
1.130 brouard 1414: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1415: double **matprod2(); /* test */
1.126 brouard 1416: double **oldm, **newm, **savm; /* Working pointers to matrices */
1417: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1418: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1419:
1.136 brouard 1420: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1421: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1422: FILE *ficlog, *ficrespow;
1.130 brouard 1423: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1424: double fretone; /* Only one call to likelihood */
1.130 brouard 1425: long ipmx=0; /* Number of contributions */
1.126 brouard 1426: double sw; /* Sum of weights */
1427: char filerespow[FILENAMELENGTH];
1428: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1429: FILE *ficresilk;
1430: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1431: FILE *ficresprobmorprev;
1432: FILE *fichtm, *fichtmcov; /* Html File */
1433: FILE *ficreseij;
1434: char filerese[FILENAMELENGTH];
1435: FILE *ficresstdeij;
1436: char fileresstde[FILENAMELENGTH];
1437: FILE *ficrescveij;
1438: char filerescve[FILENAMELENGTH];
1439: FILE *ficresvij;
1440: char fileresv[FILENAMELENGTH];
1.269 brouard 1441:
1.126 brouard 1442: char title[MAXLINE];
1.234 brouard 1443: char model[MAXLINE]; /**< The model line */
1.217 brouard 1444: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1445: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1446: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1447: char command[FILENAMELENGTH];
1448: int outcmd=0;
1449:
1.217 brouard 1450: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1451: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1452: char filelog[FILENAMELENGTH]; /* Log file */
1453: char filerest[FILENAMELENGTH];
1454: char fileregp[FILENAMELENGTH];
1455: char popfile[FILENAMELENGTH];
1456:
1457: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1458:
1.157 brouard 1459: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1460: /* struct timezone tzp; */
1461: /* extern int gettimeofday(); */
1462: struct tm tml, *gmtime(), *localtime();
1463:
1464: extern time_t time();
1465:
1466: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1467: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1468: struct tm tm;
1469:
1.126 brouard 1470: char strcurr[80], strfor[80];
1471:
1472: char *endptr;
1473: long lval;
1474: double dval;
1475:
1476: #define NR_END 1
1477: #define FREE_ARG char*
1478: #define FTOL 1.0e-10
1479:
1480: #define NRANSI
1.240 brouard 1481: #define ITMAX 200
1482: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1483:
1484: #define TOL 2.0e-4
1485:
1486: #define CGOLD 0.3819660
1487: #define ZEPS 1.0e-10
1488: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1489:
1490: #define GOLD 1.618034
1491: #define GLIMIT 100.0
1492: #define TINY 1.0e-20
1493:
1494: static double maxarg1,maxarg2;
1495: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1496: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1497:
1498: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1499: #define rint(a) floor(a+0.5)
1.166 brouard 1500: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1501: #define mytinydouble 1.0e-16
1.166 brouard 1502: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1503: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1504: /* static double dsqrarg; */
1505: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1506: static double sqrarg;
1507: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1508: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1509: int agegomp= AGEGOMP;
1510:
1511: int imx;
1512: int stepm=1;
1513: /* Stepm, step in month: minimum step interpolation*/
1514:
1515: int estepm;
1516: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1517:
1518: int m,nb;
1519: long *num;
1.197 brouard 1520: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1521: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1522: covariate for which somebody answered excluding
1523: undefined. Usually 2: 0 and 1. */
1524: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1525: covariate for which somebody answered including
1526: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1527: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1528: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1529: double ***mobaverage, ***mobaverages; /* New global variable */
1.332 brouard 1530: 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 1531: double *ageexmed,*agecens;
1532: double dateintmean=0;
1.296 brouard 1533: double anprojd, mprojd, jprojd; /* For eventual projections */
1534: double anprojf, mprojf, jprojf;
1.126 brouard 1535:
1.296 brouard 1536: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1537: double anbackf, mbackf, jbackf;
1538: double jintmean,mintmean,aintmean;
1.126 brouard 1539: double *weight;
1540: int **s; /* Status */
1.141 brouard 1541: double *agedc;
1.145 brouard 1542: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1543: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1544: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1545: double **coqvar; /* Fixed quantitative covariate nqv */
1.341 brouard 1546: double ***cotvar; /* Time varying covariate start at ncovcol + nqv + (1 to ntv) */
1.225 brouard 1547: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1548: double idx;
1549: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1550: /* Some documentation */
1551: /* Design original data
1552: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1553: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1554: * ntv=3 nqtv=1
1.330 brouard 1555: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1556: * For time varying covariate, quanti or dummies
1557: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1.341 brouard 1558: * cotvar[wav][ncovcol+nqv+ iv(1 to nqtv)][i]= [(1 to nqtv)][i]=(V12) quanti
1.319 brouard 1559: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1560: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1.332 brouard 1561: * covar[Vk,i], value of the Vkth fixed covariate dummy or quanti for individual i:
1.319 brouard 1562: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1563: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1564: * k= 1 2 3 4 5 6 7 8 9 10 11
1565: */
1566: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1567: /* 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
1568: # States 1=Coresidence, 2 Living alone, 3 Institution
1569: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1570: */
1.343 brouard 1571: /* V5+V4+ V3+V4*V3 +V5*age+V2 +V1*V2+V1*age+V1 */
1572: /* kmodel 1 2 3 4 5 6 7 8 9 */
1.319 brouard 1573: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1574: /* fixed or varying), 1 for age product, 2 for*/
1575: /* product */
1576: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1577: /*(single or product without age), 2 dummy*/
1578: /* with age product, 3 quant with age product*/
1579: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1580: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1.330 brouard 1581: /*TnsdVar[Tvar] 1 2 3 */
1.337 brouard 1582: /*Tvaraff[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1.319 brouard 1583: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1.338 brouard 1584: /*TvarsDind[nsd] 2 3 9 */ /* position K of single dummy cova */
1.319 brouard 1585: /* nsq 1 2 */ /* Counting single quantit tv */
1586: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1587: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1588: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1589: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1590: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1591: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 brouard 1592: /* 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 1593: /* 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 1594: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1595: /* Type */
1596: /* V 1 2 3 4 5 */
1597: /* F F V V V */
1598: /* D Q D D Q */
1599: /* */
1600: int *TvarsD;
1.330 brouard 1601: int *TnsdVar;
1.234 brouard 1602: int *TvarsDind;
1603: int *TvarsQ;
1604: int *TvarsQind;
1605:
1.318 brouard 1606: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1607: int nresult=0;
1.258 brouard 1608: int parameterline=0; /* # of the parameter (type) line */
1.334 brouard 1609: int TKresult[MAXRESULTLINESPONE]; /* TKresult[nres]=k for each resultline nres give the corresponding combination of dummies */
1610: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model corresponds to the k3 position in the resultline */
1611: int modelresult[MAXRESULTLINESPONE][NCOVMAX];/* modelresult[k3]=k1: k1th position in the model corresponds to the k3 position in the resultline */
1612: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1613: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1614: double TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.334 brouard 1615: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1616: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.318 brouard 1617: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1.332 brouard 1618: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.318 brouard 1619:
1620: /* 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
1621: # States 1=Coresidence, 2 Living alone, 3 Institution
1622: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1623: */
1.234 brouard 1624: /* 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 1625: 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 */
1626: 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 */
1627: 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 */
1628: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1629: 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 */
1630: 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 1631: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1632: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1633: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1634: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1635: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1636: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1637: 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 */
1638: 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 1639: int *TvarVV; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1640: int *TvarVVind; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1641: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
1642: /* model V1+V3+age*V1+age*V3+V1*V3 */
1643: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
1644: /* TvarVV={3,1,3}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1645: /* TvarVVind={2,5,5}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1.230 brouard 1646: int *Tvarsel; /**< Selected covariates for output */
1647: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1648: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1649: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1650: 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 1651: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1652: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1653: int *Tage;
1.227 brouard 1654: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1655: 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 1656: 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*/
1657: 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 1658: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1659: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1660: int **Tvard;
1.330 brouard 1661: int **Tvardk;
1.227 brouard 1662: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1663: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1664: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1665: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1666: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1667: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1668: double *lsurv, *lpop, *tpop;
1669:
1.231 brouard 1670: #define FD 1; /* Fixed dummy covariate */
1671: #define FQ 2; /* Fixed quantitative covariate */
1672: #define FP 3; /* Fixed product covariate */
1673: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1674: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1675: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1676: #define VD 10; /* Varying dummy covariate */
1677: #define VQ 11; /* Varying quantitative covariate */
1678: #define VP 12; /* Varying product covariate */
1679: #define VPDD 13; /* Varying product dummy*dummy covariate */
1680: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1681: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1682: #define APFD 16; /* Age product * fixed dummy covariate */
1683: #define APFQ 17; /* Age product * fixed quantitative covariate */
1684: #define APVD 18; /* Age product * varying dummy covariate */
1685: #define APVQ 19; /* Age product * varying quantitative covariate */
1686:
1687: #define FTYPE 1; /* Fixed covariate */
1688: #define VTYPE 2; /* Varying covariate (loop in wave) */
1689: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1690:
1691: struct kmodel{
1692: int maintype; /* main type */
1693: int subtype; /* subtype */
1694: };
1695: struct kmodel modell[NCOVMAX];
1696:
1.143 brouard 1697: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1698: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1699:
1700: /**************** split *************************/
1701: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1702: {
1703: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1704: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1705: */
1706: char *ss; /* pointer */
1.186 brouard 1707: int l1=0, l2=0; /* length counters */
1.126 brouard 1708:
1709: l1 = strlen(path ); /* length of path */
1710: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1711: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1712: if ( ss == NULL ) { /* no directory, so determine current directory */
1713: strcpy( name, path ); /* we got the fullname name because no directory */
1714: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1715: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1716: /* get current working directory */
1717: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1718: #ifdef WIN32
1719: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1720: #else
1721: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1722: #endif
1.126 brouard 1723: return( GLOCK_ERROR_GETCWD );
1724: }
1725: /* got dirc from getcwd*/
1726: printf(" DIRC = %s \n",dirc);
1.205 brouard 1727: } else { /* strip directory from path */
1.126 brouard 1728: ss++; /* after this, the filename */
1729: l2 = strlen( ss ); /* length of filename */
1730: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1731: strcpy( name, ss ); /* save file name */
1732: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1733: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1734: printf(" DIRC2 = %s \n",dirc);
1735: }
1736: /* We add a separator at the end of dirc if not exists */
1737: l1 = strlen( dirc ); /* length of directory */
1738: if( dirc[l1-1] != DIRSEPARATOR ){
1739: dirc[l1] = DIRSEPARATOR;
1740: dirc[l1+1] = 0;
1741: printf(" DIRC3 = %s \n",dirc);
1742: }
1743: ss = strrchr( name, '.' ); /* find last / */
1744: if (ss >0){
1745: ss++;
1746: strcpy(ext,ss); /* save extension */
1747: l1= strlen( name);
1748: l2= strlen(ss)+1;
1749: strncpy( finame, name, l1-l2);
1750: finame[l1-l2]= 0;
1751: }
1752:
1753: return( 0 ); /* we're done */
1754: }
1755:
1756:
1757: /******************************************/
1758:
1759: void replace_back_to_slash(char *s, char*t)
1760: {
1761: int i;
1762: int lg=0;
1763: i=0;
1764: lg=strlen(t);
1765: for(i=0; i<= lg; i++) {
1766: (s[i] = t[i]);
1767: if (t[i]== '\\') s[i]='/';
1768: }
1769: }
1770:
1.132 brouard 1771: char *trimbb(char *out, char *in)
1.137 brouard 1772: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1773: char *s;
1774: s=out;
1775: while (*in != '\0'){
1.137 brouard 1776: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1777: in++;
1778: }
1779: *out++ = *in++;
1780: }
1781: *out='\0';
1782: return s;
1783: }
1784:
1.187 brouard 1785: /* char *substrchaine(char *out, char *in, char *chain) */
1786: /* { */
1787: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1788: /* char *s, *t; */
1789: /* t=in;s=out; */
1790: /* while ((*in != *chain) && (*in != '\0')){ */
1791: /* *out++ = *in++; */
1792: /* } */
1793:
1794: /* /\* *in matches *chain *\/ */
1795: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1796: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1797: /* } */
1798: /* in--; chain--; */
1799: /* while ( (*in != '\0')){ */
1800: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1801: /* *out++ = *in++; */
1802: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1803: /* } */
1804: /* *out='\0'; */
1805: /* out=s; */
1806: /* return out; */
1807: /* } */
1808: char *substrchaine(char *out, char *in, char *chain)
1809: {
1810: /* Substract chain 'chain' from 'in', return and output 'out' */
1811: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1812:
1813: char *strloc;
1814:
1815: strcpy (out, in);
1816: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1817: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1818: if(strloc != NULL){
1819: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1820: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1821: /* strcpy (strloc, strloc +strlen(chain));*/
1822: }
1823: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1824: return out;
1825: }
1826:
1827:
1.145 brouard 1828: char *cutl(char *blocc, char *alocc, char *in, char occ)
1829: {
1.187 brouard 1830: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1831: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1832: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1833: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1834: */
1.160 brouard 1835: char *s, *t;
1.145 brouard 1836: t=in;s=in;
1837: while ((*in != occ) && (*in != '\0')){
1838: *alocc++ = *in++;
1839: }
1840: if( *in == occ){
1841: *(alocc)='\0';
1842: s=++in;
1843: }
1844:
1845: if (s == t) {/* occ not found */
1846: *(alocc-(in-s))='\0';
1847: in=s;
1848: }
1849: while ( *in != '\0'){
1850: *blocc++ = *in++;
1851: }
1852:
1853: *blocc='\0';
1854: return t;
1855: }
1.137 brouard 1856: char *cutv(char *blocc, char *alocc, char *in, char occ)
1857: {
1.187 brouard 1858: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1859: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1860: gives blocc="abcdef2ghi" and alocc="j".
1861: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1862: */
1863: char *s, *t;
1864: t=in;s=in;
1865: while (*in != '\0'){
1866: while( *in == occ){
1867: *blocc++ = *in++;
1868: s=in;
1869: }
1870: *blocc++ = *in++;
1871: }
1872: if (s == t) /* occ not found */
1873: *(blocc-(in-s))='\0';
1874: else
1875: *(blocc-(in-s)-1)='\0';
1876: in=s;
1877: while ( *in != '\0'){
1878: *alocc++ = *in++;
1879: }
1880:
1881: *alocc='\0';
1882: return s;
1883: }
1884:
1.126 brouard 1885: int nbocc(char *s, char occ)
1886: {
1887: int i,j=0;
1888: int lg=20;
1889: i=0;
1890: lg=strlen(s);
1891: for(i=0; i<= lg; i++) {
1.234 brouard 1892: if (s[i] == occ ) j++;
1.126 brouard 1893: }
1894: return j;
1895: }
1896:
1.137 brouard 1897: /* void cutv(char *u,char *v, char*t, char occ) */
1898: /* { */
1899: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1900: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1901: /* gives u="abcdef2ghi" and v="j" *\/ */
1902: /* int i,lg,j,p=0; */
1903: /* i=0; */
1904: /* lg=strlen(t); */
1905: /* for(j=0; j<=lg-1; j++) { */
1906: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1907: /* } */
1.126 brouard 1908:
1.137 brouard 1909: /* for(j=0; j<p; j++) { */
1910: /* (u[j] = t[j]); */
1911: /* } */
1912: /* u[p]='\0'; */
1.126 brouard 1913:
1.137 brouard 1914: /* for(j=0; j<= lg; j++) { */
1915: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1916: /* } */
1917: /* } */
1.126 brouard 1918:
1.160 brouard 1919: #ifdef _WIN32
1920: char * strsep(char **pp, const char *delim)
1921: {
1922: char *p, *q;
1923:
1924: if ((p = *pp) == NULL)
1925: return 0;
1926: if ((q = strpbrk (p, delim)) != NULL)
1927: {
1928: *pp = q + 1;
1929: *q = '\0';
1930: }
1931: else
1932: *pp = 0;
1933: return p;
1934: }
1935: #endif
1936:
1.126 brouard 1937: /********************** nrerror ********************/
1938:
1939: void nrerror(char error_text[])
1940: {
1941: fprintf(stderr,"ERREUR ...\n");
1942: fprintf(stderr,"%s\n",error_text);
1943: exit(EXIT_FAILURE);
1944: }
1945: /*********************** vector *******************/
1946: double *vector(int nl, int nh)
1947: {
1948: double *v;
1949: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1950: if (!v) nrerror("allocation failure in vector");
1951: return v-nl+NR_END;
1952: }
1953:
1954: /************************ free vector ******************/
1955: void free_vector(double*v, int nl, int nh)
1956: {
1957: free((FREE_ARG)(v+nl-NR_END));
1958: }
1959:
1960: /************************ivector *******************************/
1961: int *ivector(long nl,long nh)
1962: {
1963: int *v;
1964: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1965: if (!v) nrerror("allocation failure in ivector");
1966: return v-nl+NR_END;
1967: }
1968:
1969: /******************free ivector **************************/
1970: void free_ivector(int *v, long nl, long nh)
1971: {
1972: free((FREE_ARG)(v+nl-NR_END));
1973: }
1974:
1975: /************************lvector *******************************/
1976: long *lvector(long nl,long nh)
1977: {
1978: long *v;
1979: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1980: if (!v) nrerror("allocation failure in ivector");
1981: return v-nl+NR_END;
1982: }
1983:
1984: /******************free lvector **************************/
1985: void free_lvector(long *v, long nl, long nh)
1986: {
1987: free((FREE_ARG)(v+nl-NR_END));
1988: }
1989:
1990: /******************* imatrix *******************************/
1991: int **imatrix(long nrl, long nrh, long ncl, long nch)
1992: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1993: {
1994: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1995: int **m;
1996:
1997: /* allocate pointers to rows */
1998: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1999: if (!m) nrerror("allocation failure 1 in matrix()");
2000: m += NR_END;
2001: m -= nrl;
2002:
2003:
2004: /* allocate rows and set pointers to them */
2005: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
2006: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2007: m[nrl] += NR_END;
2008: m[nrl] -= ncl;
2009:
2010: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
2011:
2012: /* return pointer to array of pointers to rows */
2013: return m;
2014: }
2015:
2016: /****************** free_imatrix *************************/
2017: void free_imatrix(m,nrl,nrh,ncl,nch)
2018: int **m;
2019: long nch,ncl,nrh,nrl;
2020: /* free an int matrix allocated by imatrix() */
2021: {
2022: free((FREE_ARG) (m[nrl]+ncl-NR_END));
2023: free((FREE_ARG) (m+nrl-NR_END));
2024: }
2025:
2026: /******************* matrix *******************************/
2027: double **matrix(long nrl, long nrh, long ncl, long nch)
2028: {
2029: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
2030: double **m;
2031:
2032: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2033: if (!m) nrerror("allocation failure 1 in matrix()");
2034: m += NR_END;
2035: m -= nrl;
2036:
2037: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2038: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2039: m[nrl] += NR_END;
2040: m[nrl] -= ncl;
2041:
2042: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2043: return m;
1.145 brouard 2044: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
2045: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
2046: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 2047: */
2048: }
2049:
2050: /*************************free matrix ************************/
2051: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
2052: {
2053: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2054: free((FREE_ARG)(m+nrl-NR_END));
2055: }
2056:
2057: /******************* ma3x *******************************/
2058: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
2059: {
2060: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
2061: double ***m;
2062:
2063: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2064: if (!m) nrerror("allocation failure 1 in matrix()");
2065: m += NR_END;
2066: m -= nrl;
2067:
2068: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2069: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2070: m[nrl] += NR_END;
2071: m[nrl] -= ncl;
2072:
2073: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2074:
2075: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
2076: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
2077: m[nrl][ncl] += NR_END;
2078: m[nrl][ncl] -= nll;
2079: for (j=ncl+1; j<=nch; j++)
2080: m[nrl][j]=m[nrl][j-1]+nlay;
2081:
2082: for (i=nrl+1; i<=nrh; i++) {
2083: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
2084: for (j=ncl+1; j<=nch; j++)
2085: m[i][j]=m[i][j-1]+nlay;
2086: }
2087: return m;
2088: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
2089: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
2090: */
2091: }
2092:
2093: /*************************free ma3x ************************/
2094: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
2095: {
2096: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
2097: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2098: free((FREE_ARG)(m+nrl-NR_END));
2099: }
2100:
2101: /*************** function subdirf ***********/
2102: char *subdirf(char fileres[])
2103: {
2104: /* Caution optionfilefiname is hidden */
2105: strcpy(tmpout,optionfilefiname);
2106: strcat(tmpout,"/"); /* Add to the right */
2107: strcat(tmpout,fileres);
2108: return tmpout;
2109: }
2110:
2111: /*************** function subdirf2 ***********/
2112: char *subdirf2(char fileres[], char *preop)
2113: {
1.314 brouard 2114: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
2115: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 2116: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 2117: /* Caution optionfilefiname is hidden */
2118: strcpy(tmpout,optionfilefiname);
2119: strcat(tmpout,"/");
2120: strcat(tmpout,preop);
2121: strcat(tmpout,fileres);
2122: return tmpout;
2123: }
2124:
2125: /*************** function subdirf3 ***********/
2126: char *subdirf3(char fileres[], char *preop, char *preop2)
2127: {
2128:
2129: /* Caution optionfilefiname is hidden */
2130: strcpy(tmpout,optionfilefiname);
2131: strcat(tmpout,"/");
2132: strcat(tmpout,preop);
2133: strcat(tmpout,preop2);
2134: strcat(tmpout,fileres);
2135: return tmpout;
2136: }
1.213 brouard 2137:
2138: /*************** function subdirfext ***********/
2139: char *subdirfext(char fileres[], char *preop, char *postop)
2140: {
2141:
2142: strcpy(tmpout,preop);
2143: strcat(tmpout,fileres);
2144: strcat(tmpout,postop);
2145: return tmpout;
2146: }
1.126 brouard 2147:
1.213 brouard 2148: /*************** function subdirfext3 ***********/
2149: char *subdirfext3(char fileres[], char *preop, char *postop)
2150: {
2151:
2152: /* Caution optionfilefiname is hidden */
2153: strcpy(tmpout,optionfilefiname);
2154: strcat(tmpout,"/");
2155: strcat(tmpout,preop);
2156: strcat(tmpout,fileres);
2157: strcat(tmpout,postop);
2158: return tmpout;
2159: }
2160:
1.162 brouard 2161: char *asc_diff_time(long time_sec, char ascdiff[])
2162: {
2163: long sec_left, days, hours, minutes;
2164: days = (time_sec) / (60*60*24);
2165: sec_left = (time_sec) % (60*60*24);
2166: hours = (sec_left) / (60*60) ;
2167: sec_left = (sec_left) %(60*60);
2168: minutes = (sec_left) /60;
2169: sec_left = (sec_left) % (60);
2170: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2171: return ascdiff;
2172: }
2173:
1.126 brouard 2174: /***************** f1dim *************************/
2175: extern int ncom;
2176: extern double *pcom,*xicom;
2177: extern double (*nrfunc)(double []);
2178:
2179: double f1dim(double x)
2180: {
2181: int j;
2182: double f;
2183: double *xt;
2184:
2185: xt=vector(1,ncom);
2186: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2187: f=(*nrfunc)(xt);
2188: free_vector(xt,1,ncom);
2189: return f;
2190: }
2191:
2192: /*****************brent *************************/
2193: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2194: {
2195: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2196: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2197: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2198: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2199: * returned function value.
2200: */
1.126 brouard 2201: int iter;
2202: double a,b,d,etemp;
1.159 brouard 2203: double fu=0,fv,fw,fx;
1.164 brouard 2204: double ftemp=0.;
1.126 brouard 2205: double p,q,r,tol1,tol2,u,v,w,x,xm;
2206: double e=0.0;
2207:
2208: a=(ax < cx ? ax : cx);
2209: b=(ax > cx ? ax : cx);
2210: x=w=v=bx;
2211: fw=fv=fx=(*f)(x);
2212: for (iter=1;iter<=ITMAX;iter++) {
2213: xm=0.5*(a+b);
2214: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2215: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2216: printf(".");fflush(stdout);
2217: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2218: #ifdef DEBUGBRENT
1.126 brouard 2219: 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);
2220: 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);
2221: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2222: #endif
2223: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2224: *xmin=x;
2225: return fx;
2226: }
2227: ftemp=fu;
2228: if (fabs(e) > tol1) {
2229: r=(x-w)*(fx-fv);
2230: q=(x-v)*(fx-fw);
2231: p=(x-v)*q-(x-w)*r;
2232: q=2.0*(q-r);
2233: if (q > 0.0) p = -p;
2234: q=fabs(q);
2235: etemp=e;
2236: e=d;
2237: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2238: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2239: else {
1.224 brouard 2240: d=p/q;
2241: u=x+d;
2242: if (u-a < tol2 || b-u < tol2)
2243: d=SIGN(tol1,xm-x);
1.126 brouard 2244: }
2245: } else {
2246: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2247: }
2248: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2249: fu=(*f)(u);
2250: if (fu <= fx) {
2251: if (u >= x) a=x; else b=x;
2252: SHFT(v,w,x,u)
1.183 brouard 2253: SHFT(fv,fw,fx,fu)
2254: } else {
2255: if (u < x) a=u; else b=u;
2256: if (fu <= fw || w == x) {
1.224 brouard 2257: v=w;
2258: w=u;
2259: fv=fw;
2260: fw=fu;
1.183 brouard 2261: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2262: v=u;
2263: fv=fu;
1.183 brouard 2264: }
2265: }
1.126 brouard 2266: }
2267: nrerror("Too many iterations in brent");
2268: *xmin=x;
2269: return fx;
2270: }
2271:
2272: /****************** mnbrak ***********************/
2273:
2274: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2275: double (*func)(double))
1.183 brouard 2276: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2277: the downhill direction (defined by the function as evaluated at the initial points) and returns
2278: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2279: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2280: */
1.126 brouard 2281: double ulim,u,r,q, dum;
2282: double fu;
1.187 brouard 2283:
2284: double scale=10.;
2285: int iterscale=0;
2286:
2287: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2288: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2289:
2290:
2291: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2292: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2293: /* *bx = *ax - (*ax - *bx)/scale; */
2294: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2295: /* } */
2296:
1.126 brouard 2297: if (*fb > *fa) {
2298: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2299: SHFT(dum,*fb,*fa,dum)
2300: }
1.126 brouard 2301: *cx=(*bx)+GOLD*(*bx-*ax);
2302: *fc=(*func)(*cx);
1.183 brouard 2303: #ifdef DEBUG
1.224 brouard 2304: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2305: 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 2306: #endif
1.224 brouard 2307: 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 2308: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2309: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2310: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2311: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2312: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2313: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2314: fu=(*func)(u);
1.163 brouard 2315: #ifdef DEBUG
2316: /* f(x)=A(x-u)**2+f(u) */
2317: double A, fparabu;
2318: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2319: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2320: 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);
2321: 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 2322: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2323: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2324: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2325: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2326: #endif
1.184 brouard 2327: #ifdef MNBRAKORIGINAL
1.183 brouard 2328: #else
1.191 brouard 2329: /* if (fu > *fc) { */
2330: /* #ifdef DEBUG */
2331: /* printf("mnbrak4 fu > fc \n"); */
2332: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2333: /* #endif */
2334: /* /\* 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 *\\/ *\/ */
2335: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2336: /* dum=u; /\* Shifting c and u *\/ */
2337: /* u = *cx; */
2338: /* *cx = dum; */
2339: /* dum = fu; */
2340: /* fu = *fc; */
2341: /* *fc =dum; */
2342: /* } else { /\* end *\/ */
2343: /* #ifdef DEBUG */
2344: /* printf("mnbrak3 fu < fc \n"); */
2345: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2346: /* #endif */
2347: /* dum=u; /\* Shifting c and u *\/ */
2348: /* u = *cx; */
2349: /* *cx = dum; */
2350: /* dum = fu; */
2351: /* fu = *fc; */
2352: /* *fc =dum; */
2353: /* } */
1.224 brouard 2354: #ifdef DEBUGMNBRAK
2355: double A, fparabu;
2356: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2357: fparabu= *fa - A*(*ax-u)*(*ax-u);
2358: 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);
2359: 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 2360: #endif
1.191 brouard 2361: dum=u; /* Shifting c and u */
2362: u = *cx;
2363: *cx = dum;
2364: dum = fu;
2365: fu = *fc;
2366: *fc =dum;
1.183 brouard 2367: #endif
1.162 brouard 2368: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2369: #ifdef DEBUG
1.224 brouard 2370: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2371: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2372: #endif
1.126 brouard 2373: fu=(*func)(u);
2374: if (fu < *fc) {
1.183 brouard 2375: #ifdef DEBUG
1.224 brouard 2376: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2377: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2378: #endif
2379: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2380: SHFT(*fb,*fc,fu,(*func)(u))
2381: #ifdef DEBUG
2382: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2383: #endif
2384: }
1.162 brouard 2385: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2386: #ifdef DEBUG
1.224 brouard 2387: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2388: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2389: #endif
1.126 brouard 2390: u=ulim;
2391: fu=(*func)(u);
1.183 brouard 2392: } else { /* u could be left to b (if r > q parabola has a maximum) */
2393: #ifdef DEBUG
1.224 brouard 2394: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2395: 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 2396: #endif
1.126 brouard 2397: u=(*cx)+GOLD*(*cx-*bx);
2398: fu=(*func)(u);
1.224 brouard 2399: #ifdef DEBUG
2400: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2401: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2402: #endif
1.183 brouard 2403: } /* end tests */
1.126 brouard 2404: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2405: SHFT(*fa,*fb,*fc,fu)
2406: #ifdef DEBUG
1.224 brouard 2407: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2408: 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 2409: #endif
2410: } /* 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 2411: }
2412:
2413: /*************** linmin ************************/
1.162 brouard 2414: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2415: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2416: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2417: the value of func at the returned location p . This is actually all accomplished by calling the
2418: routines mnbrak and brent .*/
1.126 brouard 2419: int ncom;
2420: double *pcom,*xicom;
2421: double (*nrfunc)(double []);
2422:
1.224 brouard 2423: #ifdef LINMINORIGINAL
1.126 brouard 2424: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2425: #else
2426: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2427: #endif
1.126 brouard 2428: {
2429: double brent(double ax, double bx, double cx,
2430: double (*f)(double), double tol, double *xmin);
2431: double f1dim(double x);
2432: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2433: double *fc, double (*func)(double));
2434: int j;
2435: double xx,xmin,bx,ax;
2436: double fx,fb,fa;
1.187 brouard 2437:
1.203 brouard 2438: #ifdef LINMINORIGINAL
2439: #else
2440: double scale=10., axs, xxs; /* Scale added for infinity */
2441: #endif
2442:
1.126 brouard 2443: ncom=n;
2444: pcom=vector(1,n);
2445: xicom=vector(1,n);
2446: nrfunc=func;
2447: for (j=1;j<=n;j++) {
2448: pcom[j]=p[j];
1.202 brouard 2449: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2450: }
1.187 brouard 2451:
1.203 brouard 2452: #ifdef LINMINORIGINAL
2453: xx=1.;
2454: #else
2455: axs=0.0;
2456: xxs=1.;
2457: do{
2458: xx= xxs;
2459: #endif
1.187 brouard 2460: ax=0.;
2461: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2462: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2463: /* 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)) */
2464: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2465: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2466: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2467: /* 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 2468: #ifdef LINMINORIGINAL
2469: #else
2470: if (fx != fx){
1.224 brouard 2471: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2472: printf("|");
2473: fprintf(ficlog,"|");
1.203 brouard 2474: #ifdef DEBUGLINMIN
1.224 brouard 2475: 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 2476: #endif
2477: }
1.224 brouard 2478: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2479: #endif
2480:
1.191 brouard 2481: #ifdef DEBUGLINMIN
2482: 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 2483: 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 2484: #endif
1.224 brouard 2485: #ifdef LINMINORIGINAL
2486: #else
1.317 brouard 2487: if(fb == fx){ /* Flat function in the direction */
2488: xmin=xx;
1.224 brouard 2489: *flat=1;
1.317 brouard 2490: }else{
1.224 brouard 2491: *flat=0;
2492: #endif
2493: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2494: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2495: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2496: /* fmin = f(p[j] + xmin * xi[j]) */
2497: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2498: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2499: #ifdef DEBUG
1.224 brouard 2500: 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);
2501: 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);
2502: #endif
2503: #ifdef LINMINORIGINAL
2504: #else
2505: }
1.126 brouard 2506: #endif
1.191 brouard 2507: #ifdef DEBUGLINMIN
2508: printf("linmin end ");
1.202 brouard 2509: fprintf(ficlog,"linmin end ");
1.191 brouard 2510: #endif
1.126 brouard 2511: for (j=1;j<=n;j++) {
1.203 brouard 2512: #ifdef LINMINORIGINAL
2513: xi[j] *= xmin;
2514: #else
2515: #ifdef DEBUGLINMIN
2516: if(xxs <1.0)
2517: printf(" before xi[%d]=%12.8f", j,xi[j]);
2518: #endif
2519: 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) */
2520: #ifdef DEBUGLINMIN
2521: if(xxs <1.0)
2522: 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 );
2523: #endif
2524: #endif
1.187 brouard 2525: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2526: }
1.191 brouard 2527: #ifdef DEBUGLINMIN
1.203 brouard 2528: printf("\n");
1.191 brouard 2529: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2530: 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 2531: for (j=1;j<=n;j++) {
1.202 brouard 2532: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2533: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2534: if(j % ncovmodel == 0){
1.191 brouard 2535: printf("\n");
1.202 brouard 2536: fprintf(ficlog,"\n");
2537: }
1.191 brouard 2538: }
1.203 brouard 2539: #else
1.191 brouard 2540: #endif
1.126 brouard 2541: free_vector(xicom,1,n);
2542: free_vector(pcom,1,n);
2543: }
2544:
2545:
2546: /*************** powell ************************/
1.162 brouard 2547: /*
1.317 brouard 2548: Minimization of a function func of n variables. Input consists in an initial starting point
2549: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2550: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2551: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2552: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2553: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2554: */
1.224 brouard 2555: #ifdef LINMINORIGINAL
2556: #else
2557: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2558: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2559: #endif
1.126 brouard 2560: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2561: double (*func)(double []))
2562: {
1.224 brouard 2563: #ifdef LINMINORIGINAL
2564: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2565: double (*func)(double []));
1.224 brouard 2566: #else
1.241 brouard 2567: void linmin(double p[], double xi[], int n, double *fret,
2568: double (*func)(double []),int *flat);
1.224 brouard 2569: #endif
1.239 brouard 2570: int i,ibig,j,jk,k;
1.126 brouard 2571: double del,t,*pt,*ptt,*xit;
1.181 brouard 2572: double directest;
1.126 brouard 2573: double fp,fptt;
2574: double *xits;
2575: int niterf, itmp;
2576:
2577: pt=vector(1,n);
2578: ptt=vector(1,n);
2579: xit=vector(1,n);
2580: xits=vector(1,n);
2581: *fret=(*func)(p);
2582: for (j=1;j<=n;j++) pt[j]=p[j];
1.338 brouard 2583: rcurr_time = time(NULL);
2584: fp=(*fret); /* Initialisation */
1.126 brouard 2585: for (*iter=1;;++(*iter)) {
2586: ibig=0;
2587: del=0.0;
1.157 brouard 2588: rlast_time=rcurr_time;
2589: /* (void) gettimeofday(&curr_time,&tzp); */
2590: rcurr_time = time(NULL);
2591: curr_time = *localtime(&rcurr_time);
1.337 brouard 2592: /* 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); */
2593: /* 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); */
2594: printf("\nPowell iter=%d -2*LL=%.12f gain=%.3lg %ld sec. %ld sec.",*iter,*fret,fp-*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2595: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f gain=%.3lg %ld sec. %ld sec.",*iter,*fret,fp-*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
1.157 brouard 2596: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2597: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2598: for (i=1;i<=n;i++) {
1.126 brouard 2599: fprintf(ficrespow," %.12lf", p[i]);
2600: }
1.239 brouard 2601: fprintf(ficrespow,"\n");fflush(ficrespow);
2602: printf("\n#model= 1 + age ");
2603: fprintf(ficlog,"\n#model= 1 + age ");
2604: if(nagesqr==1){
1.241 brouard 2605: printf(" + age*age ");
2606: fprintf(ficlog," + age*age ");
1.239 brouard 2607: }
2608: for(j=1;j <=ncovmodel-2;j++){
2609: if(Typevar[j]==0) {
2610: printf(" + V%d ",Tvar[j]);
2611: fprintf(ficlog," + V%d ",Tvar[j]);
2612: }else if(Typevar[j]==1) {
2613: printf(" + V%d*age ",Tvar[j]);
2614: fprintf(ficlog," + V%d*age ",Tvar[j]);
2615: }else if(Typevar[j]==2) {
2616: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2617: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2618: }
2619: }
1.126 brouard 2620: printf("\n");
1.239 brouard 2621: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2622: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2623: fprintf(ficlog,"\n");
1.239 brouard 2624: for(i=1,jk=1; i <=nlstate; i++){
2625: for(k=1; k <=(nlstate+ndeath); k++){
2626: if (k != i) {
2627: printf("%d%d ",i,k);
2628: fprintf(ficlog,"%d%d ",i,k);
2629: for(j=1; j <=ncovmodel; j++){
2630: printf("%12.7f ",p[jk]);
2631: fprintf(ficlog,"%12.7f ",p[jk]);
2632: jk++;
2633: }
2634: printf("\n");
2635: fprintf(ficlog,"\n");
2636: }
2637: }
2638: }
1.241 brouard 2639: if(*iter <=3 && *iter >1){
1.157 brouard 2640: tml = *localtime(&rcurr_time);
2641: strcpy(strcurr,asctime(&tml));
2642: rforecast_time=rcurr_time;
1.126 brouard 2643: itmp = strlen(strcurr);
2644: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2645: strcurr[itmp-1]='\0';
1.162 brouard 2646: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2647: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2648: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2649: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2650: forecast_time = *localtime(&rforecast_time);
2651: strcpy(strfor,asctime(&forecast_time));
2652: itmp = strlen(strfor);
2653: if(strfor[itmp-1]=='\n')
2654: strfor[itmp-1]='\0';
2655: printf(" - if your program needs %d iterations to converge, convergence will be \n reached in %s i.e.\n on %s (current time is %s);\n",niterf, asc_diff_time(rforecast_time-rcurr_time,tmpout),strfor,strcurr);
2656: fprintf(ficlog," - if your program needs %d iterations to converge, convergence will be \n reached in %s i.e.\n on %s (current time is %s);\n",niterf, asc_diff_time(rforecast_time-rcurr_time,tmpout),strfor,strcurr);
1.126 brouard 2657: }
2658: }
1.187 brouard 2659: for (i=1;i<=n;i++) { /* For each direction i */
2660: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2661: fptt=(*fret);
2662: #ifdef DEBUG
1.203 brouard 2663: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2664: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2665: #endif
1.203 brouard 2666: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2667: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2668: #ifdef LINMINORIGINAL
1.188 brouard 2669: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2670: #else
2671: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2672: flatdir[i]=flat; /* Function is vanishing in that direction i */
2673: #endif
2674: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2675: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2676: /* because that direction will be replaced unless the gain del is small */
2677: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2678: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2679: /* with the new direction. */
2680: del=fabs(fptt-(*fret));
2681: ibig=i;
1.126 brouard 2682: }
2683: #ifdef DEBUG
2684: printf("%d %.12e",i,(*fret));
2685: fprintf(ficlog,"%d %.12e",i,(*fret));
2686: for (j=1;j<=n;j++) {
1.224 brouard 2687: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2688: printf(" x(%d)=%.12e",j,xit[j]);
2689: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2690: }
2691: for(j=1;j<=n;j++) {
1.225 brouard 2692: printf(" p(%d)=%.12e",j,p[j]);
2693: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2694: }
2695: printf("\n");
2696: fprintf(ficlog,"\n");
2697: #endif
1.187 brouard 2698: } /* end loop on each direction i */
2699: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2700: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2701: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2702: for(j=1;j<=n;j++) {
2703: if(flatdir[j] >0){
2704: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2705: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2706: }
1.319 brouard 2707: /* printf("\n"); */
2708: /* fprintf(ficlog,"\n"); */
2709: }
1.243 brouard 2710: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2711: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2712: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2713: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2714: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2715: /* decreased of more than 3.84 */
2716: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2717: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2718: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2719:
1.188 brouard 2720: /* Starting the program with initial values given by a former maximization will simply change */
2721: /* the scales of the directions and the directions, because the are reset to canonical directions */
2722: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2723: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2724: #ifdef DEBUG
2725: int k[2],l;
2726: k[0]=1;
2727: k[1]=-1;
2728: printf("Max: %.12e",(*func)(p));
2729: fprintf(ficlog,"Max: %.12e",(*func)(p));
2730: for (j=1;j<=n;j++) {
2731: printf(" %.12e",p[j]);
2732: fprintf(ficlog," %.12e",p[j]);
2733: }
2734: printf("\n");
2735: fprintf(ficlog,"\n");
2736: for(l=0;l<=1;l++) {
2737: for (j=1;j<=n;j++) {
2738: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2739: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2740: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2741: }
2742: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2743: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2744: }
2745: #endif
2746:
2747: free_vector(xit,1,n);
2748: free_vector(xits,1,n);
2749: free_vector(ptt,1,n);
2750: free_vector(pt,1,n);
2751: return;
1.192 brouard 2752: } /* enough precision */
1.240 brouard 2753: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2754: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2755: ptt[j]=2.0*p[j]-pt[j];
2756: xit[j]=p[j]-pt[j];
2757: pt[j]=p[j];
2758: }
1.181 brouard 2759: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2760: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2761: if (*iter <=4) {
1.225 brouard 2762: #else
2763: #endif
1.224 brouard 2764: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2765: #else
1.161 brouard 2766: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2767: #endif
1.162 brouard 2768: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2769: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2770: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2771: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2772: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2773: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2774: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2775: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2776: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2777: /* Even if f3 <f1, directest can be negative and t >0 */
2778: /* mu² and del² are equal when f3=f1 */
2779: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2780: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2781: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2782: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2783: #ifdef NRCORIGINAL
2784: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2785: #else
2786: 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 2787: t= t- del*SQR(fp-fptt);
1.183 brouard 2788: #endif
1.202 brouard 2789: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2790: #ifdef DEBUG
1.181 brouard 2791: 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);
2792: 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 2793: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2794: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2795: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2796: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2797: 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);
2798: 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);
2799: #endif
1.183 brouard 2800: #ifdef POWELLORIGINAL
2801: if (t < 0.0) { /* Then we use it for new direction */
2802: #else
1.182 brouard 2803: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2804: 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 2805: 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 2806: 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 2807: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2808: }
1.181 brouard 2809: if (directest < 0.0) { /* Then we use it for new direction */
2810: #endif
1.191 brouard 2811: #ifdef DEBUGLINMIN
1.234 brouard 2812: printf("Before linmin in direction P%d-P0\n",n);
2813: for (j=1;j<=n;j++) {
2814: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2815: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2816: if(j % ncovmodel == 0){
2817: printf("\n");
2818: fprintf(ficlog,"\n");
2819: }
2820: }
1.224 brouard 2821: #endif
2822: #ifdef LINMINORIGINAL
1.234 brouard 2823: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2824: #else
1.234 brouard 2825: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2826: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2827: #endif
1.234 brouard 2828:
1.191 brouard 2829: #ifdef DEBUGLINMIN
1.234 brouard 2830: for (j=1;j<=n;j++) {
2831: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2832: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2833: if(j % ncovmodel == 0){
2834: printf("\n");
2835: fprintf(ficlog,"\n");
2836: }
2837: }
1.224 brouard 2838: #endif
1.234 brouard 2839: for (j=1;j<=n;j++) {
2840: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2841: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2842: }
1.224 brouard 2843: #ifdef LINMINORIGINAL
2844: #else
1.234 brouard 2845: for (j=1, flatd=0;j<=n;j++) {
2846: if(flatdir[j]>0)
2847: flatd++;
2848: }
2849: if(flatd >0){
1.255 brouard 2850: printf("%d flat directions: ",flatd);
2851: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2852: for (j=1;j<=n;j++) {
2853: if(flatdir[j]>0){
2854: printf("%d ",j);
2855: fprintf(ficlog,"%d ",j);
2856: }
2857: }
2858: printf("\n");
2859: fprintf(ficlog,"\n");
1.319 brouard 2860: #ifdef FLATSUP
2861: free_vector(xit,1,n);
2862: free_vector(xits,1,n);
2863: free_vector(ptt,1,n);
2864: free_vector(pt,1,n);
2865: return;
2866: #endif
1.234 brouard 2867: }
1.191 brouard 2868: #endif
1.234 brouard 2869: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2870: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2871:
1.126 brouard 2872: #ifdef DEBUG
1.234 brouard 2873: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2874: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2875: for(j=1;j<=n;j++){
2876: printf(" %lf",xit[j]);
2877: fprintf(ficlog," %lf",xit[j]);
2878: }
2879: printf("\n");
2880: fprintf(ficlog,"\n");
1.126 brouard 2881: #endif
1.192 brouard 2882: } /* end of t or directest negative */
1.224 brouard 2883: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2884: #else
1.234 brouard 2885: } /* end if (fptt < fp) */
1.192 brouard 2886: #endif
1.225 brouard 2887: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2888: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2889: #else
1.224 brouard 2890: #endif
1.234 brouard 2891: } /* loop iteration */
1.126 brouard 2892: }
1.234 brouard 2893:
1.126 brouard 2894: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2895:
1.235 brouard 2896: 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 2897: {
1.338 brouard 2898: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij . Nicely done
1.279 brouard 2899: * (and selected quantitative values in nres)
2900: * by left multiplying the unit
2901: * matrix by transitions matrix until convergence is reached with precision ftolpl
2902: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2903: * Wx is row vector: population in state 1, population in state 2, population dead
2904: * or prevalence in state 1, prevalence in state 2, 0
2905: * newm is the matrix after multiplications, its rows are identical at a factor.
2906: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2907: * Output is prlim.
2908: * Initial matrix pimij
2909: */
1.206 brouard 2910: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2911: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2912: /* 0, 0 , 1} */
2913: /*
2914: * and after some iteration: */
2915: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2916: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2917: /* 0, 0 , 1} */
2918: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2919: /* {0.51571254859325999, 0.4842874514067399, */
2920: /* 0.51326036147820708, 0.48673963852179264} */
2921: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2922:
1.332 brouard 2923: int i, ii,j,k, k1;
1.209 brouard 2924: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2925: /* double **matprod2(); */ /* test */
1.218 brouard 2926: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2927: double **newm;
1.209 brouard 2928: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2929: int ncvloop=0;
1.288 brouard 2930: int first=0;
1.169 brouard 2931:
1.209 brouard 2932: min=vector(1,nlstate);
2933: max=vector(1,nlstate);
2934: meandiff=vector(1,nlstate);
2935:
1.218 brouard 2936: /* Starting with matrix unity */
1.126 brouard 2937: for (ii=1;ii<=nlstate+ndeath;ii++)
2938: for (j=1;j<=nlstate+ndeath;j++){
2939: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2940: }
1.169 brouard 2941:
2942: cov[1]=1.;
2943:
2944: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2945: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2946: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2947: ncvloop++;
1.126 brouard 2948: newm=savm;
2949: /* Covariates have to be included here again */
1.138 brouard 2950: cov[2]=agefin;
1.319 brouard 2951: if(nagesqr==1){
2952: cov[3]= agefin*agefin;
2953: }
1.332 brouard 2954: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
2955: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
2956: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
2957: if(Typevar[k1]==1){ /* A product with age */
2958: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
2959: }else{
2960: cov[2+nagesqr+k1]=precov[nres][k1];
2961: }
2962: }/* End of loop on model equation */
2963:
2964: /* Start of old code (replaced by a loop on position in the model equation */
2965: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only of the model *\/ */
2966: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
2967: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; *\/ */
2968: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])]; */
2969: /* /\* model = 1 +age + V1*V3 + age*V1 + V2 + V1 + age*V2 + V3 + V3*age + V1*V2 */
2970: /* * k 1 2 3 4 5 6 7 8 */
2971: /* *cov[] 1 2 3 4 5 6 7 8 9 10 */
2972: /* *TypeVar[k] 2 1 0 0 1 0 1 2 */
2973: /* *Dummy[k] 0 2 0 0 2 0 2 0 */
2974: /* *Tvar[k] 4 1 2 1 2 3 3 5 */
2975: /* *nsd=3 (1) (2) (3) */
2976: /* *TvarsD[nsd] [1]=2 1 3 */
2977: /* *TnsdVar [2]=2 [1]=1 [3]=3 */
2978: /* *TvarsDind[nsd](=k) [1]=3 [2]=4 [3]=6 */
2979: /* *Tage[] [1]=1 [2]=2 [3]=3 */
2980: /* *Tvard[] [1][1]=1 [2][1]=1 */
2981: /* * [1][2]=3 [2][2]=2 */
2982: /* *Tprod[](=k) [1]=1 [2]=8 */
2983: /* *TvarsDp(=Tvar) [1]=1 [2]=2 [3]=3 [4]=5 */
2984: /* *TvarD (=k) [1]=1 [2]=3 [3]=4 [3]=6 [4]=6 */
2985: /* *TvarsDpType */
2986: /* *si model= 1 + age + V3 + V2*age + V2 + V3*age */
2987: /* * nsd=1 (1) (2) */
2988: /* *TvarsD[nsd] 3 2 */
2989: /* *TnsdVar (3)=1 (2)=2 */
2990: /* *TvarsDind[nsd](=k) [1]=1 [2]=3 */
2991: /* *Tage[] [1]=2 [2]= 3 */
2992: /* *\/ */
2993: /* /\* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; *\/ */
2994: /* /\* 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)); *\/ */
2995: /* } */
2996: /* for (k=1; k<=nsq;k++) { /\* For single quantitative varying covariates only of the model *\/ */
2997: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
2998: /* /\* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline *\/ */
2999: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
3000: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][resultmodel[nres][k1]] */
3001: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
3002: /* /\* 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]); *\/ */
3003: /* } */
3004: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
3005: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
3006: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3007: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
3008: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
3009: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3010: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
3011: /* } */
3012: /* /\* 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]); *\/ */
3013: /* } */
3014: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
3015: /* /\* 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]); *\/ */
3016: /* if(Dummy[Tvard[k][1]]==0){ */
3017: /* if(Dummy[Tvard[k][2]]==0){ */
3018: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3019: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3020: /* }else{ */
3021: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3022: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
3023: /* } */
3024: /* }else{ */
3025: /* if(Dummy[Tvard[k][2]]==0){ */
3026: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3027: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
3028: /* }else{ */
3029: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3030: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
3031: /* } */
3032: /* } */
3033: /* } /\* End product without age *\/ */
3034: /* ENd of old code */
1.138 brouard 3035: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3036: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3037: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 3038: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3039: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 3040: /* age and covariate values of ij are in 'cov' */
1.142 brouard 3041: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 3042:
1.126 brouard 3043: savm=oldm;
3044: oldm=newm;
1.209 brouard 3045:
3046: for(j=1; j<=nlstate; j++){
3047: max[j]=0.;
3048: min[j]=1.;
3049: }
3050: for(i=1;i<=nlstate;i++){
3051: sumnew=0;
3052: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
3053: for(j=1; j<=nlstate; j++){
3054: prlim[i][j]= newm[i][j]/(1-sumnew);
3055: max[j]=FMAX(max[j],prlim[i][j]);
3056: min[j]=FMIN(min[j],prlim[i][j]);
3057: }
3058: }
3059:
1.126 brouard 3060: maxmax=0.;
1.209 brouard 3061: for(j=1; j<=nlstate; j++){
3062: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
3063: maxmax=FMAX(maxmax,meandiff[j]);
3064: /* 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 3065: } /* j loop */
1.203 brouard 3066: *ncvyear= (int)age- (int)agefin;
1.208 brouard 3067: /* 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 3068: if(maxmax < ftolpl){
1.209 brouard 3069: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
3070: free_vector(min,1,nlstate);
3071: free_vector(max,1,nlstate);
3072: free_vector(meandiff,1,nlstate);
1.126 brouard 3073: return prlim;
3074: }
1.288 brouard 3075: } /* agefin loop */
1.208 brouard 3076: /* After some age loop it doesn't converge */
1.288 brouard 3077: if(!first){
3078: first=1;
3079: 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 3080: 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);
3081: }else if (first >=1 && first <10){
3082: 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);
3083: first++;
3084: }else if (first ==10){
3085: 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);
3086: 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");
3087: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
3088: first++;
1.288 brouard 3089: }
3090:
1.209 brouard 3091: /* 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); */
3092: free_vector(min,1,nlstate);
3093: free_vector(max,1,nlstate);
3094: free_vector(meandiff,1,nlstate);
1.208 brouard 3095:
1.169 brouard 3096: return prlim; /* should not reach here */
1.126 brouard 3097: }
3098:
1.217 brouard 3099:
3100: /**** Back Prevalence limit (stable or period prevalence) ****************/
3101:
1.218 brouard 3102: /* 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) */
3103: /* 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 3104: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 3105: {
1.264 brouard 3106: /* 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 3107: matrix by transitions matrix until convergence is reached with precision ftolpl */
3108: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
3109: /* Wx is row vector: population in state 1, population in state 2, population dead */
3110: /* or prevalence in state 1, prevalence in state 2, 0 */
3111: /* newm is the matrix after multiplications, its rows are identical at a factor */
3112: /* Initial matrix pimij */
3113: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
3114: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
3115: /* 0, 0 , 1} */
3116: /*
3117: * and after some iteration: */
3118: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
3119: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
3120: /* 0, 0 , 1} */
3121: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
3122: /* {0.51571254859325999, 0.4842874514067399, */
3123: /* 0.51326036147820708, 0.48673963852179264} */
3124: /* If we start from prlim again, prlim tends to a constant matrix */
3125:
1.332 brouard 3126: int i, ii,j,k, k1;
1.247 brouard 3127: int first=0;
1.217 brouard 3128: double *min, *max, *meandiff, maxmax,sumnew=0.;
3129: /* double **matprod2(); */ /* test */
3130: double **out, cov[NCOVMAX+1], **bmij();
3131: double **newm;
1.218 brouard 3132: double **dnewm, **doldm, **dsavm; /* for use */
3133: double **oldm, **savm; /* for use */
3134:
1.217 brouard 3135: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
3136: int ncvloop=0;
3137:
3138: min=vector(1,nlstate);
3139: max=vector(1,nlstate);
3140: meandiff=vector(1,nlstate);
3141:
1.266 brouard 3142: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
3143: oldm=oldms; savm=savms;
3144:
3145: /* Starting with matrix unity */
3146: for (ii=1;ii<=nlstate+ndeath;ii++)
3147: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 3148: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3149: }
3150:
3151: cov[1]=1.;
3152:
3153: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3154: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 3155: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 3156: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
3157: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 3158: ncvloop++;
1.218 brouard 3159: newm=savm; /* oldm should be kept from previous iteration or unity at start */
3160: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 3161: /* Covariates have to be included here again */
3162: cov[2]=agefin;
1.319 brouard 3163: if(nagesqr==1){
1.217 brouard 3164: cov[3]= agefin*agefin;;
1.319 brouard 3165: }
1.332 brouard 3166: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3167: if(Typevar[k1]==1){ /* A product with age */
3168: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.242 brouard 3169: }else{
1.332 brouard 3170: cov[2+nagesqr+k1]=precov[nres][k1];
1.242 brouard 3171: }
1.332 brouard 3172: }/* End of loop on model equation */
3173:
3174: /* Old code */
3175:
3176: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
3177: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
3178: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; */
3179: /* /\* 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)); *\/ */
3180: /* } */
3181: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
3182: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3183: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3184: /* /\* /\\* 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])]); *\\/ *\/ */
3185: /* /\* } *\/ */
3186: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3187: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3188: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3189: /* /\* 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]); *\/ */
3190: /* } */
3191: /* /\* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; *\/ */
3192: /* /\* for (k=1; k<=cptcovprod;k++) /\\* Useless *\\/ *\/ */
3193: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\\/ *\/ */
3194: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3195: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
3196: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ ERROR ???*\/ */
3197: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3198: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3199: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3200: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3201: /* } */
3202: /* /\* 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]); *\/ */
3203: /* } */
3204: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
3205: /* /\* 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]); *\/ */
3206: /* if(Dummy[Tvard[k][1]]==0){ */
3207: /* if(Dummy[Tvard[k][2]]==0){ */
3208: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3209: /* }else{ */
3210: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3211: /* } */
3212: /* }else{ */
3213: /* if(Dummy[Tvard[k][2]]==0){ */
3214: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3215: /* }else{ */
3216: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3217: /* } */
3218: /* } */
3219: /* } */
1.217 brouard 3220:
3221: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3222: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3223: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3224: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3225: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3226: /* ij should be linked to the correct index of cov */
3227: /* age and covariate values ij are in 'cov', but we need to pass
3228: * ij for the observed prevalence at age and status and covariate
3229: * number: prevacurrent[(int)agefin][ii][ij]
3230: */
3231: /* 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 *\/ */
3232: /* 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 *\/ */
3233: 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 3234: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3235: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3236: /* for(i=1; i<=nlstate+ndeath; i++) { */
3237: /* printf("%d newm= ",i); */
3238: /* for(j=1;j<=nlstate+ndeath;j++) { */
3239: /* printf("%f ",newm[i][j]); */
3240: /* } */
3241: /* printf("oldm * "); */
3242: /* for(j=1;j<=nlstate+ndeath;j++) { */
3243: /* printf("%f ",oldm[i][j]); */
3244: /* } */
1.268 brouard 3245: /* printf(" bmmij "); */
1.266 brouard 3246: /* for(j=1;j<=nlstate+ndeath;j++) { */
3247: /* printf("%f ",pmmij[i][j]); */
3248: /* } */
3249: /* printf("\n"); */
3250: /* } */
3251: /* } */
1.217 brouard 3252: savm=oldm;
3253: oldm=newm;
1.266 brouard 3254:
1.217 brouard 3255: for(j=1; j<=nlstate; j++){
3256: max[j]=0.;
3257: min[j]=1.;
3258: }
3259: for(j=1; j<=nlstate; j++){
3260: for(i=1;i<=nlstate;i++){
1.234 brouard 3261: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3262: bprlim[i][j]= newm[i][j];
3263: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3264: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3265: }
3266: }
1.218 brouard 3267:
1.217 brouard 3268: maxmax=0.;
3269: for(i=1; i<=nlstate; i++){
1.318 brouard 3270: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3271: maxmax=FMAX(maxmax,meandiff[i]);
3272: /* 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 3273: } /* i loop */
1.217 brouard 3274: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3275: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3276: if(maxmax < ftolpl){
1.220 brouard 3277: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3278: free_vector(min,1,nlstate);
3279: free_vector(max,1,nlstate);
3280: free_vector(meandiff,1,nlstate);
3281: return bprlim;
3282: }
1.288 brouard 3283: } /* agefin loop */
1.217 brouard 3284: /* After some age loop it doesn't converge */
1.288 brouard 3285: if(!first){
1.247 brouard 3286: first=1;
3287: 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\
3288: 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);
3289: }
3290: 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 3291: 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);
3292: /* 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); */
3293: free_vector(min,1,nlstate);
3294: free_vector(max,1,nlstate);
3295: free_vector(meandiff,1,nlstate);
3296:
3297: return bprlim; /* should not reach here */
3298: }
3299:
1.126 brouard 3300: /*************** transition probabilities ***************/
3301:
3302: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3303: {
1.138 brouard 3304: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3305: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3306: model to the ncovmodel covariates (including constant and age).
3307: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3308: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3309: ncth covariate in the global vector x is given by the formula:
3310: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3311: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3312: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3313: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3314: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3315: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3316: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3317: */
3318: double s1, lnpijopii;
1.126 brouard 3319: /*double t34;*/
1.164 brouard 3320: int i,j, nc, ii, jj;
1.126 brouard 3321:
1.223 brouard 3322: for(i=1; i<= nlstate; i++){
3323: for(j=1; j<i;j++){
3324: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3325: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3326: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3327: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3328: }
3329: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3330: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3331: }
3332: for(j=i+1; j<=nlstate+ndeath;j++){
3333: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3334: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3335: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3336: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3337: }
3338: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3339: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3340: }
3341: }
1.218 brouard 3342:
1.223 brouard 3343: for(i=1; i<= nlstate; i++){
3344: s1=0;
3345: for(j=1; j<i; j++){
1.339 brouard 3346: /* printf("debug1 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 3347: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3348: }
3349: for(j=i+1; j<=nlstate+ndeath; j++){
1.339 brouard 3350: /* printf("debug2 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 3351: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3352: }
3353: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3354: ps[i][i]=1./(s1+1.);
3355: /* Computing other pijs */
3356: for(j=1; j<i; j++)
1.325 brouard 3357: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3358: for(j=i+1; j<=nlstate+ndeath; j++)
3359: ps[i][j]= exp(ps[i][j])*ps[i][i];
3360: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3361: } /* end i */
1.218 brouard 3362:
1.223 brouard 3363: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3364: for(jj=1; jj<= nlstate+ndeath; jj++){
3365: ps[ii][jj]=0;
3366: ps[ii][ii]=1;
3367: }
3368: }
1.294 brouard 3369:
3370:
1.223 brouard 3371: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3372: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3373: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3374: /* } */
3375: /* printf("\n "); */
3376: /* } */
3377: /* printf("\n ");printf("%lf ",cov[2]);*/
3378: /*
3379: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3380: goto end;*/
1.266 brouard 3381: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3382: }
3383:
1.218 brouard 3384: /*************** backward transition probabilities ***************/
3385:
3386: /* 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 ) */
3387: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3388: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3389: {
1.302 brouard 3390: /* 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 3391: * 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 3392: */
1.218 brouard 3393: int i, ii, j,k;
1.222 brouard 3394:
3395: double **out, **pmij();
3396: double sumnew=0.;
1.218 brouard 3397: double agefin;
1.292 brouard 3398: 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 3399: double **dnewm, **dsavm, **doldm;
3400: double **bbmij;
3401:
1.218 brouard 3402: doldm=ddoldms; /* global pointers */
1.222 brouard 3403: dnewm=ddnewms;
3404: dsavm=ddsavms;
1.318 brouard 3405:
3406: /* Debug */
3407: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3408: agefin=cov[2];
1.268 brouard 3409: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3410: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3411: the observed prevalence (with this covariate ij) at beginning of transition */
3412: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3413:
3414: /* P_x */
1.325 brouard 3415: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3416: /* outputs pmmij which is a stochastic matrix in row */
3417:
3418: /* Diag(w_x) */
1.292 brouard 3419: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3420: sumnew=0.;
1.269 brouard 3421: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3422: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3423: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3424: sumnew+=prevacurrent[(int)agefin][ii][ij];
3425: }
3426: if(sumnew >0.01){ /* At least some value in the prevalence */
3427: for (ii=1;ii<=nlstate+ndeath;ii++){
3428: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3429: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3430: }
3431: }else{
3432: for (ii=1;ii<=nlstate+ndeath;ii++){
3433: for (j=1;j<=nlstate+ndeath;j++)
3434: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3435: }
3436: /* if(sumnew <0.9){ */
3437: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3438: /* } */
3439: }
3440: k3=0.0; /* We put the last diagonal to 0 */
3441: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3442: doldm[ii][ii]= k3;
3443: }
3444: /* End doldm, At the end doldm is diag[(w_i)] */
3445:
1.292 brouard 3446: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3447: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3448:
1.292 brouard 3449: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3450: /* 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 3451: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3452: sumnew=0.;
1.222 brouard 3453: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3454: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3455: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3456: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3457: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3458: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3459: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3460: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3461: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3462: /* }else */
1.268 brouard 3463: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3464: } /*End ii */
3465: } /* 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 */
3466:
1.292 brouard 3467: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3468: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3469: /* end bmij */
1.266 brouard 3470: return ps; /*pointer is unchanged */
1.218 brouard 3471: }
1.217 brouard 3472: /*************** transition probabilities ***************/
3473:
1.218 brouard 3474: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3475: {
3476: /* According to parameters values stored in x and the covariate's values stored in cov,
3477: computes the probability to be observed in state j being in state i by appying the
3478: model to the ncovmodel covariates (including constant and age).
3479: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3480: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3481: ncth covariate in the global vector x is given by the formula:
3482: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3483: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3484: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3485: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3486: Outputs ps[i][j] the probability to be observed in j being in j according to
3487: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3488: */
3489: double s1, lnpijopii;
3490: /*double t34;*/
3491: int i,j, nc, ii, jj;
3492:
1.234 brouard 3493: for(i=1; i<= nlstate; i++){
3494: for(j=1; j<i;j++){
3495: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3496: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3497: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3498: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3499: }
3500: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3501: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3502: }
3503: for(j=i+1; j<=nlstate+ndeath;j++){
3504: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3505: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3506: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3507: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3508: }
3509: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3510: }
3511: }
3512:
3513: for(i=1; i<= nlstate; i++){
3514: s1=0;
3515: for(j=1; j<i; j++){
3516: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3517: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3518: }
3519: for(j=i+1; j<=nlstate+ndeath; j++){
3520: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3521: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3522: }
3523: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3524: ps[i][i]=1./(s1+1.);
3525: /* Computing other pijs */
3526: for(j=1; j<i; j++)
3527: ps[i][j]= exp(ps[i][j])*ps[i][i];
3528: for(j=i+1; j<=nlstate+ndeath; j++)
3529: ps[i][j]= exp(ps[i][j])*ps[i][i];
3530: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3531: } /* end i */
3532:
3533: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3534: for(jj=1; jj<= nlstate+ndeath; jj++){
3535: ps[ii][jj]=0;
3536: ps[ii][ii]=1;
3537: }
3538: }
1.296 brouard 3539: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3540: for(jj=1; jj<= nlstate+ndeath; jj++){
3541: s1=0.;
3542: for(ii=1; ii<= nlstate+ndeath; ii++){
3543: s1+=ps[ii][jj];
3544: }
3545: for(ii=1; ii<= nlstate; ii++){
3546: ps[ii][jj]=ps[ii][jj]/s1;
3547: }
3548: }
3549: /* Transposition */
3550: for(jj=1; jj<= nlstate+ndeath; jj++){
3551: for(ii=jj; ii<= nlstate+ndeath; ii++){
3552: s1=ps[ii][jj];
3553: ps[ii][jj]=ps[jj][ii];
3554: ps[jj][ii]=s1;
3555: }
3556: }
3557: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3558: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3559: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3560: /* } */
3561: /* printf("\n "); */
3562: /* } */
3563: /* printf("\n ");printf("%lf ",cov[2]);*/
3564: /*
3565: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3566: goto end;*/
3567: return ps;
1.217 brouard 3568: }
3569:
3570:
1.126 brouard 3571: /**************** Product of 2 matrices ******************/
3572:
1.145 brouard 3573: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3574: {
3575: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3576: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3577: /* in, b, out are matrice of pointers which should have been initialized
3578: before: only the contents of out is modified. The function returns
3579: a pointer to pointers identical to out */
1.145 brouard 3580: int i, j, k;
1.126 brouard 3581: for(i=nrl; i<= nrh; i++)
1.145 brouard 3582: for(k=ncolol; k<=ncoloh; k++){
3583: out[i][k]=0.;
3584: for(j=ncl; j<=nch; j++)
3585: out[i][k] +=in[i][j]*b[j][k];
3586: }
1.126 brouard 3587: return out;
3588: }
3589:
3590:
3591: /************* Higher Matrix Product ***************/
3592:
1.235 brouard 3593: 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 3594: {
1.336 brouard 3595: /* Already optimized with precov.
3596: 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 3597: 'nhstepm*hstepm*stepm' months (i.e. until
3598: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3599: nhstepm*hstepm matrices.
3600: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3601: (typically every 2 years instead of every month which is too big
3602: for the memory).
3603: Model is determined by parameters x and covariates have to be
3604: included manually here.
3605:
3606: */
3607:
1.330 brouard 3608: int i, j, d, h, k, k1;
1.131 brouard 3609: double **out, cov[NCOVMAX+1];
1.126 brouard 3610: double **newm;
1.187 brouard 3611: double agexact;
1.214 brouard 3612: double agebegin, ageend;
1.126 brouard 3613:
3614: /* Hstepm could be zero and should return the unit matrix */
3615: for (i=1;i<=nlstate+ndeath;i++)
3616: for (j=1;j<=nlstate+ndeath;j++){
3617: oldm[i][j]=(i==j ? 1.0 : 0.0);
3618: po[i][j][0]=(i==j ? 1.0 : 0.0);
3619: }
3620: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3621: for(h=1; h <=nhstepm; h++){
3622: for(d=1; d <=hstepm; d++){
3623: newm=savm;
3624: /* Covariates have to be included here again */
3625: cov[1]=1.;
1.214 brouard 3626: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3627: cov[2]=agexact;
1.319 brouard 3628: if(nagesqr==1){
1.227 brouard 3629: cov[3]= agexact*agexact;
1.319 brouard 3630: }
1.330 brouard 3631: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
3632: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
3633: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.332 brouard 3634: if(Typevar[k1]==1){ /* A product with age */
3635: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
3636: }else{
3637: cov[2+nagesqr+k1]=precov[nres][k1];
3638: }
3639: }/* End of loop on model equation */
3640: /* Old code */
3641: /* if( Dummy[k1]==0 && Typevar[k1]==0 ){ /\* Single dummy *\/ */
3642: /* /\* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) *\/ */
3643: /* /\* for (k=1; k<=nsd;k++) { /\\* For single dummy covariates only *\\/ *\/ */
3644: /* /\* /\\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\\/ *\/ */
3645: /* /\* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 *\/ */
3646: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3647: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3648: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
3649: /* /\* nsd 1 2 3 *\/ /\* Counting single dummies covar fixed or tv *\/ */
3650: /* /\*TvarsD[nsd] 4 3 1 *\/ /\* ID of single dummy cova fixed or timevary*\/ */
3651: /* /\*TvarsDind[k] 2 3 9 *\/ /\* position K of single dummy cova *\/ */
3652: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] *\/ */
3653: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3654: /* /\* 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]])); *\/ */
3655: /* 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); */
3656: /* printf("hpxij new Dummy precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3657: /* }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /\* Single quantitative variables *\/ */
3658: /* /\* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline *\/ */
3659: /* cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]]; */
3660: /* /\* for (k=1; k<=nsq;k++) { /\\* For single varying covariates only *\\/ *\/ */
3661: /* /\* /\\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\\/ *\/ */
3662: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
3663: /* 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]]); */
3664: /* printf("hpxij new Quanti precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3665: /* }else if( Dummy[k1]==2 ){ /\* For dummy with age product *\/ */
3666: /* /\* Tvar[k1] Variable in the age product age*V1 is 1 *\/ */
3667: /* /\* [Tinvresult[nres][V1] is its value in the resultline nres *\/ */
3668: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvar[k1]]*cov[2]; */
3669: /* 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]); */
3670: /* printf("hpxij new Dummy with age product precov[nres=%d][k1=%d]=%.4f * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3671:
3672: /* /\* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; *\/ */
3673: /* /\* for (k=1; k<=cptcovage;k++){ /\\* For product with age V1+V1*age +V4 +age*V3 *\\/ *\/ */
3674: /* /\* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*\/ */
3675: /* /\* *\/ */
1.330 brouard 3676: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3677: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3678: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
1.332 brouard 3679: /* /\*cptcovage=2 1 2 *\/ */
3680: /* /\*Tage[k]= 5 8 *\/ */
3681: /* }else if( Dummy[k1]==3 ){ /\* For quant with age product *\/ */
3682: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3683: /* 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]]); */
3684: /* printf("hpxij new Quanti with age product precov[nres=%d][k1=%d] * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3685: /* /\* if(Dummy[Tage[k]]== 2){ /\\* dummy with age *\\/ *\/ */
3686: /* /\* /\\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\\* dummy with age *\\\/ *\\/ *\/ */
3687: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\\/ *\/ */
3688: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\\/ *\/ */
3689: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
3690: /* /\* 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); *\/ */
3691: /* /\* } else if(Dummy[Tage[k]]== 3){ /\\* quantitative with age *\\/ *\/ */
3692: /* /\* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; *\/ */
3693: /* /\* } *\/ */
3694: /* /\* 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]); *\/ */
3695: /* }else if(Typevar[k1]==2 ){ /\* For product (not with age) *\/ */
3696: /* /\* for (k=1; k<=cptcovprod;k++){ /\\* For product without age *\\/ *\/ */
3697: /* /\* /\\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\\/ *\/ */
3698: /* /\* /\\* k 1 2 3 4 5 6 7 8 9 *\\/ *\/ */
3699: /* /\* /\\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\\/ *\/ */
3700: /* /\* /\\*cptcovprod=1 1 2 *\\/ *\/ */
3701: /* /\* /\\*Tprod[]= 4 7 *\\/ *\/ */
3702: /* /\* /\\*Tvard[][1] 4 1 *\\/ *\/ */
3703: /* /\* /\\*Tvard[][2] 3 2 *\\/ *\/ */
1.330 brouard 3704:
1.332 brouard 3705: /* /\* 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])]); *\/ */
3706: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3707: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]]; */
3708: /* 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]]); */
3709: /* printf("hpxij new Product no age product precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3710:
3711: /* /\* if(Dummy[Tvardk[k1][1]]==0){ *\/ */
3712: /* /\* if(Dummy[Tvardk[k1][2]]==0){ /\\* Product of dummies *\\/ *\/ */
3713: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3714: /* /\* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; *\/ */
3715: /* /\* 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]])]; *\/ */
3716: /* /\* }else{ /\\* Product of dummy by quantitative *\\/ *\/ */
3717: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; *\/ */
3718: /* /\* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; *\/ */
3719: /* /\* } *\/ */
3720: /* /\* }else{ /\\* Product of quantitative by...*\\/ *\/ */
3721: /* /\* if(Dummy[Tvard[k][2]]==0){ /\\* quant by dummy *\\/ *\/ */
3722: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\\/ *\/ */
3723: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3724: /* /\* }else{ /\\* Product of two quant *\\/ *\/ */
3725: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\\/ *\/ */
3726: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3727: /* /\* } *\/ */
3728: /* /\* }/\\*end of products quantitative *\\/ *\/ */
3729: /* }/\*end of products *\/ */
3730: /* } /\* End of loop on model equation *\/ */
1.235 brouard 3731: /* for (k=1; k<=cptcovn;k++) */
3732: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3733: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3734: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3735: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3736: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3737:
3738:
1.126 brouard 3739: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3740: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3741: /* right multiplication of oldm by the current matrix */
1.126 brouard 3742: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3743: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3744: /* if((int)age == 70){ */
3745: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3746: /* for(i=1; i<=nlstate+ndeath; i++) { */
3747: /* printf("%d pmmij ",i); */
3748: /* for(j=1;j<=nlstate+ndeath;j++) { */
3749: /* printf("%f ",pmmij[i][j]); */
3750: /* } */
3751: /* printf(" oldm "); */
3752: /* for(j=1;j<=nlstate+ndeath;j++) { */
3753: /* printf("%f ",oldm[i][j]); */
3754: /* } */
3755: /* printf("\n"); */
3756: /* } */
3757: /* } */
1.126 brouard 3758: savm=oldm;
3759: oldm=newm;
3760: }
3761: for(i=1; i<=nlstate+ndeath; i++)
3762: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3763: po[i][j][h]=newm[i][j];
3764: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3765: }
1.128 brouard 3766: /*printf("h=%d ",h);*/
1.126 brouard 3767: } /* end h */
1.267 brouard 3768: /* printf("\n H=%d \n",h); */
1.126 brouard 3769: return po;
3770: }
3771:
1.217 brouard 3772: /************* Higher Back Matrix Product ***************/
1.218 brouard 3773: /* 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 3774: 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 3775: {
1.332 brouard 3776: /* For dummy covariates given in each resultline (for historical, computes the corresponding combination ij),
3777: computes the transition matrix starting at age 'age' over
1.217 brouard 3778: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3779: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3780: nhstepm*hstepm matrices.
3781: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3782: (typically every 2 years instead of every month which is too big
1.217 brouard 3783: for the memory).
1.218 brouard 3784: Model is determined by parameters x and covariates have to be
1.266 brouard 3785: included manually here. Then we use a call to bmij(x and cov)
3786: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3787: */
1.217 brouard 3788:
1.332 brouard 3789: int i, j, d, h, k, k1;
1.266 brouard 3790: double **out, cov[NCOVMAX+1], **bmij();
3791: double **newm, ***newmm;
1.217 brouard 3792: double agexact;
3793: double agebegin, ageend;
1.222 brouard 3794: double **oldm, **savm;
1.217 brouard 3795:
1.266 brouard 3796: newmm=po; /* To be saved */
3797: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3798: /* Hstepm could be zero and should return the unit matrix */
3799: for (i=1;i<=nlstate+ndeath;i++)
3800: for (j=1;j<=nlstate+ndeath;j++){
3801: oldm[i][j]=(i==j ? 1.0 : 0.0);
3802: po[i][j][0]=(i==j ? 1.0 : 0.0);
3803: }
3804: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3805: for(h=1; h <=nhstepm; h++){
3806: for(d=1; d <=hstepm; d++){
3807: newm=savm;
3808: /* Covariates have to be included here again */
3809: cov[1]=1.;
1.271 brouard 3810: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3811: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3812: /* Debug */
3813: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3814: cov[2]=agexact;
1.332 brouard 3815: if(nagesqr==1){
1.222 brouard 3816: cov[3]= agexact*agexact;
1.332 brouard 3817: }
3818: /** New code */
3819: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3820: if(Typevar[k1]==1){ /* A product with age */
3821: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.325 brouard 3822: }else{
1.332 brouard 3823: cov[2+nagesqr+k1]=precov[nres][k1];
1.325 brouard 3824: }
1.332 brouard 3825: }/* End of loop on model equation */
3826: /** End of new code */
3827: /** This was old code */
3828: /* for (k=1; k<=nsd;k++){ /\* For single dummy covariates only *\//\* cptcovn error *\/ */
3829: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3830: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3831: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/\* Bug valgrind *\/ */
3832: /* /\* 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)); *\/ */
3833: /* } */
3834: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3835: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3836: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3837: /* /\* 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]); *\/ */
3838: /* } */
3839: /* for (k=1; k<=cptcovage;k++){ /\* Should start at cptcovn+1 *\//\* For product with age *\/ */
3840: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age error!!!*\\/ *\/ */
3841: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3842: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3843: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3844: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3845: /* } */
3846: /* /\* 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]); *\/ */
3847: /* } */
3848: /* for (k=1; k<=cptcovprod;k++){ /\* Useless because included in cptcovn *\/ */
3849: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3850: /* if(Dummy[Tvard[k][1]]==0){ */
3851: /* if(Dummy[Tvard[k][2]]==0){ */
3852: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]; */
3853: /* }else{ */
3854: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3855: /* } */
3856: /* }else{ */
3857: /* if(Dummy[Tvard[k][2]]==0){ */
3858: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3859: /* }else{ */
3860: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3861: /* } */
3862: /* } */
3863: /* } */
3864: /* /\*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*\/ */
3865: /* /\*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*\/ */
3866: /** End of old code */
3867:
1.218 brouard 3868: /* Careful transposed matrix */
1.266 brouard 3869: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3870: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3871: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3872: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3873: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3874: /* if((int)age == 70){ */
3875: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3876: /* for(i=1; i<=nlstate+ndeath; i++) { */
3877: /* printf("%d pmmij ",i); */
3878: /* for(j=1;j<=nlstate+ndeath;j++) { */
3879: /* printf("%f ",pmmij[i][j]); */
3880: /* } */
3881: /* printf(" oldm "); */
3882: /* for(j=1;j<=nlstate+ndeath;j++) { */
3883: /* printf("%f ",oldm[i][j]); */
3884: /* } */
3885: /* printf("\n"); */
3886: /* } */
3887: /* } */
3888: savm=oldm;
3889: oldm=newm;
3890: }
3891: for(i=1; i<=nlstate+ndeath; i++)
3892: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3893: po[i][j][h]=newm[i][j];
1.268 brouard 3894: /* if(h==nhstepm) */
3895: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3896: }
1.268 brouard 3897: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3898: } /* end h */
1.268 brouard 3899: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3900: return po;
3901: }
3902:
3903:
1.162 brouard 3904: #ifdef NLOPT
3905: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3906: double fret;
3907: double *xt;
3908: int j;
3909: myfunc_data *d2 = (myfunc_data *) pd;
3910: /* xt = (p1-1); */
3911: xt=vector(1,n);
3912: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3913:
3914: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3915: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3916: printf("Function = %.12lf ",fret);
3917: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3918: printf("\n");
3919: free_vector(xt,1,n);
3920: return fret;
3921: }
3922: #endif
1.126 brouard 3923:
3924: /*************** log-likelihood *************/
3925: double func( double *x)
3926: {
1.336 brouard 3927: int i, ii, j, k, mi, d, kk, kf=0;
1.226 brouard 3928: int ioffset=0;
1.339 brouard 3929: int ipos=0,iposold=0,ncovv=0;
3930:
1.340 brouard 3931: double cotvarv, cotvarvold;
1.226 brouard 3932: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3933: double **out;
3934: double lli; /* Individual log likelihood */
3935: int s1, s2;
1.228 brouard 3936: 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 3937:
1.226 brouard 3938: double bbh, survp;
3939: double agexact;
1.336 brouard 3940: double agebegin, ageend;
1.226 brouard 3941: /*extern weight */
3942: /* We are differentiating ll according to initial status */
3943: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3944: /*for(i=1;i<imx;i++)
3945: printf(" %d\n",s[4][i]);
3946: */
1.162 brouard 3947:
1.226 brouard 3948: ++countcallfunc;
1.162 brouard 3949:
1.226 brouard 3950: cov[1]=1.;
1.126 brouard 3951:
1.226 brouard 3952: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3953: ioffset=0;
1.226 brouard 3954: if(mle==1){
3955: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3956: /* Computes the values of the ncovmodel covariates of the model
3957: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3958: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3959: to be observed in j being in i according to the model.
3960: */
1.243 brouard 3961: ioffset=2+nagesqr ;
1.233 brouard 3962: /* Fixed */
1.345 brouard 3963: for (kf=1; kf<=ncovf;kf++){ /* For each fixed covariate dummy or quant or prod */
1.319 brouard 3964: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3965: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3966: /* 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 3967: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.336 brouard 3968: 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 3969: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3970: }
1.226 brouard 3971: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3972: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3973: has been calculated etc */
3974: /* For an individual i, wav[i] gives the number of effective waves */
3975: /* We compute the contribution to Likelihood of each effective transition
3976: mw[mi][i] is real wave of the mi th effectve wave */
3977: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3978: s2=s[mw[mi+1][i]][i];
1.341 brouard 3979: 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 3980: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3981: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3982: */
1.336 brouard 3983: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
3984: /* Wave varying (but not age varying) */
1.339 brouard 3985: /* 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*\/ */
3986: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? *\/ */
3987: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
3988: /* } */
1.340 brouard 3989: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age )*/
3990: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
3991: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
1.345 brouard 3992: if(FixedV[itv]!=0){ /* Not a fixed covariate */
1.341 brouard 3993: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* cotvar[wav][ncovcol+nqv+iv][i] */
1.340 brouard 3994: }else{ /* fixed covariate */
1.345 brouard 3995: 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 3996: }
1.339 brouard 3997: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 3998: cotvarvold=cotvarv;
3999: }else{ /* A second product */
4000: cotvarv=cotvarv*cotvarvold;
1.339 brouard 4001: }
4002: iposold=ipos;
1.340 brouard 4003: cov[ioffset+ipos]=cotvarv;
1.234 brouard 4004: }
1.339 brouard 4005: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
4006: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4007: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
4008: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
4009: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
4010: /* printf(" i=%d,mi=%d,itv=%d,TmodelInvind[itv]=%d,cotvar[mw[mi][i]][TmodelInvind[itv]][i]=%f\n", i, mi, itv, TmodelInvind[itv],cotvar[mw[mi][i]][TmodelInvind[itv]][i]); */
4011: /* } */
4012: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
4013: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4014: /* /\* 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]); *\/ */
4015: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
4016: /* } */
4017: /* for products of time varying to be done */
1.234 brouard 4018: for (ii=1;ii<=nlstate+ndeath;ii++)
4019: for (j=1;j<=nlstate+ndeath;j++){
4020: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4021: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4022: }
1.336 brouard 4023:
4024: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4025: 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 4026: for(d=0; d<dh[mi][i]; d++){
4027: newm=savm;
4028: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4029: cov[2]=agexact;
4030: if(nagesqr==1)
4031: cov[3]= agexact*agexact; /* Should be changed here */
4032: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 4033: if(!FixedV[Tvar[Tage[kk]]])
4034: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4035: else
1.341 brouard 4036: 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) */
1.234 brouard 4037: }
4038: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4039: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4040: savm=oldm;
4041: oldm=newm;
4042: } /* end mult */
4043:
4044: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
4045: /* But now since version 0.9 we anticipate for bias at large stepm.
4046: * If stepm is larger than one month (smallest stepm) and if the exact delay
4047: * (in months) between two waves is not a multiple of stepm, we rounded to
4048: * the nearest (and in case of equal distance, to the lowest) interval but now
4049: * we keep into memory the bias bh[mi][i] and also the previous matrix product
4050: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
4051: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 4052: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
4053: * -stepm/2 to stepm/2 .
4054: * For stepm=1 the results are the same as for previous versions of Imach.
4055: * For stepm > 1 the results are less biased than in previous versions.
4056: */
1.234 brouard 4057: s1=s[mw[mi][i]][i];
4058: s2=s[mw[mi+1][i]][i];
4059: bbh=(double)bh[mi][i]/(double)stepm;
4060: /* bias bh is positive if real duration
4061: * is higher than the multiple of stepm and negative otherwise.
4062: */
4063: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
4064: if( s2 > nlstate){
4065: /* i.e. if s2 is a death state and if the date of death is known
4066: then the contribution to the likelihood is the probability to
4067: die between last step unit time and current step unit time,
4068: which is also equal to probability to die before dh
4069: minus probability to die before dh-stepm .
4070: In version up to 0.92 likelihood was computed
4071: as if date of death was unknown. Death was treated as any other
4072: health state: the date of the interview describes the actual state
4073: and not the date of a change in health state. The former idea was
4074: to consider that at each interview the state was recorded
4075: (healthy, disable or death) and IMaCh was corrected; but when we
4076: introduced the exact date of death then we should have modified
4077: the contribution of an exact death to the likelihood. This new
4078: contribution is smaller and very dependent of the step unit
4079: stepm. It is no more the probability to die between last interview
4080: and month of death but the probability to survive from last
4081: interview up to one month before death multiplied by the
4082: probability to die within a month. Thanks to Chris
4083: Jackson for correcting this bug. Former versions increased
4084: mortality artificially. The bad side is that we add another loop
4085: which slows down the processing. The difference can be up to 10%
4086: lower mortality.
4087: */
4088: /* If, at the beginning of the maximization mostly, the
4089: cumulative probability or probability to be dead is
4090: constant (ie = 1) over time d, the difference is equal to
4091: 0. out[s1][3] = savm[s1][3]: probability, being at state
4092: s1 at precedent wave, to be dead a month before current
4093: wave is equal to probability, being at state s1 at
4094: precedent wave, to be dead at mont of the current
4095: wave. Then the observed probability (that this person died)
4096: is null according to current estimated parameter. In fact,
4097: it should be very low but not zero otherwise the log go to
4098: infinity.
4099: */
1.183 brouard 4100: /* #ifdef INFINITYORIGINAL */
4101: /* lli=log(out[s1][s2] - savm[s1][s2]); */
4102: /* #else */
4103: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
4104: /* lli=log(mytinydouble); */
4105: /* else */
4106: /* lli=log(out[s1][s2] - savm[s1][s2]); */
4107: /* #endif */
1.226 brouard 4108: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4109:
1.226 brouard 4110: } else if ( s2==-1 ) { /* alive */
4111: for (j=1,survp=0. ; j<=nlstate; j++)
4112: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4113: /*survp += out[s1][j]; */
4114: lli= log(survp);
4115: }
1.336 brouard 4116: /* else if (s2==-4) { */
4117: /* for (j=3,survp=0. ; j<=nlstate; j++) */
4118: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4119: /* lli= log(survp); */
4120: /* } */
4121: /* else if (s2==-5) { */
4122: /* for (j=1,survp=0. ; j<=2; j++) */
4123: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4124: /* lli= log(survp); */
4125: /* } */
1.226 brouard 4126: else{
4127: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4128: /* 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 */
4129: }
4130: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
4131: /*if(lli ==000.0)*/
1.340 brouard 4132: /* 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 4133: ipmx +=1;
4134: sw += weight[i];
4135: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4136: /* if (lli < log(mytinydouble)){ */
4137: /* 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); */
4138: /* 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]); */
4139: /* } */
4140: } /* end of wave */
4141: } /* end of individual */
4142: } else if(mle==2){
4143: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 4144: ioffset=2+nagesqr ;
4145: for (k=1; k<=ncovf;k++)
4146: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 4147: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 4148: for(k=1; k <= ncovv ; k++){
1.341 brouard 4149: 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 4150: }
1.226 brouard 4151: for (ii=1;ii<=nlstate+ndeath;ii++)
4152: for (j=1;j<=nlstate+ndeath;j++){
4153: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4154: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4155: }
4156: for(d=0; d<=dh[mi][i]; d++){
4157: newm=savm;
4158: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4159: cov[2]=agexact;
4160: if(nagesqr==1)
4161: cov[3]= agexact*agexact;
4162: for (kk=1; kk<=cptcovage;kk++) {
4163: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4164: }
4165: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4166: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4167: savm=oldm;
4168: oldm=newm;
4169: } /* end mult */
4170:
4171: s1=s[mw[mi][i]][i];
4172: s2=s[mw[mi+1][i]][i];
4173: bbh=(double)bh[mi][i]/(double)stepm;
4174: 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 */
4175: ipmx +=1;
4176: sw += weight[i];
4177: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4178: } /* end of wave */
4179: } /* end of individual */
4180: } else if(mle==3){ /* exponential inter-extrapolation */
4181: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4182: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4183: for(mi=1; mi<= wav[i]-1; mi++){
4184: for (ii=1;ii<=nlstate+ndeath;ii++)
4185: for (j=1;j<=nlstate+ndeath;j++){
4186: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4187: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4188: }
4189: for(d=0; d<dh[mi][i]; d++){
4190: newm=savm;
4191: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4192: cov[2]=agexact;
4193: if(nagesqr==1)
4194: cov[3]= agexact*agexact;
4195: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 4196: if(!FixedV[Tvar[Tage[kk]]])
4197: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4198: else
1.341 brouard 4199: 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 4200: }
4201: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4202: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4203: savm=oldm;
4204: oldm=newm;
4205: } /* end mult */
4206:
4207: s1=s[mw[mi][i]][i];
4208: s2=s[mw[mi+1][i]][i];
4209: bbh=(double)bh[mi][i]/(double)stepm;
4210: 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 */
4211: ipmx +=1;
4212: sw += weight[i];
4213: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4214: } /* end of wave */
4215: } /* end of individual */
4216: }else if (mle==4){ /* ml=4 no inter-extrapolation */
4217: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4218: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4219: for(mi=1; mi<= wav[i]-1; mi++){
4220: for (ii=1;ii<=nlstate+ndeath;ii++)
4221: for (j=1;j<=nlstate+ndeath;j++){
4222: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4223: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4224: }
4225: for(d=0; d<dh[mi][i]; d++){
4226: newm=savm;
4227: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4228: cov[2]=agexact;
4229: if(nagesqr==1)
4230: cov[3]= agexact*agexact;
4231: for (kk=1; kk<=cptcovage;kk++) {
4232: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4233: }
1.126 brouard 4234:
1.226 brouard 4235: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4236: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4237: savm=oldm;
4238: oldm=newm;
4239: } /* end mult */
4240:
4241: s1=s[mw[mi][i]][i];
4242: s2=s[mw[mi+1][i]][i];
4243: if( s2 > nlstate){
4244: lli=log(out[s1][s2] - savm[s1][s2]);
4245: } else if ( s2==-1 ) { /* alive */
4246: for (j=1,survp=0. ; j<=nlstate; j++)
4247: survp += out[s1][j];
4248: lli= log(survp);
4249: }else{
4250: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4251: }
4252: ipmx +=1;
4253: sw += weight[i];
4254: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.343 brouard 4255: /* 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 4256: } /* end of wave */
4257: } /* end of individual */
4258: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
4259: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4260: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4261: for(mi=1; mi<= wav[i]-1; mi++){
4262: for (ii=1;ii<=nlstate+ndeath;ii++)
4263: for (j=1;j<=nlstate+ndeath;j++){
4264: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4265: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4266: }
4267: for(d=0; d<dh[mi][i]; d++){
4268: newm=savm;
4269: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4270: cov[2]=agexact;
4271: if(nagesqr==1)
4272: cov[3]= agexact*agexact;
4273: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 4274: if(!FixedV[Tvar[Tage[kk]]])
4275: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4276: else
1.341 brouard 4277: 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 4278: }
1.126 brouard 4279:
1.226 brouard 4280: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4281: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4282: savm=oldm;
4283: oldm=newm;
4284: } /* end mult */
4285:
4286: s1=s[mw[mi][i]][i];
4287: s2=s[mw[mi+1][i]][i];
4288: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4289: ipmx +=1;
4290: sw += weight[i];
4291: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4292: /*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]);*/
4293: } /* end of wave */
4294: } /* end of individual */
4295: } /* End of if */
4296: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4297: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4298: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4299: return -l;
1.126 brouard 4300: }
4301:
4302: /*************** log-likelihood *************/
4303: double funcone( double *x)
4304: {
1.228 brouard 4305: /* Same as func but slower because of a lot of printf and if */
1.335 brouard 4306: int i, ii, j, k, mi, d, kk, kf=0;
1.228 brouard 4307: int ioffset=0;
1.339 brouard 4308: int ipos=0,iposold=0,ncovv=0;
4309:
1.340 brouard 4310: double cotvarv, cotvarvold;
1.131 brouard 4311: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 4312: double **out;
4313: double lli; /* Individual log likelihood */
4314: double llt;
4315: int s1, s2;
1.228 brouard 4316: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
4317:
1.126 brouard 4318: double bbh, survp;
1.187 brouard 4319: double agexact;
1.214 brouard 4320: double agebegin, ageend;
1.126 brouard 4321: /*extern weight */
4322: /* We are differentiating ll according to initial status */
4323: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4324: /*for(i=1;i<imx;i++)
4325: printf(" %d\n",s[4][i]);
4326: */
4327: cov[1]=1.;
4328:
4329: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4330: ioffset=0;
4331: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.336 brouard 4332: /* Computes the values of the ncovmodel covariates of the model
4333: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
4334: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
4335: to be observed in j being in i according to the model.
4336: */
1.243 brouard 4337: /* ioffset=2+nagesqr+cptcovage; */
4338: ioffset=2+nagesqr;
1.232 brouard 4339: /* Fixed */
1.224 brouard 4340: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4341: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.335 brouard 4342: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
1.339 brouard 4343: /* printf("Debug3 TvarFind[%d]=%d",kf, TvarFind[kf]); */
4344: /* printf(" Tvar[TvarFind[kf]]=%d", Tvar[TvarFind[kf]]); */
4345: /* printf(" i=%d covar[Tvar[TvarFind[kf]]][i]=%f\n",i,covar[Tvar[TvarFind[kf]]][i]); */
1.335 brouard 4346: 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 4347: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4348: /* cov[2+6]=covar[Tvar[6]][i]; */
4349: /* cov[2+6]=covar[2][i]; V2 */
4350: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4351: /* cov[2+7]=covar[Tvar[7]][i]; */
4352: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4353: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4354: /* cov[2+9]=covar[Tvar[9]][i]; */
4355: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4356: }
1.336 brouard 4357: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
4358: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
4359: has been calculated etc */
4360: /* For an individual i, wav[i] gives the number of effective waves */
4361: /* We compute the contribution to Likelihood of each effective transition
4362: mw[mi][i] is real wave of the mi th effectve wave */
4363: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
4364: s2=s[mw[mi+1][i]][i];
1.341 brouard 4365: And the iv th varying covariate in the DATA is the cotvar[mw[mi+1][i]][ncovcol+nqv+iv][i]
1.336 brouard 4366: */
4367: /* This part may be useless now because everythin should be in covar */
1.232 brouard 4368: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4369: /* 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?)*\/ */
4370: /* } */
1.231 brouard 4371: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4372: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4373: /* } */
1.225 brouard 4374:
1.233 brouard 4375:
4376: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.339 brouard 4377: /* 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 */
4378: /* for(k=1; k <= ncovv ; k++){ /\* Varying covariates (single and product but no age )*\/ */
4379: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; *\/ */
4380: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
4381: /* } */
4382:
4383: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
4384: /* model V1+V3+age*V1+age*V3+V1*V3 */
4385: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
4386: /* TvarVV[1]=V3 (first time varying in the model equation, TvarVV[2]=V1 (in V1*V3) TvarVV[3]=3(V3) */
4387: /* We need the position of the time varying or product in the model */
4388: /* 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 */
4389: /* TvarVV gives the variable name */
1.340 brouard 4390: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
4391: * k= 1 2 3 4 5 6 7 8 9
4392: * varying 1 2 3 4 5
4393: * ncovv 1 2 3 4 5 6 7 8
1.343 brouard 4394: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
1.340 brouard 4395: * TvarVVind 2 3 7 7 8 8 9 9
4396: * TvarFind[k] 1 0 0 0 0 0 0 0 0
4397: */
1.345 brouard 4398: /* Other model ncovcol=5 nqv=0 ntv=3 nqtv=0 nlstate=3
1.346 brouard 4399: * 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[4]=6
1.345 brouard 4400: * FixedV[ncovcol+qv+ntv+nqtv] V5
4401: * V1 V2 V3 V4 V5 V6 V7 V8
4402: * 0 0 0 0 0 1 1 1
4403: * model= V2 + V3 + V4 + V6 + V7 + V6*V2 + V7*V2 + V6*V3 + V7*V3 + V6*V4 + V7*V4
4404: * kmodel 1 2 3 4 5 6 7 8 9 10 11
4405: * ncovf 1 2 3
4406: * ncovvt=14 1 2 3 4 5 6 7 8 9 10 11 12 13 14
4407: * TvarVV[1]@14 = itv {6, 7, 6, 2, 7, 2, 6, 3, 7, 3, 6, 4, 7, 4}
4408: * TvarVVind[1]@14= {4, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11}
4409: * TvarFind[1]@14= {1, 2, 3, 0 <repeats 12 times>}
4410: * Tvar[1]@20= {2, 3, 4, 6, 7, 9, 10, 11, 12, 13, 14}
4411: * TvarFind[itv] 0 0 0
4412: * FixedV[itv] 1 1 1 0 1 0 1 0 1 0 0
4413: * Tvar[TvarFind[ncovf]]=[1]=2 [2]=3 [4]=4
4414: * Tvar[TvarFind[itv]] [0]=? ?ncovv 1 à ncovvt]
4415: * Not a fixed cotvar[mw][itv][i] 6 7 6 2 7, 2, 6, 3, 7, 3, 6, 4, 7, 4}
4416: * fixed covar[itv] [6] [7] [6][2]
4417: */
4418:
1.340 brouard 4419: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
1.345 brouard 4420: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate, exploding product */
1.340 brouard 4421: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
1.345 brouard 4422: /* if(TvarFind[itv]==0){ /\* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv *\/ */
4423: if(FixedV[itv]!=0){ /* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv */
4424: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* because cotvar starts now at first ncovcol+nqv+ntv+nqtv (1 to nqtv) */
1.340 brouard 4425: }else{ /* fixed covariate */
1.345 brouard 4426: /* cotvarv=covar[Tvar[TvarFind[itv]]][i]; /\* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model *\/ */
4427: 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 4428: }
1.339 brouard 4429: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 4430: cotvarvold=cotvarv;
4431: }else{ /* A second product */
4432: cotvarv=cotvarv*cotvarvold;
1.339 brouard 4433: }
4434: iposold=ipos;
1.340 brouard 4435: cov[ioffset+ipos]=cotvarv;
1.339 brouard 4436: /* For products */
4437: }
4438: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates single *\/ */
4439: /* iv=TvarVDind[itv]; /\* iv, position in the model equation of time varying covariate itv *\/ */
4440: /* /\* "V1+V3+age*V1+age*V3+V1*V3" with V3 time varying *\/ */
4441: /* /\* 1 2 3 4 5 *\/ */
4442: /* /\*itv 1 *\/ */
4443: /* /\* TvarVInd[1]= 2 *\/ */
4444: /* /\* iv= Tvar[Tmodelind[itv]]-ncovcol-nqv; /\\* Counting the # varying covariate from 1 to ntveff *\\/ *\/ */
4445: /* /\* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; *\/ */
4446: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; *\/ */
4447: /* /\* k=ioffset-2-nagesqr-cptcovage+itv; /\\* position in simple model *\\/ *\/ */
4448: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; *\/ */
4449: /* cov[ioffset+iv]=cotvar[mw[mi][i]][itv][i]; */
4450: /* /\* 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]); *\/ */
4451: /* } */
1.232 brouard 4452: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4453: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4454: /* /\* 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]); *\/ */
4455: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4456: /* } */
1.126 brouard 4457: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4458: for (j=1;j<=nlstate+ndeath;j++){
4459: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4460: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4461: }
1.214 brouard 4462:
4463: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4464: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4465: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4466: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4467: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4468: and mw[mi+1][i]. dh depends on stepm.*/
4469: newm=savm;
1.247 brouard 4470: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4471: cov[2]=agexact;
4472: if(nagesqr==1)
4473: cov[3]= agexact*agexact;
4474: for (kk=1; kk<=cptcovage;kk++) {
4475: if(!FixedV[Tvar[Tage[kk]]])
4476: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4477: else
1.341 brouard 4478: 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) */
1.242 brouard 4479: }
4480: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4481: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4482: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4483: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4484: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4485: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4486: savm=oldm;
4487: oldm=newm;
1.126 brouard 4488: } /* end mult */
1.336 brouard 4489: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
4490: /* But now since version 0.9 we anticipate for bias at large stepm.
4491: * If stepm is larger than one month (smallest stepm) and if the exact delay
4492: * (in months) between two waves is not a multiple of stepm, we rounded to
4493: * the nearest (and in case of equal distance, to the lowest) interval but now
4494: * we keep into memory the bias bh[mi][i] and also the previous matrix product
4495: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
4496: * probability in order to take into account the bias as a fraction of the way
4497: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
4498: * -stepm/2 to stepm/2 .
4499: * For stepm=1 the results are the same as for previous versions of Imach.
4500: * For stepm > 1 the results are less biased than in previous versions.
4501: */
1.126 brouard 4502: s1=s[mw[mi][i]][i];
4503: s2=s[mw[mi+1][i]][i];
1.217 brouard 4504: /* if(s2==-1){ */
1.268 brouard 4505: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4506: /* /\* exit(1); *\/ */
4507: /* } */
1.126 brouard 4508: bbh=(double)bh[mi][i]/(double)stepm;
4509: /* bias is positive if real duration
4510: * is higher than the multiple of stepm and negative otherwise.
4511: */
4512: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4513: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4514: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4515: for (j=1,survp=0. ; j<=nlstate; j++)
4516: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4517: lli= log(survp);
1.126 brouard 4518: }else if (mle==1){
1.242 brouard 4519: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4520: } else if(mle==2){
1.242 brouard 4521: 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 4522: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4523: 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 4524: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4525: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4526: } else{ /* mle=0 back to 1 */
1.242 brouard 4527: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4528: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4529: } /* End of if */
4530: ipmx +=1;
4531: sw += weight[i];
4532: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.342 brouard 4533: /* Printing covariates values for each contribution for checking */
1.343 brouard 4534: /* 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 4535: if(globpr){
1.246 brouard 4536: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4537: %11.6f %11.6f %11.6f ", \
1.242 brouard 4538: 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 4539: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.343 brouard 4540: /* printf("%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\ */
4541: /* %11.6f %11.6f %11.6f ", \ */
4542: /* num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw, */
4543: /* 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.242 brouard 4544: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4545: llt +=ll[k]*gipmx/gsw;
4546: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
1.335 brouard 4547: /* printf(" %10.6f",-ll[k]*gipmx/gsw); */
1.242 brouard 4548: }
1.343 brouard 4549: fprintf(ficresilk," %10.6f ", -llt);
1.335 brouard 4550: /* printf(" %10.6f\n", -llt); */
1.342 brouard 4551: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
1.343 brouard 4552: /* fprintf(ficresilk,"%09ld ", num[i]); */ /* not necessary */
4553: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
4554: fprintf(ficresilk," %g",covar[Tvar[TvarFind[kf]]][i]);
4555: }
4556: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
4557: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
4558: if(ipos!=iposold){ /* Not a product or first of a product */
4559: fprintf(ficresilk," %g",cov[ioffset+ipos]);
4560: /* printf(" %g",cov[ioffset+ipos]); */
4561: }else{
4562: fprintf(ficresilk,"*");
4563: /* printf("*"); */
1.342 brouard 4564: }
1.343 brouard 4565: iposold=ipos;
4566: }
4567: for (kk=1; kk<=cptcovage;kk++) {
4568: if(!FixedV[Tvar[Tage[kk]]]){
4569: fprintf(ficresilk," %g*age",covar[Tvar[Tage[kk]]][i]);
4570: /* printf(" %g*age",covar[Tvar[Tage[kk]]][i]); */
4571: }else{
4572: fprintf(ficresilk," %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
4573: /* printf(" %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
1.342 brouard 4574: }
1.343 brouard 4575: }
4576: /* printf("\n"); */
1.342 brouard 4577: /* } /\* End debugILK *\/ */
4578: fprintf(ficresilk,"\n");
4579: } /* End if globpr */
1.335 brouard 4580: } /* end of wave */
4581: } /* end of individual */
4582: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
1.232 brouard 4583: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
1.335 brouard 4584: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4585: if(globpr==0){ /* First time we count the contributions and weights */
4586: gipmx=ipmx;
4587: gsw=sw;
4588: }
1.343 brouard 4589: return -l;
1.126 brouard 4590: }
4591:
4592:
4593: /*************** function likelione ***********/
1.292 brouard 4594: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4595: {
4596: /* This routine should help understanding what is done with
4597: the selection of individuals/waves and
4598: to check the exact contribution to the likelihood.
4599: Plotting could be done.
1.342 brouard 4600: */
4601: void pstamp(FILE *ficres);
1.343 brouard 4602: int k, kf, kk, kvar, ncovv, iposold, ipos;
1.126 brouard 4603:
4604: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4605: strcpy(fileresilk,"ILK_");
1.202 brouard 4606: strcat(fileresilk,fileresu);
1.126 brouard 4607: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4608: printf("Problem with resultfile: %s\n", fileresilk);
4609: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4610: }
1.342 brouard 4611: pstamp(ficresilk);fprintf(ficresilk,"# model=1+age+%s\n",model);
1.214 brouard 4612: 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");
4613: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4614: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4615: for(k=1; k<=nlstate; k++)
4616: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
1.342 brouard 4617: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total) ");
4618:
4619: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
4620: for(kf=1;kf <= ncovf; kf++){
4621: fprintf(ficresilk,"V%d",Tvar[TvarFind[kf]]);
4622: /* printf("V%d",Tvar[TvarFind[kf]]); */
4623: }
4624: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){
1.343 brouard 4625: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
1.342 brouard 4626: if(ipos!=iposold){ /* Not a product or first of a product */
4627: /* printf(" %d",ipos); */
4628: fprintf(ficresilk," V%d",TvarVV[ncovv]);
4629: }else{
4630: /* printf("*"); */
4631: fprintf(ficresilk,"*");
1.343 brouard 4632: }
1.342 brouard 4633: iposold=ipos;
4634: }
4635: for (kk=1; kk<=cptcovage;kk++) {
4636: if(!FixedV[Tvar[Tage[kk]]]){
4637: /* printf(" %d*age(Fixed)",Tvar[Tage[kk]]); */
4638: fprintf(ficresilk," %d*age(Fixed)",Tvar[Tage[kk]]);
4639: }else{
4640: fprintf(ficresilk," %d*age(Varying)",Tvar[Tage[kk]]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
4641: /* printf(" %d*age(Varying)",Tvar[Tage[kk]]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
4642: }
4643: }
4644: /* } /\* End if debugILK *\/ */
4645: /* printf("\n"); */
4646: fprintf(ficresilk,"\n");
4647: } /* End glogpri */
1.126 brouard 4648:
1.292 brouard 4649: *fretone=(*func)(p);
1.126 brouard 4650: if(*globpri !=0){
4651: fclose(ficresilk);
1.205 brouard 4652: if (mle ==0)
4653: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4654: else if(mle >=1)
4655: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4656: 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 4657: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4658:
1.207 brouard 4659: 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 4660: <img src=\"%s-ori.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4661: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.343 brouard 4662: <img src=\"%s-dest.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
4663:
4664: for (k=1; k<= nlstate ; k++) {
4665: 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 \
4666: <img src=\"%s-p%dj.png\">\n",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4667: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
4668: /* kvar=Tvar[TvarFind[kf]]; */ /* variable */
4669: 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): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \
4670: <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]]);
4671: }
4672: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Loop on the time varying extended covariates (with extension of Vn*Vm */
4673: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
4674: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
4675: /* 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]); */
4676: if(ipos!=iposold){ /* Not a product or first of a product */
4677: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
4678: /* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); */
4679: 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) */
4680: 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> \
4681: <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);
4682: } /* End only for dummies time varying (single?) */
4683: }else{ /* Useless product */
4684: /* printf("*"); */
4685: /* fprintf(ficresilk,"*"); */
4686: }
4687: iposold=ipos;
4688: } /* For each time varying covariate */
4689: } /* End loop on states */
4690:
4691: /* if(debugILK){ */
4692: /* for(kf=1; kf <= ncovf; kf++){ /\* For each simple dummy covariate of the model *\/ */
4693: /* /\* kvar=Tvar[TvarFind[kf]]; *\/ /\* variable *\/ */
4694: /* for (k=1; k<= nlstate ; k++) { */
4695: /* 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> \ */
4696: /* <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]]); */
4697: /* } */
4698: /* } */
4699: /* for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /\* Loop on the time varying extended covariates (with extension of Vn*Vm *\/ */
4700: /* ipos=TvarVVind[ncovv]; /\* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate *\/ */
4701: /* kvar=TvarVV[ncovv]; /\* TvarVV={3, 1, 3} gives the name of each varying covariate *\/ */
4702: /* /\* 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]); *\/ */
4703: /* if(ipos!=iposold){ /\* Not a product or first of a product *\/ */
4704: /* /\* fprintf(ficresilk," V%d",TvarVV[ncovv]); *\/ */
4705: /* /\* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); *\/ */
4706: /* 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) *\/ */
4707: /* for (k=1; k<= nlstate ; k++) { */
4708: /* 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> \ */
4709: /* <img src=\"%s-p%dj-%d.png\">",k,k,kvar,kvar,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,kvar); */
4710: /* } /\* End state *\/ */
4711: /* } /\* End only for dummies time varying (single?) *\/ */
4712: /* }else{ /\* Useless product *\/ */
4713: /* /\* printf("*"); *\/ */
4714: /* /\* fprintf(ficresilk,"*"); *\/ */
4715: /* } */
4716: /* iposold=ipos; */
4717: /* } /\* For each time varying covariate *\/ */
4718: /* }/\* End debugILK *\/ */
1.207 brouard 4719: fflush(fichtm);
1.343 brouard 4720: }/* End globpri */
1.126 brouard 4721: return;
4722: }
4723:
4724:
4725: /*********** Maximum Likelihood Estimation ***************/
4726:
4727: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4728: {
1.319 brouard 4729: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4730: double **xi;
4731: double fret;
4732: double fretone; /* Only one call to likelihood */
4733: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4734:
4735: #ifdef NLOPT
4736: int creturn;
4737: nlopt_opt opt;
4738: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4739: double *lb;
4740: double minf; /* the minimum objective value, upon return */
4741: double * p1; /* Shifted parameters from 0 instead of 1 */
4742: myfunc_data dinst, *d = &dinst;
4743: #endif
4744:
4745:
1.126 brouard 4746: xi=matrix(1,npar,1,npar);
4747: for (i=1;i<=npar;i++)
4748: for (j=1;j<=npar;j++)
4749: xi[i][j]=(i==j ? 1.0 : 0.0);
4750: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4751: strcpy(filerespow,"POW_");
1.126 brouard 4752: strcat(filerespow,fileres);
4753: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4754: printf("Problem with resultfile: %s\n", filerespow);
4755: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4756: }
4757: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4758: for (i=1;i<=nlstate;i++)
4759: for(j=1;j<=nlstate+ndeath;j++)
4760: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4761: fprintf(ficrespow,"\n");
1.162 brouard 4762: #ifdef POWELL
1.319 brouard 4763: #ifdef LINMINORIGINAL
4764: #else /* LINMINORIGINAL */
4765:
4766: flatdir=ivector(1,npar);
4767: for (j=1;j<=npar;j++) flatdir[j]=0;
4768: #endif /*LINMINORIGINAL */
4769:
4770: #ifdef FLATSUP
4771: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4772: /* reorganizing p by suppressing flat directions */
4773: for(i=1, jk=1; i <=nlstate; i++){
4774: for(k=1; k <=(nlstate+ndeath); k++){
4775: if (k != i) {
4776: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4777: if(flatdir[jk]==1){
4778: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4779: }
4780: for(j=1; j <=ncovmodel; j++){
4781: printf("%12.7f ",p[jk]);
4782: jk++;
4783: }
4784: printf("\n");
4785: }
4786: }
4787: }
4788: /* skipping */
4789: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4790: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4791: for(k=1; k <=(nlstate+ndeath); k++){
4792: if (k != i) {
4793: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4794: if(flatdir[jk]==1){
4795: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4796: for(j=1; j <=ncovmodel; jk++,j++){
4797: printf(" p[%d]=%12.7f",jk, p[jk]);
4798: /*q[jjk]=p[jk];*/
4799: }
4800: }else{
4801: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4802: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4803: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4804: /*q[jjk]=p[jk];*/
4805: }
4806: }
4807: printf("\n");
4808: }
4809: fflush(stdout);
4810: }
4811: }
4812: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4813: #else /* FLATSUP */
1.126 brouard 4814: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4815: #endif /* FLATSUP */
4816:
4817: #ifdef LINMINORIGINAL
4818: #else
4819: free_ivector(flatdir,1,npar);
4820: #endif /* LINMINORIGINAL*/
4821: #endif /* POWELL */
1.126 brouard 4822:
1.162 brouard 4823: #ifdef NLOPT
4824: #ifdef NEWUOA
4825: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4826: #else
4827: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4828: #endif
4829: lb=vector(0,npar-1);
4830: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4831: nlopt_set_lower_bounds(opt, lb);
4832: nlopt_set_initial_step1(opt, 0.1);
4833:
4834: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4835: d->function = func;
4836: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4837: nlopt_set_min_objective(opt, myfunc, d);
4838: nlopt_set_xtol_rel(opt, ftol);
4839: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4840: printf("nlopt failed! %d\n",creturn);
4841: }
4842: else {
4843: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4844: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4845: iter=1; /* not equal */
4846: }
4847: nlopt_destroy(opt);
4848: #endif
1.319 brouard 4849: #ifdef FLATSUP
4850: /* npared = npar -flatd/ncovmodel; */
4851: /* xired= matrix(1,npared,1,npared); */
4852: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4853: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4854: /* free_matrix(xire,1,npared,1,npared); */
4855: #else /* FLATSUP */
4856: #endif /* FLATSUP */
1.126 brouard 4857: free_matrix(xi,1,npar,1,npar);
4858: fclose(ficrespow);
1.203 brouard 4859: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4860: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4861: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4862:
4863: }
4864:
4865: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4866: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4867: {
4868: double **a,**y,*x,pd;
1.203 brouard 4869: /* double **hess; */
1.164 brouard 4870: int i, j;
1.126 brouard 4871: int *indx;
4872:
4873: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4874: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4875: void lubksb(double **a, int npar, int *indx, double b[]) ;
4876: void ludcmp(double **a, int npar, int *indx, double *d) ;
4877: double gompertz(double p[]);
1.203 brouard 4878: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4879:
4880: printf("\nCalculation of the hessian matrix. Wait...\n");
4881: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4882: for (i=1;i<=npar;i++){
1.203 brouard 4883: printf("%d-",i);fflush(stdout);
4884: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4885:
4886: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4887:
4888: /* printf(" %f ",p[i]);
4889: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4890: }
4891:
4892: for (i=1;i<=npar;i++) {
4893: for (j=1;j<=npar;j++) {
4894: if (j>i) {
1.203 brouard 4895: printf(".%d-%d",i,j);fflush(stdout);
4896: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4897: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4898:
4899: hess[j][i]=hess[i][j];
4900: /*printf(" %lf ",hess[i][j]);*/
4901: }
4902: }
4903: }
4904: printf("\n");
4905: fprintf(ficlog,"\n");
4906:
4907: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4908: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4909:
4910: a=matrix(1,npar,1,npar);
4911: y=matrix(1,npar,1,npar);
4912: x=vector(1,npar);
4913: indx=ivector(1,npar);
4914: for (i=1;i<=npar;i++)
4915: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4916: ludcmp(a,npar,indx,&pd);
4917:
4918: for (j=1;j<=npar;j++) {
4919: for (i=1;i<=npar;i++) x[i]=0;
4920: x[j]=1;
4921: lubksb(a,npar,indx,x);
4922: for (i=1;i<=npar;i++){
4923: matcov[i][j]=x[i];
4924: }
4925: }
4926:
4927: printf("\n#Hessian matrix#\n");
4928: fprintf(ficlog,"\n#Hessian matrix#\n");
4929: for (i=1;i<=npar;i++) {
4930: for (j=1;j<=npar;j++) {
1.203 brouard 4931: printf("%.6e ",hess[i][j]);
4932: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4933: }
4934: printf("\n");
4935: fprintf(ficlog,"\n");
4936: }
4937:
1.203 brouard 4938: /* printf("\n#Covariance matrix#\n"); */
4939: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4940: /* for (i=1;i<=npar;i++) { */
4941: /* for (j=1;j<=npar;j++) { */
4942: /* printf("%.6e ",matcov[i][j]); */
4943: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4944: /* } */
4945: /* printf("\n"); */
4946: /* fprintf(ficlog,"\n"); */
4947: /* } */
4948:
1.126 brouard 4949: /* Recompute Inverse */
1.203 brouard 4950: /* for (i=1;i<=npar;i++) */
4951: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4952: /* ludcmp(a,npar,indx,&pd); */
4953:
4954: /* printf("\n#Hessian matrix recomputed#\n"); */
4955:
4956: /* for (j=1;j<=npar;j++) { */
4957: /* for (i=1;i<=npar;i++) x[i]=0; */
4958: /* x[j]=1; */
4959: /* lubksb(a,npar,indx,x); */
4960: /* for (i=1;i<=npar;i++){ */
4961: /* y[i][j]=x[i]; */
4962: /* printf("%.3e ",y[i][j]); */
4963: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4964: /* } */
4965: /* printf("\n"); */
4966: /* fprintf(ficlog,"\n"); */
4967: /* } */
4968:
4969: /* Verifying the inverse matrix */
4970: #ifdef DEBUGHESS
4971: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4972:
1.203 brouard 4973: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4974: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4975:
4976: for (j=1;j<=npar;j++) {
4977: for (i=1;i<=npar;i++){
1.203 brouard 4978: printf("%.2f ",y[i][j]);
4979: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4980: }
4981: printf("\n");
4982: fprintf(ficlog,"\n");
4983: }
1.203 brouard 4984: #endif
1.126 brouard 4985:
4986: free_matrix(a,1,npar,1,npar);
4987: free_matrix(y,1,npar,1,npar);
4988: free_vector(x,1,npar);
4989: free_ivector(indx,1,npar);
1.203 brouard 4990: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4991:
4992:
4993: }
4994:
4995: /*************** hessian matrix ****************/
4996: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4997: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4998: int i;
4999: int l=1, lmax=20;
1.203 brouard 5000: double k1,k2, res, fx;
1.132 brouard 5001: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 5002: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
5003: int k=0,kmax=10;
5004: double l1;
5005:
5006: fx=func(x);
5007: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 5008: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 5009: l1=pow(10,l);
5010: delts=delt;
5011: for(k=1 ; k <kmax; k=k+1){
5012: delt = delta*(l1*k);
5013: p2[theta]=x[theta] +delt;
1.145 brouard 5014: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 5015: p2[theta]=x[theta]-delt;
5016: k2=func(p2)-fx;
5017: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 5018: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 5019:
1.203 brouard 5020: #ifdef DEBUGHESSII
1.126 brouard 5021: 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);
5022: 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);
5023: #endif
5024: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
5025: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
5026: k=kmax;
5027: }
5028: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 5029: k=kmax; l=lmax*10;
1.126 brouard 5030: }
5031: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
5032: delts=delt;
5033: }
1.203 brouard 5034: } /* End loop k */
1.126 brouard 5035: }
5036: delti[theta]=delts;
5037: return res;
5038:
5039: }
5040:
1.203 brouard 5041: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 5042: {
5043: int i;
1.164 brouard 5044: int l=1, lmax=20;
1.126 brouard 5045: double k1,k2,k3,k4,res,fx;
1.132 brouard 5046: double p2[MAXPARM+1];
1.203 brouard 5047: int k, kmax=1;
5048: double v1, v2, cv12, lc1, lc2;
1.208 brouard 5049:
5050: int firstime=0;
1.203 brouard 5051:
1.126 brouard 5052: fx=func(x);
1.203 brouard 5053: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 5054: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 5055: p2[thetai]=x[thetai]+delti[thetai]*k;
5056: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 5057: k1=func(p2)-fx;
5058:
1.203 brouard 5059: p2[thetai]=x[thetai]+delti[thetai]*k;
5060: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 5061: k2=func(p2)-fx;
5062:
1.203 brouard 5063: p2[thetai]=x[thetai]-delti[thetai]*k;
5064: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 5065: k3=func(p2)-fx;
5066:
1.203 brouard 5067: p2[thetai]=x[thetai]-delti[thetai]*k;
5068: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 5069: k4=func(p2)-fx;
1.203 brouard 5070: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
5071: if(k1*k2*k3*k4 <0.){
1.208 brouard 5072: firstime=1;
1.203 brouard 5073: kmax=kmax+10;
1.208 brouard 5074: }
5075: if(kmax >=10 || firstime ==1){
1.246 brouard 5076: 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);
5077: 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 5078: 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);
5079: 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);
5080: }
5081: #ifdef DEBUGHESSIJ
5082: v1=hess[thetai][thetai];
5083: v2=hess[thetaj][thetaj];
5084: cv12=res;
5085: /* Computing eigen value of Hessian matrix */
5086: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
5087: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
5088: if ((lc2 <0) || (lc1 <0) ){
5089: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
5090: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
5091: 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);
5092: 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);
5093: }
1.126 brouard 5094: #endif
5095: }
5096: return res;
5097: }
5098:
1.203 brouard 5099: /* Not done yet: Was supposed to fix if not exactly at the maximum */
5100: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
5101: /* { */
5102: /* int i; */
5103: /* int l=1, lmax=20; */
5104: /* double k1,k2,k3,k4,res,fx; */
5105: /* double p2[MAXPARM+1]; */
5106: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
5107: /* int k=0,kmax=10; */
5108: /* double l1; */
5109:
5110: /* fx=func(x); */
5111: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
5112: /* l1=pow(10,l); */
5113: /* delts=delt; */
5114: /* for(k=1 ; k <kmax; k=k+1){ */
5115: /* delt = delti*(l1*k); */
5116: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
5117: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
5118: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
5119: /* k1=func(p2)-fx; */
5120:
5121: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
5122: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
5123: /* k2=func(p2)-fx; */
5124:
5125: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
5126: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
5127: /* k3=func(p2)-fx; */
5128:
5129: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
5130: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
5131: /* k4=func(p2)-fx; */
5132: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
5133: /* #ifdef DEBUGHESSIJ */
5134: /* 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); */
5135: /* 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); */
5136: /* #endif */
5137: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
5138: /* k=kmax; */
5139: /* } */
5140: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
5141: /* k=kmax; l=lmax*10; */
5142: /* } */
5143: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
5144: /* delts=delt; */
5145: /* } */
5146: /* } /\* End loop k *\/ */
5147: /* } */
5148: /* delti[theta]=delts; */
5149: /* return res; */
5150: /* } */
5151:
5152:
1.126 brouard 5153: /************** Inverse of matrix **************/
5154: void ludcmp(double **a, int n, int *indx, double *d)
5155: {
5156: int i,imax,j,k;
5157: double big,dum,sum,temp;
5158: double *vv;
5159:
5160: vv=vector(1,n);
5161: *d=1.0;
5162: for (i=1;i<=n;i++) {
5163: big=0.0;
5164: for (j=1;j<=n;j++)
5165: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 5166: if (big == 0.0){
5167: printf(" Singular Hessian matrix at row %d:\n",i);
5168: for (j=1;j<=n;j++) {
5169: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
5170: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
5171: }
5172: fflush(ficlog);
5173: fclose(ficlog);
5174: nrerror("Singular matrix in routine ludcmp");
5175: }
1.126 brouard 5176: vv[i]=1.0/big;
5177: }
5178: for (j=1;j<=n;j++) {
5179: for (i=1;i<j;i++) {
5180: sum=a[i][j];
5181: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
5182: a[i][j]=sum;
5183: }
5184: big=0.0;
5185: for (i=j;i<=n;i++) {
5186: sum=a[i][j];
5187: for (k=1;k<j;k++)
5188: sum -= a[i][k]*a[k][j];
5189: a[i][j]=sum;
5190: if ( (dum=vv[i]*fabs(sum)) >= big) {
5191: big=dum;
5192: imax=i;
5193: }
5194: }
5195: if (j != imax) {
5196: for (k=1;k<=n;k++) {
5197: dum=a[imax][k];
5198: a[imax][k]=a[j][k];
5199: a[j][k]=dum;
5200: }
5201: *d = -(*d);
5202: vv[imax]=vv[j];
5203: }
5204: indx[j]=imax;
5205: if (a[j][j] == 0.0) a[j][j]=TINY;
5206: if (j != n) {
5207: dum=1.0/(a[j][j]);
5208: for (i=j+1;i<=n;i++) a[i][j] *= dum;
5209: }
5210: }
5211: free_vector(vv,1,n); /* Doesn't work */
5212: ;
5213: }
5214:
5215: void lubksb(double **a, int n, int *indx, double b[])
5216: {
5217: int i,ii=0,ip,j;
5218: double sum;
5219:
5220: for (i=1;i<=n;i++) {
5221: ip=indx[i];
5222: sum=b[ip];
5223: b[ip]=b[i];
5224: if (ii)
5225: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
5226: else if (sum) ii=i;
5227: b[i]=sum;
5228: }
5229: for (i=n;i>=1;i--) {
5230: sum=b[i];
5231: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
5232: b[i]=sum/a[i][i];
5233: }
5234: }
5235:
5236: void pstamp(FILE *fichier)
5237: {
1.196 brouard 5238: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 5239: }
5240:
1.297 brouard 5241: void date2dmy(double date,double *day, double *month, double *year){
5242: double yp=0., yp1=0., yp2=0.;
5243:
5244: yp1=modf(date,&yp);/* extracts integral of date in yp and
5245: fractional in yp1 */
5246: *year=yp;
5247: yp2=modf((yp1*12),&yp);
5248: *month=yp;
5249: yp1=modf((yp2*30.5),&yp);
5250: *day=yp;
5251: if(*day==0) *day=1;
5252: if(*month==0) *month=1;
5253: }
5254:
1.253 brouard 5255:
5256:
1.126 brouard 5257: /************ Frequencies ********************/
1.251 brouard 5258: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 5259: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
5260: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 5261: { /* Some frequencies as well as proposing some starting values */
1.332 brouard 5262: /* Frequencies of any combination of dummy covariate used in the model equation */
1.265 brouard 5263: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 5264: int iind=0, iage=0;
5265: int mi; /* Effective wave */
5266: int first;
5267: double ***freq; /* Frequencies */
1.268 brouard 5268: 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 */
5269: 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 5270: double *meanq, *stdq, *idq;
1.226 brouard 5271: double **meanqt;
5272: double *pp, **prop, *posprop, *pospropt;
5273: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
5274: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
5275: double agebegin, ageend;
5276:
5277: pp=vector(1,nlstate);
1.251 brouard 5278: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5279: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
5280: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
5281: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
5282: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 5283: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 5284: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 5285: meanqt=matrix(1,lastpass,1,nqtveff);
5286: strcpy(fileresp,"P_");
5287: strcat(fileresp,fileresu);
5288: /*strcat(fileresphtm,fileresu);*/
5289: if((ficresp=fopen(fileresp,"w"))==NULL) {
5290: printf("Problem with prevalence resultfile: %s\n", fileresp);
5291: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
5292: exit(0);
5293: }
1.240 brouard 5294:
1.226 brouard 5295: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
5296: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
5297: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5298: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5299: fflush(ficlog);
5300: exit(70);
5301: }
5302: else{
5303: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 5304: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5305: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5306: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5307: }
1.319 brouard 5308: 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 5309:
1.226 brouard 5310: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
5311: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
5312: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5313: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5314: fflush(ficlog);
5315: exit(70);
1.240 brouard 5316: } else{
1.226 brouard 5317: 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 5318: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5319: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5320: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5321: }
1.319 brouard 5322: 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 5323:
1.253 brouard 5324: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
5325: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 5326: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5327: j1=0;
1.126 brouard 5328:
1.227 brouard 5329: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
1.335 brouard 5330: j=cptcoveff; /* Only simple dummy covariates used in the model */
1.330 brouard 5331: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 5332: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 5333:
5334:
1.226 brouard 5335: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
5336: reference=low_education V1=0,V2=0
5337: med_educ V1=1 V2=0,
5338: high_educ V1=0 V2=1
1.330 brouard 5339: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 5340: */
1.249 brouard 5341: dateintsum=0;
5342: k2cpt=0;
5343:
1.253 brouard 5344: if(cptcoveff == 0 )
1.265 brouard 5345: nl=1; /* Constant and age model only */
1.253 brouard 5346: else
5347: nl=2;
1.265 brouard 5348:
5349: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
5350: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.335 brouard 5351: * Loop on j1(1 to 2**cptcoveff) covariate combination
1.265 brouard 5352: * freq[s1][s2][iage] =0.
5353: * Loop on iind
5354: * ++freq[s1][s2][iage] weighted
5355: * end iind
5356: * if covariate and j!0
5357: * headers Variable on one line
5358: * endif cov j!=0
5359: * header of frequency table by age
5360: * Loop on age
5361: * pp[s1]+=freq[s1][s2][iage] weighted
5362: * pos+=freq[s1][s2][iage] weighted
5363: * Loop on s1 initial state
5364: * fprintf(ficresp
5365: * end s1
5366: * end age
5367: * if j!=0 computes starting values
5368: * end compute starting values
5369: * end j1
5370: * end nl
5371: */
1.253 brouard 5372: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
5373: if(nj==1)
5374: j=0; /* First pass for the constant */
1.265 brouard 5375: else{
1.335 brouard 5376: 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 5377: }
1.251 brouard 5378: first=1;
1.332 brouard 5379: 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 5380: posproptt=0.;
1.330 brouard 5381: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 5382: scanf("%d", i);*/
5383: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 5384: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 5385: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 5386: freq[i][s2][m]=0;
1.251 brouard 5387:
5388: for (i=1; i<=nlstate; i++) {
1.240 brouard 5389: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 5390: prop[i][m]=0;
5391: posprop[i]=0;
5392: pospropt[i]=0;
5393: }
1.283 brouard 5394: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 5395: idq[z1]=0.;
5396: meanq[z1]=0.;
5397: stdq[z1]=0.;
1.283 brouard 5398: }
5399: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 5400: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 5401: /* meanqt[m][z1]=0.; */
5402: /* } */
5403: /* } */
1.251 brouard 5404: /* dateintsum=0; */
5405: /* k2cpt=0; */
5406:
1.265 brouard 5407: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 5408: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
5409: bool=1;
5410: if(j !=0){
5411: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.335 brouard 5412: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
5413: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
1.251 brouard 5414: /* if(Tvaraff[z1] ==-20){ */
5415: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
5416: /* }else if(Tvaraff[z1] ==-10){ */
5417: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 brouard 5418: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
1.335 brouard 5419: /* 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); */
5420: if(Tvaraff[z1]<1 || Tvaraff[z1]>=NCOVMAX)
1.338 brouard 5421: printf("Error Tvaraff[z1]=%d<1 or >=%d, cptcoveff=%d model=1+age+%s\n",Tvaraff[z1],NCOVMAX, cptcoveff, model);
1.332 brouard 5422: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]){ /* for combination j1 of covariates */
1.265 brouard 5423: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 5424: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.332 brouard 5425: /* 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", */
5426: /* bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),*/
5427: /* j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
1.251 brouard 5428: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
5429: } /* Onlyf fixed */
5430: } /* end z1 */
1.335 brouard 5431: } /* cptcoveff > 0 */
1.251 brouard 5432: } /* end any */
5433: }/* end j==0 */
1.265 brouard 5434: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 5435: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 5436: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 5437: m=mw[mi][iind];
5438: if(j!=0){
5439: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.335 brouard 5440: for (z1=1; z1<=cptcoveff; z1++) {
1.251 brouard 5441: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 5442: /* iv= Tvar[Tmodelind[z1]]-ncovcol-nqv; /\* Good *\/ */
5443: iv= Tvar[Tmodelind[z1]]; /* Good *//* because cotvar starts now at first at ncovcol+nqv+ntv */
1.332 brouard 5444: 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 5445: value is -1, we don't select. It differs from the
5446: constant and age model which counts them. */
5447: bool=0; /* not selected */
5448: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.334 brouard 5449: /* i1=Tvaraff[z1]; */
5450: /* i2=TnsdVar[i1]; */
5451: /* i3=nbcode[i1][i2]; */
5452: /* i4=covar[i1][iind]; */
5453: /* if(i4 != i3){ */
5454: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) { /* Bug valgrind */
1.251 brouard 5455: bool=0;
5456: }
5457: }
5458: }
5459: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
5460: } /* end j==0 */
5461: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 5462: if(bool==1){ /*Selected */
1.251 brouard 5463: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
5464: and mw[mi+1][iind]. dh depends on stepm. */
5465: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
5466: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
5467: if(m >=firstpass && m <=lastpass){
5468: k2=anint[m][iind]+(mint[m][iind]/12.);
5469: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
5470: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
5471: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
5472: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
5473: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
5474: if (m<lastpass) {
5475: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
5476: /* 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]); */
5477: if(s[m][iind]==-1)
5478: 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.));
5479: 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 5480: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
5481: if(!isnan(covar[ncovcol+z1][iind])){
1.332 brouard 5482: idq[z1]=idq[z1]+weight[iind];
5483: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
5484: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
5485: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
1.311 brouard 5486: }
1.284 brouard 5487: }
1.251 brouard 5488: /* if((int)agev[m][iind] == 55) */
5489: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
5490: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
5491: 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 5492: }
1.251 brouard 5493: } /* end if between passes */
5494: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
5495: dateintsum=dateintsum+k2; /* on all covariates ?*/
5496: k2cpt++;
5497: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 5498: }
1.251 brouard 5499: }else{
5500: bool=1;
5501: }/* end bool 2 */
5502: } /* end m */
1.284 brouard 5503: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
5504: /* idq[z1]=idq[z1]+weight[iind]; */
5505: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
5506: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
5507: /* } */
1.251 brouard 5508: } /* end bool */
5509: } /* end iind = 1 to imx */
1.319 brouard 5510: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 5511: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
5512:
5513:
5514: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.335 brouard 5515: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
1.265 brouard 5516: pstamp(ficresp);
1.335 brouard 5517: if (cptcoveff>0 && j!=0){
1.265 brouard 5518: pstamp(ficresp);
1.251 brouard 5519: printf( "\n#********** Variable ");
5520: fprintf(ficresp, "\n#********** Variable ");
5521: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
5522: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
5523: fprintf(ficlog, "\n#********** Variable ");
1.340 brouard 5524: for (z1=1; z1<=cptcoveff; z1++){
1.251 brouard 5525: if(!FixedV[Tvaraff[z1]]){
1.330 brouard 5526: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5527: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5528: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5529: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5530: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 5531: }else{
1.330 brouard 5532: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5533: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5534: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5535: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5536: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 5537: }
5538: }
5539: printf( "**********\n#");
5540: fprintf(ficresp, "**********\n#");
5541: fprintf(ficresphtm, "**********</h3>\n");
5542: fprintf(ficresphtmfr, "**********</h3>\n");
5543: fprintf(ficlog, "**********\n");
5544: }
1.284 brouard 5545: /*
5546: Printing means of quantitative variables if any
5547: */
5548: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5549: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5550: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5551: if(weightopt==1){
5552: printf(" Weighted mean and standard deviation of");
5553: fprintf(ficlog," Weighted mean and standard deviation of");
5554: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5555: }
1.311 brouard 5556: /* mu = \frac{w x}{\sum w}
5557: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5558: */
5559: 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]));
5560: 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]));
5561: 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 5562: }
5563: /* for (z1=1; z1<= nqtveff; z1++) { */
5564: /* for(m=1;m<=lastpass;m++){ */
5565: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5566: /* } */
5567: /* } */
1.283 brouard 5568:
1.251 brouard 5569: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.335 brouard 5570: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 5571: fprintf(ficresp, " Age");
1.335 brouard 5572: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) {
5573: 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]]);
5574: fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5575: }
1.251 brouard 5576: for(i=1; i<=nlstate;i++) {
1.335 brouard 5577: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5578: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5579: }
1.335 brouard 5580: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5581: fprintf(ficresphtm, "\n");
5582:
5583: /* Header of frequency table by age */
5584: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5585: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5586: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5587: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5588: if(s2!=0 && m!=0)
5589: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5590: }
1.226 brouard 5591: }
1.251 brouard 5592: fprintf(ficresphtmfr, "\n");
5593:
5594: /* For each age */
5595: for(iage=iagemin; iage <= iagemax+3; iage++){
5596: fprintf(ficresphtm,"<tr>");
5597: if(iage==iagemax+1){
5598: fprintf(ficlog,"1");
5599: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5600: }else if(iage==iagemax+2){
5601: fprintf(ficlog,"0");
5602: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5603: }else if(iage==iagemax+3){
5604: fprintf(ficlog,"Total");
5605: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5606: }else{
1.240 brouard 5607: if(first==1){
1.251 brouard 5608: first=0;
5609: printf("See log file for details...\n");
5610: }
5611: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5612: fprintf(ficlog,"Age %d", iage);
5613: }
1.265 brouard 5614: for(s1=1; s1 <=nlstate ; s1++){
5615: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5616: pp[s1] += freq[s1][m][iage];
1.251 brouard 5617: }
1.265 brouard 5618: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5619: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5620: pos += freq[s1][m][iage];
5621: if(pp[s1]>=1.e-10){
1.251 brouard 5622: if(first==1){
1.265 brouard 5623: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5624: }
1.265 brouard 5625: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5626: }else{
5627: if(first==1)
1.265 brouard 5628: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5629: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5630: }
5631: }
5632:
1.265 brouard 5633: for(s1=1; s1 <=nlstate ; s1++){
5634: /* posprop[s1]=0; */
5635: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5636: pp[s1] += freq[s1][m][iage];
5637: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5638:
5639: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5640: pos += pp[s1]; /* pos is the total number of transitions until this age */
5641: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5642: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5643: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5644: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5645: }
5646:
5647: /* Writing ficresp */
1.335 brouard 5648: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5649: if( iage <= iagemax){
5650: fprintf(ficresp," %d",iage);
5651: }
5652: }else if( nj==2){
5653: if( iage <= iagemax){
5654: fprintf(ficresp," %d",iage);
1.335 brouard 5655: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.265 brouard 5656: }
1.240 brouard 5657: }
1.265 brouard 5658: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5659: if(pos>=1.e-5){
1.251 brouard 5660: if(first==1)
1.265 brouard 5661: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5662: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5663: }else{
5664: if(first==1)
1.265 brouard 5665: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5666: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5667: }
5668: if( iage <= iagemax){
5669: if(pos>=1.e-5){
1.335 brouard 5670: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5671: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5672: }else if( nj==2){
5673: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5674: }
5675: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5676: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5677: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5678: } else{
1.335 brouard 5679: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 5680: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5681: }
1.240 brouard 5682: }
1.265 brouard 5683: pospropt[s1] +=posprop[s1];
5684: } /* end loop s1 */
1.251 brouard 5685: /* pospropt=0.; */
1.265 brouard 5686: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5687: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5688: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5689: if(first==1){
1.265 brouard 5690: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5691: }
1.265 brouard 5692: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5693: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5694: }
1.265 brouard 5695: if(s1!=0 && m!=0)
5696: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5697: }
1.265 brouard 5698: } /* end loop s1 */
1.251 brouard 5699: posproptt=0.;
1.265 brouard 5700: for(s1=1; s1 <=nlstate; s1++){
5701: posproptt += pospropt[s1];
1.251 brouard 5702: }
5703: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5704: fprintf(ficresphtm,"</tr>\n");
1.335 brouard 5705: if((cptcoveff==0 && nj==1)|| nj==2 ) {
1.265 brouard 5706: if(iage <= iagemax)
5707: fprintf(ficresp,"\n");
1.240 brouard 5708: }
1.251 brouard 5709: if(first==1)
5710: printf("Others in log...\n");
5711: fprintf(ficlog,"\n");
5712: } /* end loop age iage */
1.265 brouard 5713:
1.251 brouard 5714: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5715: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5716: if(posproptt < 1.e-5){
1.265 brouard 5717: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5718: }else{
1.265 brouard 5719: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5720: }
1.226 brouard 5721: }
1.251 brouard 5722: fprintf(ficresphtm,"</tr>\n");
5723: fprintf(ficresphtm,"</table>\n");
5724: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5725: if(posproptt < 1.e-5){
1.251 brouard 5726: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5727: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5728: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5729: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5730: invalidvarcomb[j1]=1;
1.226 brouard 5731: }else{
1.338 brouard 5732: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced (or no resultline).</p>",j1);
1.251 brouard 5733: invalidvarcomb[j1]=0;
1.226 brouard 5734: }
1.251 brouard 5735: fprintf(ficresphtmfr,"</table>\n");
5736: fprintf(ficlog,"\n");
5737: if(j!=0){
5738: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5739: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5740: for(k=1; k <=(nlstate+ndeath); k++){
5741: if (k != i) {
1.265 brouard 5742: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5743: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5744: if(j1==1){ /* All dummy covariates to zero */
5745: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5746: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5747: printf("%d%d ",i,k);
5748: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5749: 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]));
5750: 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]));
5751: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5752: }
1.253 brouard 5753: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5754: for(iage=iagemin; iage <= iagemax+3; iage++){
5755: x[iage]= (double)iage;
5756: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5757: /* 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 5758: }
1.268 brouard 5759: /* Some are not finite, but linreg will ignore these ages */
5760: no=0;
1.253 brouard 5761: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5762: pstart[s1]=b;
5763: pstart[s1-1]=a;
1.252 brouard 5764: }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 */
5765: 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]);
5766: 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 5767: 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 5768: printf("%d%d ",i,k);
5769: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5770: 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 5771: }else{ /* Other cases, like quantitative fixed or varying covariates */
5772: ;
5773: }
5774: /* printf("%12.7f )", param[i][jj][k]); */
5775: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5776: s1++;
1.251 brouard 5777: } /* end jj */
5778: } /* end k!= i */
5779: } /* end k */
1.265 brouard 5780: } /* end i, s1 */
1.251 brouard 5781: } /* end j !=0 */
5782: } /* end selected combination of covariate j1 */
5783: if(j==0){ /* We can estimate starting values from the occurences in each case */
5784: printf("#Freqsummary: Starting values for the constants:\n");
5785: fprintf(ficlog,"\n");
1.265 brouard 5786: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5787: for(k=1; k <=(nlstate+ndeath); k++){
5788: if (k != i) {
5789: printf("%d%d ",i,k);
5790: fprintf(ficlog,"%d%d ",i,k);
5791: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5792: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5793: if(jj==1){ /* Age has to be done */
1.265 brouard 5794: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5795: 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]));
5796: 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 5797: }
5798: /* printf("%12.7f )", param[i][jj][k]); */
5799: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5800: s1++;
1.250 brouard 5801: }
1.251 brouard 5802: printf("\n");
5803: fprintf(ficlog,"\n");
1.250 brouard 5804: }
5805: }
1.284 brouard 5806: } /* end of state i */
1.251 brouard 5807: printf("#Freqsummary\n");
5808: fprintf(ficlog,"\n");
1.265 brouard 5809: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5810: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5811: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5812: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5813: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5814: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5815: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5816: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5817: /* } */
5818: }
1.265 brouard 5819: } /* end loop s1 */
1.251 brouard 5820:
5821: printf("\n");
5822: fprintf(ficlog,"\n");
5823: } /* end j=0 */
1.249 brouard 5824: } /* end j */
1.252 brouard 5825:
1.253 brouard 5826: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5827: for(i=1, jk=1; i <=nlstate; i++){
5828: for(j=1; j <=nlstate+ndeath; j++){
5829: if(j!=i){
5830: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5831: printf("%1d%1d",i,j);
5832: fprintf(ficparo,"%1d%1d",i,j);
5833: for(k=1; k<=ncovmodel;k++){
5834: /* printf(" %lf",param[i][j][k]); */
5835: /* fprintf(ficparo," %lf",param[i][j][k]); */
5836: p[jk]=pstart[jk];
5837: printf(" %f ",pstart[jk]);
5838: fprintf(ficparo," %f ",pstart[jk]);
5839: jk++;
5840: }
5841: printf("\n");
5842: fprintf(ficparo,"\n");
5843: }
5844: }
5845: }
5846: } /* end mle=-2 */
1.226 brouard 5847: dateintmean=dateintsum/k2cpt;
1.296 brouard 5848: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5849:
1.226 brouard 5850: fclose(ficresp);
5851: fclose(ficresphtm);
5852: fclose(ficresphtmfr);
1.283 brouard 5853: free_vector(idq,1,nqfveff);
1.226 brouard 5854: free_vector(meanq,1,nqfveff);
1.284 brouard 5855: free_vector(stdq,1,nqfveff);
1.226 brouard 5856: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5857: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5858: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5859: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5860: free_vector(pospropt,1,nlstate);
5861: free_vector(posprop,1,nlstate);
1.251 brouard 5862: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5863: free_vector(pp,1,nlstate);
5864: /* End of freqsummary */
5865: }
1.126 brouard 5866:
1.268 brouard 5867: /* Simple linear regression */
5868: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5869:
5870: /* y=a+bx regression */
5871: double sumx = 0.0; /* sum of x */
5872: double sumx2 = 0.0; /* sum of x**2 */
5873: double sumxy = 0.0; /* sum of x * y */
5874: double sumy = 0.0; /* sum of y */
5875: double sumy2 = 0.0; /* sum of y**2 */
5876: double sume2 = 0.0; /* sum of square or residuals */
5877: double yhat;
5878:
5879: double denom=0;
5880: int i;
5881: int ne=*no;
5882:
5883: for ( i=ifi, ne=0;i<=ila;i++) {
5884: if(!isfinite(x[i]) || !isfinite(y[i])){
5885: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5886: continue;
5887: }
5888: ne=ne+1;
5889: sumx += x[i];
5890: sumx2 += x[i]*x[i];
5891: sumxy += x[i] * y[i];
5892: sumy += y[i];
5893: sumy2 += y[i]*y[i];
5894: denom = (ne * sumx2 - sumx*sumx);
5895: /* 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); */
5896: }
5897:
5898: denom = (ne * sumx2 - sumx*sumx);
5899: if (denom == 0) {
5900: // vertical, slope m is infinity
5901: *b = INFINITY;
5902: *a = 0;
5903: if (r) *r = 0;
5904: return 1;
5905: }
5906:
5907: *b = (ne * sumxy - sumx * sumy) / denom;
5908: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5909: if (r!=NULL) {
5910: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5911: sqrt((sumx2 - sumx*sumx/ne) *
5912: (sumy2 - sumy*sumy/ne));
5913: }
5914: *no=ne;
5915: for ( i=ifi, ne=0;i<=ila;i++) {
5916: if(!isfinite(x[i]) || !isfinite(y[i])){
5917: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5918: continue;
5919: }
5920: ne=ne+1;
5921: yhat = y[i] - *a -*b* x[i];
5922: sume2 += yhat * yhat ;
5923:
5924: denom = (ne * sumx2 - sumx*sumx);
5925: /* 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); */
5926: }
5927: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5928: *sa= *sb * sqrt(sumx2/ne);
5929:
5930: return 0;
5931: }
5932:
1.126 brouard 5933: /************ Prevalence ********************/
1.227 brouard 5934: 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)
5935: {
5936: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5937: in each health status at the date of interview (if between dateprev1 and dateprev2).
5938: We still use firstpass and lastpass as another selection.
5939: */
1.126 brouard 5940:
1.227 brouard 5941: int i, m, jk, j1, bool, z1,j, iv;
5942: int mi; /* Effective wave */
5943: int iage;
5944: double agebegin, ageend;
5945:
5946: double **prop;
5947: double posprop;
5948: double y2; /* in fractional years */
5949: int iagemin, iagemax;
5950: int first; /** to stop verbosity which is redirected to log file */
5951:
5952: iagemin= (int) agemin;
5953: iagemax= (int) agemax;
5954: /*pp=vector(1,nlstate);*/
1.251 brouard 5955: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5956: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5957: j1=0;
1.222 brouard 5958:
1.227 brouard 5959: /*j=cptcoveff;*/
5960: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5961:
1.288 brouard 5962: first=0;
1.335 brouard 5963: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of simple dummy covariates */
1.227 brouard 5964: for (i=1; i<=nlstate; i++)
1.251 brouard 5965: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5966: prop[i][iage]=0.0;
5967: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5968: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5969: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5970:
5971: for (i=1; i<=imx; i++) { /* Each individual */
5972: bool=1;
5973: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5974: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5975: m=mw[mi][i];
5976: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5977: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5978: for (z1=1; z1<=cptcoveff; z1++){
5979: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 5980: iv= Tvar[Tmodelind[z1]];/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.332 brouard 5981: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality */
1.227 brouard 5982: bool=0;
5983: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.332 brouard 5984: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.227 brouard 5985: bool=0;
5986: }
5987: }
5988: if(bool==1){ /* Otherwise we skip that wave/person */
5989: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5990: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5991: if(m >=firstpass && m <=lastpass){
5992: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5993: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5994: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5995: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5996: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5997: 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);
5998: exit(1);
5999: }
6000: if (s[m][i]>0 && s[m][i]<=nlstate) {
6001: /*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]]);*/
6002: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
6003: prop[s[m][i]][iagemax+3] += weight[i];
6004: } /* end valid statuses */
6005: } /* end selection of dates */
6006: } /* end selection of waves */
6007: } /* end bool */
6008: } /* end wave */
6009: } /* end individual */
6010: for(i=iagemin; i <= iagemax+3; i++){
6011: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
6012: posprop += prop[jk][i];
6013: }
6014:
6015: for(jk=1; jk <=nlstate ; jk++){
6016: if( i <= iagemax){
6017: if(posprop>=1.e-5){
6018: probs[i][jk][j1]= prop[jk][i]/posprop;
6019: } else{
1.288 brouard 6020: if(!first){
6021: first=1;
1.266 brouard 6022: 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]);
6023: }else{
1.288 brouard 6024: 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 6025: }
6026: }
6027: }
6028: }/* end jk */
6029: }/* end i */
1.222 brouard 6030: /*} *//* end i1 */
1.227 brouard 6031: } /* end j1 */
1.222 brouard 6032:
1.227 brouard 6033: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
6034: /*free_vector(pp,1,nlstate);*/
1.251 brouard 6035: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 6036: } /* End of prevalence */
1.126 brouard 6037:
6038: /************* Waves Concatenation ***************/
6039:
6040: 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)
6041: {
1.298 brouard 6042: /* 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 6043: Death is a valid wave (if date is known).
6044: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
6045: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 6046: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 6047: */
1.126 brouard 6048:
1.224 brouard 6049: int i=0, mi=0, m=0, mli=0;
1.126 brouard 6050: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
6051: double sum=0., jmean=0.;*/
1.224 brouard 6052: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 6053: int j, k=0,jk, ju, jl;
6054: double sum=0.;
6055: first=0;
1.214 brouard 6056: firstwo=0;
1.217 brouard 6057: firsthree=0;
1.218 brouard 6058: firstfour=0;
1.164 brouard 6059: jmin=100000;
1.126 brouard 6060: jmax=-1;
6061: jmean=0.;
1.224 brouard 6062:
6063: /* Treating live states */
1.214 brouard 6064: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 6065: mi=0; /* First valid wave */
1.227 brouard 6066: mli=0; /* Last valid wave */
1.309 brouard 6067: m=firstpass; /* Loop on waves */
6068: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 6069: 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 */
6070: mli=m-1;/* mw[++mi][i]=m-1; */
6071: }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 6072: 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 6073: mli=m;
1.224 brouard 6074: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
6075: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 6076: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 6077: }
1.309 brouard 6078: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 6079: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 6080: break;
1.224 brouard 6081: #else
1.317 brouard 6082: 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 6083: if(firsthree == 0){
1.302 brouard 6084: 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 6085: firsthree=1;
1.317 brouard 6086: }else if(firsthree >=1 && firsthree < 10){
6087: 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);
6088: firsthree++;
6089: }else if(firsthree == 10){
6090: printf("Information, too many Information flags: no more reported to log either\n");
6091: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
6092: firsthree++;
6093: }else{
6094: firsthree++;
1.227 brouard 6095: }
1.309 brouard 6096: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 6097: mli=m;
6098: }
6099: if(s[m][i]==-2){ /* Vital status is really unknown */
6100: nbwarn++;
1.309 brouard 6101: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 6102: 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);
6103: 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);
6104: }
6105: break;
6106: }
6107: break;
1.224 brouard 6108: #endif
1.227 brouard 6109: }/* End m >= lastpass */
1.126 brouard 6110: }/* end while */
1.224 brouard 6111:
1.227 brouard 6112: /* 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 6113: /* After last pass */
1.224 brouard 6114: /* Treating death states */
1.214 brouard 6115: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 6116: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
6117: /* } */
1.126 brouard 6118: mi++; /* Death is another wave */
6119: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 6120: /* Only death is a correct wave */
1.126 brouard 6121: mw[mi][i]=m;
1.257 brouard 6122: } /* else not in a death state */
1.224 brouard 6123: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 6124: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 6125: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 6126: 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 6127: nbwarn++;
6128: if(firstfiv==0){
1.309 brouard 6129: 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 6130: firstfiv=1;
6131: }else{
1.309 brouard 6132: 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 6133: }
1.309 brouard 6134: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
6135: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 6136: nberr++;
6137: if(firstwo==0){
1.309 brouard 6138: 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 6139: firstwo=1;
6140: }
1.309 brouard 6141: 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 6142: }
1.257 brouard 6143: }else{ /* if date of interview is unknown */
1.227 brouard 6144: /* death is known but not confirmed by death status at any wave */
6145: if(firstfour==0){
1.309 brouard 6146: 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 6147: firstfour=1;
6148: }
1.309 brouard 6149: 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 6150: }
1.224 brouard 6151: } /* end if date of death is known */
6152: #endif
1.309 brouard 6153: wav[i]=mi; /* mi should be the last effective wave (or mli), */
6154: /* wav[i]=mw[mi][i]; */
1.126 brouard 6155: if(mi==0){
6156: nbwarn++;
6157: if(first==0){
1.227 brouard 6158: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
6159: first=1;
1.126 brouard 6160: }
6161: if(first==1){
1.227 brouard 6162: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 6163: }
6164: } /* end mi==0 */
6165: } /* End individuals */
1.214 brouard 6166: /* wav and mw are no more changed */
1.223 brouard 6167:
1.317 brouard 6168: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
6169: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
6170:
6171:
1.126 brouard 6172: for(i=1; i<=imx; i++){
6173: for(mi=1; mi<wav[i];mi++){
6174: if (stepm <=0)
1.227 brouard 6175: dh[mi][i]=1;
1.126 brouard 6176: else{
1.260 brouard 6177: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 6178: if (agedc[i] < 2*AGESUP) {
6179: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
6180: if(j==0) j=1; /* Survives at least one month after exam */
6181: else if(j<0){
6182: nberr++;
6183: printf("Error! Negative delay (%d to death) between waves %d and %d of individual %ld at 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]);
6184: j=1; /* Temporary Dangerous patch */
6185: 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);
6186: fprintf(ficlog,"Error! Negative delay (%d to death) between waves %d and %d of individual %ld at 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]);
6187: 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);
6188: }
6189: k=k+1;
6190: if (j >= jmax){
6191: jmax=j;
6192: ijmax=i;
6193: }
6194: if (j <= jmin){
6195: jmin=j;
6196: ijmin=i;
6197: }
6198: sum=sum+j;
6199: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
6200: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
6201: }
6202: }
6203: else{
6204: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 6205: /* 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 6206:
1.227 brouard 6207: k=k+1;
6208: if (j >= jmax) {
6209: jmax=j;
6210: ijmax=i;
6211: }
6212: else if (j <= jmin){
6213: jmin=j;
6214: ijmin=i;
6215: }
6216: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
6217: /*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]);*/
6218: if(j<0){
6219: nberr++;
6220: printf("Error! Negative delay (%d) between waves %d and %d of individual %ld at 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]);
6221: fprintf(ficlog,"Error! Negative delay (%d) between waves %d and %d of individual %ld at 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]);
6222: }
6223: sum=sum+j;
6224: }
6225: jk= j/stepm;
6226: jl= j -jk*stepm;
6227: ju= j -(jk+1)*stepm;
6228: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
6229: if(jl==0){
6230: dh[mi][i]=jk;
6231: bh[mi][i]=0;
6232: }else{ /* We want a negative bias in order to only have interpolation ie
6233: * to avoid the price of an extra matrix product in likelihood */
6234: dh[mi][i]=jk+1;
6235: bh[mi][i]=ju;
6236: }
6237: }else{
6238: if(jl <= -ju){
6239: dh[mi][i]=jk;
6240: bh[mi][i]=jl; /* bias is positive if real duration
6241: * is higher than the multiple of stepm and negative otherwise.
6242: */
6243: }
6244: else{
6245: dh[mi][i]=jk+1;
6246: bh[mi][i]=ju;
6247: }
6248: if(dh[mi][i]==0){
6249: dh[mi][i]=1; /* At least one step */
6250: bh[mi][i]=ju; /* At least one step */
6251: /* 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);*/
6252: }
6253: } /* end if mle */
1.126 brouard 6254: }
6255: } /* end wave */
6256: }
6257: jmean=sum/k;
6258: 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 6259: 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 6260: }
1.126 brouard 6261:
6262: /*********** Tricode ****************************/
1.220 brouard 6263: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 6264: {
6265: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
6266: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
6267: * Boring subroutine which should only output nbcode[Tvar[j]][k]
6268: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
6269: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
6270: */
1.130 brouard 6271:
1.242 brouard 6272: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
6273: int modmaxcovj=0; /* Modality max of covariates j */
6274: int cptcode=0; /* Modality max of covariates j */
6275: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 6276:
6277:
1.242 brouard 6278: /* cptcoveff=0; */
6279: /* *cptcov=0; */
1.126 brouard 6280:
1.242 brouard 6281: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 6282: for (k=1; k <= maxncov; k++)
6283: for(j=1; j<=2; j++)
6284: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 6285:
1.242 brouard 6286: /* Loop on covariates without age and products and no quantitative variable */
1.335 brouard 6287: 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 6288: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
1.343 brouard 6289: /* printf("Testing k=%d, cptcovt=%d\n",k, cptcovt); */
1.339 brouard 6290: if(Dummy[k]==0 && Typevar[k] !=1 && Typevar[k] != 2){ /* Dummy covariate and not age product nor fixed product */
1.242 brouard 6291: switch(Fixed[k]) {
6292: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 6293: modmaxcovj=0;
6294: modmincovj=0;
1.242 brouard 6295: 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 6296: /* 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 6297: ij=(int)(covar[Tvar[k]][i]);
6298: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
6299: * If product of Vn*Vm, still boolean *:
6300: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
6301: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
6302: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
6303: modality of the nth covariate of individual i. */
6304: if (ij > modmaxcovj)
6305: modmaxcovj=ij;
6306: else if (ij < modmincovj)
6307: modmincovj=ij;
1.287 brouard 6308: if (ij <0 || ij >1 ){
1.311 brouard 6309: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6310: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6311: fflush(ficlog);
6312: exit(1);
1.287 brouard 6313: }
6314: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 6315: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
6316: exit(1);
6317: }else
6318: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
6319: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
6320: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
6321: /* getting the maximum value of the modality of the covariate
6322: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
6323: female ies 1, then modmaxcovj=1.
6324: */
6325: } /* end for loop on individuals i */
6326: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6327: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6328: cptcode=modmaxcovj;
6329: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
6330: /*for (i=0; i<=cptcode; i++) {*/
6331: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
6332: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6333: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6334: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
6335: if( j != -1){
6336: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
6337: covariate for which somebody answered excluding
6338: undefined. Usually 2: 0 and 1. */
6339: }
6340: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
6341: covariate for which somebody answered including
6342: undefined. Usually 3: -1, 0 and 1. */
6343: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
6344: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
6345: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 6346:
1.242 brouard 6347: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
6348: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
6349: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
6350: /* modmincovj=3; modmaxcovj = 7; */
6351: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
6352: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
6353: /* defining two dummy variables: variables V1_1 and V1_2.*/
6354: /* nbcode[Tvar[j]][ij]=k; */
6355: /* nbcode[Tvar[j]][1]=0; */
6356: /* nbcode[Tvar[j]][2]=1; */
6357: /* nbcode[Tvar[j]][3]=2; */
6358: /* To be continued (not working yet). */
6359: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 6360:
6361: /* 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*/
6362: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
6363: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
6364: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
6365: /*, could be restored in the future */
6366: 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 6367: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
6368: break;
6369: }
6370: ij++;
1.287 brouard 6371: 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 6372: cptcode = ij; /* New max modality for covar j */
6373: } /* end of loop on modality i=-1 to 1 or more */
6374: break;
6375: case 1: /* Testing on varying covariate, could be simple and
6376: * should look at waves or product of fixed *
6377: * varying. No time to test -1, assuming 0 and 1 only */
6378: ij=0;
6379: for(i=0; i<=1;i++){
6380: nbcode[Tvar[k]][++ij]=i;
6381: }
6382: break;
6383: default:
6384: break;
6385: } /* end switch */
6386: } /* end dummy test */
1.342 brouard 6387: if(Dummy[k]==1 && Typevar[k] !=1 && Fixed ==0){ /* Fixed Quantitative covariate and not age product */
1.311 brouard 6388: 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 6389: if(Tvar[k]<=0 || Tvar[k]>=NCOVMAX){
6390: printf("Error k=%d \n",k);
6391: exit(1);
6392: }
1.311 brouard 6393: if(isnan(covar[Tvar[k]][i])){
6394: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6395: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6396: fflush(ficlog);
6397: exit(1);
6398: }
6399: }
1.335 brouard 6400: } /* end Quanti */
1.287 brouard 6401: } /* 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 6402:
6403: for (k=-1; k< maxncov; k++) Ndum[k]=0;
6404: /* Look at fixed dummy (single or product) covariates to check empty modalities */
6405: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
6406: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
6407: 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 */
6408: 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 */
6409: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
6410: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
6411:
6412: ij=0;
6413: /* 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 6414: 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 */
6415: /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
1.242 brouard 6416: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
6417: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
1.335 brouard 6418: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy simple and non empty in the model */
6419: /* Typevar[k] =0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
6420: /* 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 6421: /* If product not in single variable we don't print results */
6422: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
1.335 brouard 6423: ++ij;/* V5 + V4 + V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V1, *//* V5 quanti, V2 quanti, V4, V3, V1 dummies */
6424: /* k= 1 2 3 4 5 6 7 8 9 */
6425: /* Tvar[k]= 5 4 3 6 5 2 7 1 1 */
6426: /* ij 1 2 3 */
6427: /* Tvaraff[ij]= 4 3 1 */
6428: /* Tmodelind[ij]=2 3 9 */
6429: /* TmodelInvind[ij]=2 1 1 */
1.242 brouard 6430: 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*/
6431: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
6432: 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 */
6433: if(Fixed[k]!=0)
6434: anyvaryingduminmodel=1;
6435: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
6436: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
6437: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
6438: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
6439: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
6440: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
6441: }
6442: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
6443: /* ij--; */
6444: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.335 brouard 6445: *cptcov=ij; /* cptcov= Number of total real effective simple dummies (fixed or time arying) effective (used as cptcoveff in other functions)
1.242 brouard 6446: * because they can be excluded from the model and real
6447: * if in the model but excluded because missing values, but how to get k from ij?*/
6448: for(j=ij+1; j<= cptcovt; j++){
6449: Tvaraff[j]=0;
6450: Tmodelind[j]=0;
6451: }
6452: for(j=ntveff+1; j<= cptcovt; j++){
6453: TmodelInvind[j]=0;
6454: }
6455: /* To be sorted */
6456: ;
6457: }
1.126 brouard 6458:
1.145 brouard 6459:
1.126 brouard 6460: /*********** Health Expectancies ****************/
6461:
1.235 brouard 6462: 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 6463:
6464: {
6465: /* Health expectancies, no variances */
1.329 brouard 6466: /* cij is the combination in the list of combination of dummy covariates */
6467: /* strstart is a string of time at start of computing */
1.164 brouard 6468: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 6469: int nhstepma, nstepma; /* Decreasing with age */
6470: double age, agelim, hf;
6471: double ***p3mat;
6472: double eip;
6473:
1.238 brouard 6474: /* pstamp(ficreseij); */
1.126 brouard 6475: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
6476: fprintf(ficreseij,"# Age");
6477: for(i=1; i<=nlstate;i++){
6478: for(j=1; j<=nlstate;j++){
6479: fprintf(ficreseij," e%1d%1d ",i,j);
6480: }
6481: fprintf(ficreseij," e%1d. ",i);
6482: }
6483: fprintf(ficreseij,"\n");
6484:
6485:
6486: if(estepm < stepm){
6487: printf ("Problem %d lower than %d\n",estepm, stepm);
6488: }
6489: else hstepm=estepm;
6490: /* We compute the life expectancy from trapezoids spaced every estepm months
6491: * This is mainly to measure the difference between two models: for example
6492: * if stepm=24 months pijx are given only every 2 years and by summing them
6493: * we are calculating an estimate of the Life Expectancy assuming a linear
6494: * progression in between and thus overestimating or underestimating according
6495: * to the curvature of the survival function. If, for the same date, we
6496: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6497: * to compare the new estimate of Life expectancy with the same linear
6498: * hypothesis. A more precise result, taking into account a more precise
6499: * curvature will be obtained if estepm is as small as stepm. */
6500:
6501: /* For example we decided to compute the life expectancy with the smallest unit */
6502: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6503: nhstepm is the number of hstepm from age to agelim
6504: nstepm is the number of stepm from age to agelin.
1.270 brouard 6505: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 6506: and note for a fixed period like estepm months */
6507: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6508: survival function given by stepm (the optimization length). Unfortunately it
6509: means that if the survival funtion is printed only each two years of age and if
6510: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6511: results. So we changed our mind and took the option of the best precision.
6512: */
6513: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6514:
6515: agelim=AGESUP;
6516: /* If stepm=6 months */
6517: /* Computed by stepm unit matrices, product of hstepm matrices, stored
6518: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
6519:
6520: /* nhstepm age range expressed in number of stepm */
6521: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6522: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6523: /* if (stepm >= YEARM) hstepm=1;*/
6524: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6525: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6526:
6527: for (age=bage; age<=fage; age ++){
6528: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6529: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6530: /* if (stepm >= YEARM) hstepm=1;*/
6531: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
6532:
6533: /* If stepm=6 months */
6534: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6535: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 brouard 6536: /* 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 6537: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 6538:
6539: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6540:
6541: printf("%d|",(int)age);fflush(stdout);
6542: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6543:
6544: /* Computing expectancies */
6545: for(i=1; i<=nlstate;i++)
6546: for(j=1; j<=nlstate;j++)
6547: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6548: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
6549:
6550: /* 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]);*/
6551:
6552: }
6553:
6554: fprintf(ficreseij,"%3.0f",age );
6555: for(i=1; i<=nlstate;i++){
6556: eip=0;
6557: for(j=1; j<=nlstate;j++){
6558: eip +=eij[i][j][(int)age];
6559: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
6560: }
6561: fprintf(ficreseij,"%9.4f", eip );
6562: }
6563: fprintf(ficreseij,"\n");
6564:
6565: }
6566: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6567: printf("\n");
6568: fprintf(ficlog,"\n");
6569:
6570: }
6571:
1.235 brouard 6572: 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 6573:
6574: {
6575: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6576: to initial status i, ei. .
1.126 brouard 6577: */
1.336 brouard 6578: /* Very time consuming function, but already optimized with precov */
1.126 brouard 6579: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6580: int nhstepma, nstepma; /* Decreasing with age */
6581: double age, agelim, hf;
6582: double ***p3matp, ***p3matm, ***varhe;
6583: double **dnewm,**doldm;
6584: double *xp, *xm;
6585: double **gp, **gm;
6586: double ***gradg, ***trgradg;
6587: int theta;
6588:
6589: double eip, vip;
6590:
6591: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6592: xp=vector(1,npar);
6593: xm=vector(1,npar);
6594: dnewm=matrix(1,nlstate*nlstate,1,npar);
6595: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6596:
6597: pstamp(ficresstdeij);
6598: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6599: fprintf(ficresstdeij,"# Age");
6600: for(i=1; i<=nlstate;i++){
6601: for(j=1; j<=nlstate;j++)
6602: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6603: fprintf(ficresstdeij," e%1d. ",i);
6604: }
6605: fprintf(ficresstdeij,"\n");
6606:
6607: pstamp(ficrescveij);
6608: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6609: fprintf(ficrescveij,"# Age");
6610: for(i=1; i<=nlstate;i++)
6611: for(j=1; j<=nlstate;j++){
6612: cptj= (j-1)*nlstate+i;
6613: for(i2=1; i2<=nlstate;i2++)
6614: for(j2=1; j2<=nlstate;j2++){
6615: cptj2= (j2-1)*nlstate+i2;
6616: if(cptj2 <= cptj)
6617: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6618: }
6619: }
6620: fprintf(ficrescveij,"\n");
6621:
6622: if(estepm < stepm){
6623: printf ("Problem %d lower than %d\n",estepm, stepm);
6624: }
6625: else hstepm=estepm;
6626: /* We compute the life expectancy from trapezoids spaced every estepm months
6627: * This is mainly to measure the difference between two models: for example
6628: * if stepm=24 months pijx are given only every 2 years and by summing them
6629: * we are calculating an estimate of the Life Expectancy assuming a linear
6630: * progression in between and thus overestimating or underestimating according
6631: * to the curvature of the survival function. If, for the same date, we
6632: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6633: * to compare the new estimate of Life expectancy with the same linear
6634: * hypothesis. A more precise result, taking into account a more precise
6635: * curvature will be obtained if estepm is as small as stepm. */
6636:
6637: /* For example we decided to compute the life expectancy with the smallest unit */
6638: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6639: nhstepm is the number of hstepm from age to agelim
6640: nstepm is the number of stepm from age to agelin.
6641: Look at hpijx to understand the reason of that which relies in memory size
6642: and note for a fixed period like estepm months */
6643: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6644: survival function given by stepm (the optimization length). Unfortunately it
6645: means that if the survival funtion is printed only each two years of age and if
6646: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6647: results. So we changed our mind and took the option of the best precision.
6648: */
6649: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6650:
6651: /* If stepm=6 months */
6652: /* nhstepm age range expressed in number of stepm */
6653: agelim=AGESUP;
6654: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6655: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6656: /* if (stepm >= YEARM) hstepm=1;*/
6657: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6658:
6659: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6660: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6661: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6662: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6663: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6664: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6665:
6666: for (age=bage; age<=fage; age ++){
6667: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6668: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6669: /* if (stepm >= YEARM) hstepm=1;*/
6670: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6671:
1.126 brouard 6672: /* If stepm=6 months */
6673: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6674: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6675:
6676: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6677:
1.126 brouard 6678: /* Computing Variances of health expectancies */
6679: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6680: decrease memory allocation */
6681: for(theta=1; theta <=npar; theta++){
6682: for(i=1; i<=npar; i++){
1.222 brouard 6683: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6684: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6685: }
1.235 brouard 6686: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6687: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6688:
1.126 brouard 6689: for(j=1; j<= nlstate; j++){
1.222 brouard 6690: for(i=1; i<=nlstate; i++){
6691: for(h=0; h<=nhstepm-1; h++){
6692: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6693: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6694: }
6695: }
1.126 brouard 6696: }
1.218 brouard 6697:
1.126 brouard 6698: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6699: for(h=0; h<=nhstepm-1; h++){
6700: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6701: }
1.126 brouard 6702: }/* End theta */
6703:
6704:
6705: for(h=0; h<=nhstepm-1; h++)
6706: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6707: for(theta=1; theta <=npar; theta++)
6708: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6709:
1.218 brouard 6710:
1.222 brouard 6711: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6712: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6713: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6714:
1.222 brouard 6715: printf("%d|",(int)age);fflush(stdout);
6716: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6717: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6718: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6719: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6720: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6721: for(ij=1;ij<=nlstate*nlstate;ij++)
6722: for(ji=1;ji<=nlstate*nlstate;ji++)
6723: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6724: }
6725: }
1.320 brouard 6726: /* if((int)age ==50){ */
6727: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6728: /* } */
1.126 brouard 6729: /* Computing expectancies */
1.235 brouard 6730: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6731: for(i=1; i<=nlstate;i++)
6732: for(j=1; j<=nlstate;j++)
1.222 brouard 6733: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6734: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6735:
1.222 brouard 6736: /* 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 6737:
1.222 brouard 6738: }
1.269 brouard 6739:
6740: /* Standard deviation of expectancies ij */
1.126 brouard 6741: fprintf(ficresstdeij,"%3.0f",age );
6742: for(i=1; i<=nlstate;i++){
6743: eip=0.;
6744: vip=0.;
6745: for(j=1; j<=nlstate;j++){
1.222 brouard 6746: eip += eij[i][j][(int)age];
6747: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6748: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6749: 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 6750: }
6751: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6752: }
6753: fprintf(ficresstdeij,"\n");
1.218 brouard 6754:
1.269 brouard 6755: /* Variance of expectancies ij */
1.126 brouard 6756: fprintf(ficrescveij,"%3.0f",age );
6757: for(i=1; i<=nlstate;i++)
6758: for(j=1; j<=nlstate;j++){
1.222 brouard 6759: cptj= (j-1)*nlstate+i;
6760: for(i2=1; i2<=nlstate;i2++)
6761: for(j2=1; j2<=nlstate;j2++){
6762: cptj2= (j2-1)*nlstate+i2;
6763: if(cptj2 <= cptj)
6764: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6765: }
1.126 brouard 6766: }
6767: fprintf(ficrescveij,"\n");
1.218 brouard 6768:
1.126 brouard 6769: }
6770: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6771: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6772: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6773: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6774: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6775: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6776: printf("\n");
6777: fprintf(ficlog,"\n");
1.218 brouard 6778:
1.126 brouard 6779: free_vector(xm,1,npar);
6780: free_vector(xp,1,npar);
6781: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6782: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6783: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6784: }
1.218 brouard 6785:
1.126 brouard 6786: /************ Variance ******************/
1.235 brouard 6787: 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 6788: {
1.279 brouard 6789: /** Variance of health expectancies
6790: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6791: * double **newm;
6792: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6793: */
1.218 brouard 6794:
6795: /* int movingaverage(); */
6796: double **dnewm,**doldm;
6797: double **dnewmp,**doldmp;
6798: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6799: int first=0;
1.218 brouard 6800: int k;
6801: double *xp;
1.279 brouard 6802: double **gp, **gm; /**< for var eij */
6803: double ***gradg, ***trgradg; /**< for var eij */
6804: double **gradgp, **trgradgp; /**< for var p point j */
6805: double *gpp, *gmp; /**< for var p point j */
6806: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6807: double ***p3mat;
6808: double age,agelim, hf;
6809: /* double ***mobaverage; */
6810: int theta;
6811: char digit[4];
6812: char digitp[25];
6813:
6814: char fileresprobmorprev[FILENAMELENGTH];
6815:
6816: if(popbased==1){
6817: if(mobilav!=0)
6818: strcpy(digitp,"-POPULBASED-MOBILAV_");
6819: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6820: }
6821: else
6822: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6823:
1.218 brouard 6824: /* if (mobilav!=0) { */
6825: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6826: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6827: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6828: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6829: /* } */
6830: /* } */
6831:
6832: strcpy(fileresprobmorprev,"PRMORPREV-");
6833: sprintf(digit,"%-d",ij);
6834: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6835: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6836: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6837: strcat(fileresprobmorprev,fileresu);
6838: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6839: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6840: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6841: }
6842: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6843: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6844: pstamp(ficresprobmorprev);
6845: 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 6846: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
1.337 brouard 6847:
6848: /* We use TinvDoQresult[nres][resultmodel[nres][j] we sort according to the equation model and the resultline: it is a choice */
6849: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ /\* To be done*\/ */
6850: /* fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
6851: /* } */
6852: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */ /* To be done*/
1.344 brouard 6853: /* fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); */
1.337 brouard 6854: fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 6855: }
1.337 brouard 6856: /* for(j=1;j<=cptcoveff;j++) */
6857: /* fprintf(ficresprobmorprev," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,TnsdVar[Tvaraff[j]])]); */
1.238 brouard 6858: fprintf(ficresprobmorprev,"\n");
6859:
1.218 brouard 6860: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6861: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6862: fprintf(ficresprobmorprev," p.%-d SE",j);
6863: for(i=1; i<=nlstate;i++)
6864: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6865: }
6866: fprintf(ficresprobmorprev,"\n");
6867:
6868: fprintf(ficgp,"\n# Routine varevsij");
6869: fprintf(ficgp,"\nunset title \n");
6870: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6871: 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");
6872: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6873:
1.218 brouard 6874: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6875: pstamp(ficresvij);
6876: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6877: if(popbased==1)
6878: 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);
6879: else
6880: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6881: fprintf(ficresvij,"# Age");
6882: for(i=1; i<=nlstate;i++)
6883: for(j=1; j<=nlstate;j++)
6884: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6885: fprintf(ficresvij,"\n");
6886:
6887: xp=vector(1,npar);
6888: dnewm=matrix(1,nlstate,1,npar);
6889: doldm=matrix(1,nlstate,1,nlstate);
6890: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6891: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6892:
6893: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6894: gpp=vector(nlstate+1,nlstate+ndeath);
6895: gmp=vector(nlstate+1,nlstate+ndeath);
6896: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6897:
1.218 brouard 6898: if(estepm < stepm){
6899: printf ("Problem %d lower than %d\n",estepm, stepm);
6900: }
6901: else hstepm=estepm;
6902: /* For example we decided to compute the life expectancy with the smallest unit */
6903: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6904: nhstepm is the number of hstepm from age to agelim
6905: nstepm is the number of stepm from age to agelim.
6906: Look at function hpijx to understand why because of memory size limitations,
6907: we decided (b) to get a life expectancy respecting the most precise curvature of the
6908: survival function given by stepm (the optimization length). Unfortunately it
6909: means that if the survival funtion is printed every two years of age and if
6910: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6911: results. So we changed our mind and took the option of the best precision.
6912: */
6913: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6914: agelim = AGESUP;
6915: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6916: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6917: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6918: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6919: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6920: gp=matrix(0,nhstepm,1,nlstate);
6921: gm=matrix(0,nhstepm,1,nlstate);
6922:
6923:
6924: for(theta=1; theta <=npar; theta++){
6925: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6926: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6927: }
1.279 brouard 6928: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6929: * returns into prlim .
1.288 brouard 6930: */
1.242 brouard 6931: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6932:
6933: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6934: if (popbased==1) {
6935: if(mobilav ==0){
6936: for(i=1; i<=nlstate;i++)
6937: prlim[i][i]=probs[(int)age][i][ij];
6938: }else{ /* mobilav */
6939: for(i=1; i<=nlstate;i++)
6940: prlim[i][i]=mobaverage[(int)age][i][ij];
6941: }
6942: }
1.295 brouard 6943: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6944: */
6945: 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 6946: /**< 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 6947: * at horizon h in state j including mortality.
6948: */
1.218 brouard 6949: for(j=1; j<= nlstate; j++){
6950: for(h=0; h<=nhstepm; h++){
6951: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6952: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6953: }
6954: }
1.279 brouard 6955: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6956: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6957: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6958: */
6959: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6960: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6961: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6962: }
6963:
6964: /* Again with minus shift */
1.218 brouard 6965:
6966: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6967: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6968:
1.242 brouard 6969: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6970:
6971: if (popbased==1) {
6972: if(mobilav ==0){
6973: for(i=1; i<=nlstate;i++)
6974: prlim[i][i]=probs[(int)age][i][ij];
6975: }else{ /* mobilav */
6976: for(i=1; i<=nlstate;i++)
6977: prlim[i][i]=mobaverage[(int)age][i][ij];
6978: }
6979: }
6980:
1.235 brouard 6981: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6982:
6983: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6984: for(h=0; h<=nhstepm; h++){
6985: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6986: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6987: }
6988: }
6989: /* This for computing probability of death (h=1 means
6990: computed over hstepm matrices product = hstepm*stepm months)
6991: as a weighted average of prlim.
6992: */
6993: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6994: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6995: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6996: }
1.279 brouard 6997: /* end shifting computations */
6998:
6999: /**< Computing gradient matrix at horizon h
7000: */
1.218 brouard 7001: for(j=1; j<= nlstate; j++) /* vareij */
7002: for(h=0; h<=nhstepm; h++){
7003: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
7004: }
1.279 brouard 7005: /**< Gradient of overall mortality p.3 (or p.j)
7006: */
7007: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 7008: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
7009: }
7010:
7011: } /* End theta */
1.279 brouard 7012:
7013: /* We got the gradient matrix for each theta and state j */
1.218 brouard 7014: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
7015:
7016: for(h=0; h<=nhstepm; h++) /* veij */
7017: for(j=1; j<=nlstate;j++)
7018: for(theta=1; theta <=npar; theta++)
7019: trgradg[h][j][theta]=gradg[h][theta][j];
7020:
7021: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
7022: for(theta=1; theta <=npar; theta++)
7023: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 7024: /**< as well as its transposed matrix
7025: */
1.218 brouard 7026:
7027: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
7028: for(i=1;i<=nlstate;i++)
7029: for(j=1;j<=nlstate;j++)
7030: vareij[i][j][(int)age] =0.;
1.279 brouard 7031:
7032: /* Computing trgradg by matcov by gradg at age and summing over h
7033: * and k (nhstepm) formula 15 of article
7034: * Lievre-Brouard-Heathcote
7035: */
7036:
1.218 brouard 7037: for(h=0;h<=nhstepm;h++){
7038: for(k=0;k<=nhstepm;k++){
7039: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
7040: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
7041: for(i=1;i<=nlstate;i++)
7042: for(j=1;j<=nlstate;j++)
7043: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
7044: }
7045: }
7046:
1.279 brouard 7047: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
7048: * p.j overall mortality formula 49 but computed directly because
7049: * we compute the grad (wix pijx) instead of grad (pijx),even if
7050: * wix is independent of theta.
7051: */
1.218 brouard 7052: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
7053: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
7054: for(j=nlstate+1;j<=nlstate+ndeath;j++)
7055: for(i=nlstate+1;i<=nlstate+ndeath;i++)
7056: varppt[j][i]=doldmp[j][i];
7057: /* end ppptj */
7058: /* x centered again */
7059:
1.242 brouard 7060: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 7061:
7062: if (popbased==1) {
7063: if(mobilav ==0){
7064: for(i=1; i<=nlstate;i++)
7065: prlim[i][i]=probs[(int)age][i][ij];
7066: }else{ /* mobilav */
7067: for(i=1; i<=nlstate;i++)
7068: prlim[i][i]=mobaverage[(int)age][i][ij];
7069: }
7070: }
7071:
7072: /* This for computing probability of death (h=1 means
7073: computed over hstepm (estepm) matrices product = hstepm*stepm months)
7074: as a weighted average of prlim.
7075: */
1.235 brouard 7076: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 7077: for(j=nlstate+1;j<=nlstate+ndeath;j++){
7078: for(i=1,gmp[j]=0.;i<= nlstate; i++)
7079: gmp[j] += prlim[i][i]*p3mat[i][j][1];
7080: }
7081: /* end probability of death */
7082:
7083: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
7084: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
7085: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
7086: for(i=1; i<=nlstate;i++){
7087: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
7088: }
7089: }
7090: fprintf(ficresprobmorprev,"\n");
7091:
7092: fprintf(ficresvij,"%.0f ",age );
7093: for(i=1; i<=nlstate;i++)
7094: for(j=1; j<=nlstate;j++){
7095: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
7096: }
7097: fprintf(ficresvij,"\n");
7098: free_matrix(gp,0,nhstepm,1,nlstate);
7099: free_matrix(gm,0,nhstepm,1,nlstate);
7100: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
7101: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
7102: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7103: } /* End age */
7104: free_vector(gpp,nlstate+1,nlstate+ndeath);
7105: free_vector(gmp,nlstate+1,nlstate+ndeath);
7106: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
7107: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
7108: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
7109: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
7110: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
7111: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
7112: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
7113: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
7114: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
7115: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
7116: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
7117: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
7118: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
7119: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
7120: 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);
7121: /* 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 7122: */
1.218 brouard 7123: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
7124: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 7125:
1.218 brouard 7126: free_vector(xp,1,npar);
7127: free_matrix(doldm,1,nlstate,1,nlstate);
7128: free_matrix(dnewm,1,nlstate,1,npar);
7129: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
7130: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
7131: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
7132: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7133: fclose(ficresprobmorprev);
7134: fflush(ficgp);
7135: fflush(fichtm);
7136: } /* end varevsij */
1.126 brouard 7137:
7138: /************ Variance of prevlim ******************/
1.269 brouard 7139: 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 7140: {
1.205 brouard 7141: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 7142: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 7143:
1.268 brouard 7144: double **dnewmpar,**doldm;
1.126 brouard 7145: int i, j, nhstepm, hstepm;
7146: double *xp;
7147: double *gp, *gm;
7148: double **gradg, **trgradg;
1.208 brouard 7149: double **mgm, **mgp;
1.126 brouard 7150: double age,agelim;
7151: int theta;
7152:
7153: pstamp(ficresvpl);
1.288 brouard 7154: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 7155: fprintf(ficresvpl,"# Age ");
7156: if(nresult >=1)
7157: fprintf(ficresvpl," Result# ");
1.126 brouard 7158: for(i=1; i<=nlstate;i++)
7159: fprintf(ficresvpl," %1d-%1d",i,i);
7160: fprintf(ficresvpl,"\n");
7161:
7162: xp=vector(1,npar);
1.268 brouard 7163: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 7164: doldm=matrix(1,nlstate,1,nlstate);
7165:
7166: hstepm=1*YEARM; /* Every year of age */
7167: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
7168: agelim = AGESUP;
7169: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
7170: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
7171: if (stepm >= YEARM) hstepm=1;
7172: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
7173: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 7174: mgp=matrix(1,npar,1,nlstate);
7175: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 7176: gp=vector(1,nlstate);
7177: gm=vector(1,nlstate);
7178:
7179: for(theta=1; theta <=npar; theta++){
7180: for(i=1; i<=npar; i++){ /* Computes gradient */
7181: xp[i] = x[i] + (i==theta ?delti[theta]:0);
7182: }
1.288 brouard 7183: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
7184: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
7185: /* else */
7186: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 7187: for(i=1;i<=nlstate;i++){
1.126 brouard 7188: gp[i] = prlim[i][i];
1.208 brouard 7189: mgp[theta][i] = prlim[i][i];
7190: }
1.126 brouard 7191: for(i=1; i<=npar; i++) /* Computes gradient */
7192: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 7193: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
7194: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
7195: /* else */
7196: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 7197: for(i=1;i<=nlstate;i++){
1.126 brouard 7198: gm[i] = prlim[i][i];
1.208 brouard 7199: mgm[theta][i] = prlim[i][i];
7200: }
1.126 brouard 7201: for(i=1;i<=nlstate;i++)
7202: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 7203: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 7204: } /* End theta */
7205:
7206: trgradg =matrix(1,nlstate,1,npar);
7207:
7208: for(j=1; j<=nlstate;j++)
7209: for(theta=1; theta <=npar; theta++)
7210: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 7211: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7212: /* printf("\nmgm mgp %d ",(int)age); */
7213: /* for(j=1; j<=nlstate;j++){ */
7214: /* printf(" %d ",j); */
7215: /* for(theta=1; theta <=npar; theta++) */
7216: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7217: /* printf("\n "); */
7218: /* } */
7219: /* } */
7220: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7221: /* printf("\n gradg %d ",(int)age); */
7222: /* for(j=1; j<=nlstate;j++){ */
7223: /* printf("%d ",j); */
7224: /* for(theta=1; theta <=npar; theta++) */
7225: /* printf("%d %lf ",theta,gradg[theta][j]); */
7226: /* printf("\n "); */
7227: /* } */
7228: /* } */
1.126 brouard 7229:
7230: for(i=1;i<=nlstate;i++)
7231: varpl[i][(int)age] =0.;
1.209 brouard 7232: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 7233: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7234: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 7235: }else{
1.268 brouard 7236: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7237: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 7238: }
1.126 brouard 7239: for(i=1;i<=nlstate;i++)
7240: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7241:
7242: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 7243: if(nresult >=1)
7244: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 7245: for(i=1; i<=nlstate;i++){
1.126 brouard 7246: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 7247: /* for(j=1;j<=nlstate;j++) */
7248: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
7249: }
1.126 brouard 7250: fprintf(ficresvpl,"\n");
7251: free_vector(gp,1,nlstate);
7252: free_vector(gm,1,nlstate);
1.208 brouard 7253: free_matrix(mgm,1,npar,1,nlstate);
7254: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 7255: free_matrix(gradg,1,npar,1,nlstate);
7256: free_matrix(trgradg,1,nlstate,1,npar);
7257: } /* End age */
7258:
7259: free_vector(xp,1,npar);
7260: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 7261: free_matrix(dnewmpar,1,nlstate,1,nlstate);
7262:
7263: }
7264:
7265:
7266: /************ Variance of backprevalence limit ******************/
1.269 brouard 7267: 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 7268: {
7269: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
7270: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
7271:
7272: double **dnewmpar,**doldm;
7273: int i, j, nhstepm, hstepm;
7274: double *xp;
7275: double *gp, *gm;
7276: double **gradg, **trgradg;
7277: double **mgm, **mgp;
7278: double age,agelim;
7279: int theta;
7280:
7281: pstamp(ficresvbl);
7282: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
7283: fprintf(ficresvbl,"# Age ");
7284: if(nresult >=1)
7285: fprintf(ficresvbl," Result# ");
7286: for(i=1; i<=nlstate;i++)
7287: fprintf(ficresvbl," %1d-%1d",i,i);
7288: fprintf(ficresvbl,"\n");
7289:
7290: xp=vector(1,npar);
7291: dnewmpar=matrix(1,nlstate,1,npar);
7292: doldm=matrix(1,nlstate,1,nlstate);
7293:
7294: hstepm=1*YEARM; /* Every year of age */
7295: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
7296: agelim = AGEINF;
7297: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
7298: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
7299: if (stepm >= YEARM) hstepm=1;
7300: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
7301: gradg=matrix(1,npar,1,nlstate);
7302: mgp=matrix(1,npar,1,nlstate);
7303: mgm=matrix(1,npar,1,nlstate);
7304: gp=vector(1,nlstate);
7305: gm=vector(1,nlstate);
7306:
7307: for(theta=1; theta <=npar; theta++){
7308: for(i=1; i<=npar; i++){ /* Computes gradient */
7309: xp[i] = x[i] + (i==theta ?delti[theta]:0);
7310: }
7311: if(mobilavproj > 0 )
7312: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7313: else
7314: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7315: for(i=1;i<=nlstate;i++){
7316: gp[i] = bprlim[i][i];
7317: mgp[theta][i] = bprlim[i][i];
7318: }
7319: for(i=1; i<=npar; i++) /* Computes gradient */
7320: xp[i] = x[i] - (i==theta ?delti[theta]:0);
7321: if(mobilavproj > 0 )
7322: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7323: else
7324: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7325: for(i=1;i<=nlstate;i++){
7326: gm[i] = bprlim[i][i];
7327: mgm[theta][i] = bprlim[i][i];
7328: }
7329: for(i=1;i<=nlstate;i++)
7330: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
7331: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
7332: } /* End theta */
7333:
7334: trgradg =matrix(1,nlstate,1,npar);
7335:
7336: for(j=1; j<=nlstate;j++)
7337: for(theta=1; theta <=npar; theta++)
7338: trgradg[j][theta]=gradg[theta][j];
7339: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7340: /* printf("\nmgm mgp %d ",(int)age); */
7341: /* for(j=1; j<=nlstate;j++){ */
7342: /* printf(" %d ",j); */
7343: /* for(theta=1; theta <=npar; theta++) */
7344: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7345: /* printf("\n "); */
7346: /* } */
7347: /* } */
7348: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7349: /* printf("\n gradg %d ",(int)age); */
7350: /* for(j=1; j<=nlstate;j++){ */
7351: /* printf("%d ",j); */
7352: /* for(theta=1; theta <=npar; theta++) */
7353: /* printf("%d %lf ",theta,gradg[theta][j]); */
7354: /* printf("\n "); */
7355: /* } */
7356: /* } */
7357:
7358: for(i=1;i<=nlstate;i++)
7359: varbpl[i][(int)age] =0.;
7360: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
7361: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7362: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7363: }else{
7364: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7365: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7366: }
7367: for(i=1;i<=nlstate;i++)
7368: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7369:
7370: fprintf(ficresvbl,"%.0f ",age );
7371: if(nresult >=1)
7372: fprintf(ficresvbl,"%d ",nres );
7373: for(i=1; i<=nlstate;i++)
7374: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
7375: fprintf(ficresvbl,"\n");
7376: free_vector(gp,1,nlstate);
7377: free_vector(gm,1,nlstate);
7378: free_matrix(mgm,1,npar,1,nlstate);
7379: free_matrix(mgp,1,npar,1,nlstate);
7380: free_matrix(gradg,1,npar,1,nlstate);
7381: free_matrix(trgradg,1,nlstate,1,npar);
7382: } /* End age */
7383:
7384: free_vector(xp,1,npar);
7385: free_matrix(doldm,1,nlstate,1,npar);
7386: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 7387:
7388: }
7389:
7390: /************ Variance of one-step probabilities ******************/
7391: 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 7392: {
7393: int i, j=0, k1, l1, tj;
7394: int k2, l2, j1, z1;
7395: int k=0, l;
7396: int first=1, first1, first2;
1.326 brouard 7397: int nres=0; /* New */
1.222 brouard 7398: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
7399: double **dnewm,**doldm;
7400: double *xp;
7401: double *gp, *gm;
7402: double **gradg, **trgradg;
7403: double **mu;
7404: double age, cov[NCOVMAX+1];
7405: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
7406: int theta;
7407: char fileresprob[FILENAMELENGTH];
7408: char fileresprobcov[FILENAMELENGTH];
7409: char fileresprobcor[FILENAMELENGTH];
7410: double ***varpij;
7411:
7412: strcpy(fileresprob,"PROB_");
7413: strcat(fileresprob,fileres);
7414: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
7415: printf("Problem with resultfile: %s\n", fileresprob);
7416: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
7417: }
7418: strcpy(fileresprobcov,"PROBCOV_");
7419: strcat(fileresprobcov,fileresu);
7420: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
7421: printf("Problem with resultfile: %s\n", fileresprobcov);
7422: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
7423: }
7424: strcpy(fileresprobcor,"PROBCOR_");
7425: strcat(fileresprobcor,fileresu);
7426: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
7427: printf("Problem with resultfile: %s\n", fileresprobcor);
7428: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
7429: }
7430: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7431: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7432: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7433: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7434: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7435: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7436: pstamp(ficresprob);
7437: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
7438: fprintf(ficresprob,"# Age");
7439: pstamp(ficresprobcov);
7440: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
7441: fprintf(ficresprobcov,"# Age");
7442: pstamp(ficresprobcor);
7443: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
7444: fprintf(ficresprobcor,"# Age");
1.126 brouard 7445:
7446:
1.222 brouard 7447: for(i=1; i<=nlstate;i++)
7448: for(j=1; j<=(nlstate+ndeath);j++){
7449: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
7450: fprintf(ficresprobcov," p%1d-%1d ",i,j);
7451: fprintf(ficresprobcor," p%1d-%1d ",i,j);
7452: }
7453: /* fprintf(ficresprob,"\n");
7454: fprintf(ficresprobcov,"\n");
7455: fprintf(ficresprobcor,"\n");
7456: */
7457: xp=vector(1,npar);
7458: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7459: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7460: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
7461: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
7462: first=1;
7463: fprintf(ficgp,"\n# Routine varprob");
7464: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
7465: fprintf(fichtm,"\n");
7466:
1.288 brouard 7467: 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 7468: 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);
7469: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 7470: and drawn. It helps understanding how is the covariance between two incidences.\
7471: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 7472: 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 7473: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
7474: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
7475: standard deviations wide on each axis. <br>\
7476: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
7477: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
7478: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
7479:
1.222 brouard 7480: cov[1]=1;
7481: /* tj=cptcoveff; */
1.225 brouard 7482: tj = (int) pow(2,cptcoveff);
1.222 brouard 7483: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
7484: j1=0;
1.332 brouard 7485:
7486: for(nres=1;nres <=nresult; nres++){ /* For each resultline */
7487: for(j1=1; j1<=tj;j1++){ /* For any combination of dummy covariates, fixed and varying */
1.342 brouard 7488: /* 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 7489: if(tj != 1 && TKresult[nres]!= j1)
7490: continue;
7491:
7492: /* for(j1=1; j1<=tj;j1++){ /\* For each valid combination of covariates or only once*\/ */
7493: /* for(nres=1;nres <=1; nres++){ /\* For each resultline *\/ */
7494: /* /\* for(nres=1;nres <=nresult; nres++){ /\\* For each resultline *\\/ *\/ */
1.222 brouard 7495: if (cptcovn>0) {
1.334 brouard 7496: fprintf(ficresprob, "\n#********** Variable ");
7497: fprintf(ficresprobcov, "\n#********** Variable ");
7498: fprintf(ficgp, "\n#********** Variable ");
7499: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
7500: fprintf(ficresprobcor, "\n#********** Variable ");
7501:
7502: /* Including quantitative variables of the resultline to be done */
7503: for (z1=1; z1<=cptcovs; z1++){ /* Loop on each variable of this resultline */
1.343 brouard 7504: /* printf("Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model); */
1.338 brouard 7505: fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model);
7506: /* 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 7507: if(Dummy[modelresult[nres][z1]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to z1 in resultline */
7508: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed referenced to model equation */
7509: 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 */
7510: 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 */
7511: 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 */
7512: 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 */
7513: 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 */
7514: fprintf(ficresprob,"fixed ");
7515: fprintf(ficresprobcov,"fixed ");
7516: fprintf(ficgp,"fixed ");
7517: fprintf(fichtmcov,"fixed ");
7518: fprintf(ficresprobcor,"fixed ");
7519: }else{
7520: fprintf(ficresprob,"varyi ");
7521: fprintf(ficresprobcov,"varyi ");
7522: fprintf(ficgp,"varyi ");
7523: fprintf(fichtmcov,"varyi ");
7524: fprintf(ficresprobcor,"varyi ");
7525: }
7526: }else if(Dummy[modelresult[nres][z1]]==1){ /* Quanti variable */
7527: /* For each selected (single) quantitative value */
1.337 brouard 7528: fprintf(ficresprob," V%d=%lg ",Tvqresult[nres][z1],Tqresult[nres][z1]);
1.334 brouard 7529: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed */
7530: fprintf(ficresprob,"fixed ");
7531: fprintf(ficresprobcov,"fixed ");
7532: fprintf(ficgp,"fixed ");
7533: fprintf(fichtmcov,"fixed ");
7534: fprintf(ficresprobcor,"fixed ");
7535: }else{
7536: fprintf(ficresprob,"varyi ");
7537: fprintf(ficresprobcov,"varyi ");
7538: fprintf(ficgp,"varyi ");
7539: fprintf(fichtmcov,"varyi ");
7540: fprintf(ficresprobcor,"varyi ");
7541: }
7542: }else{
7543: 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 */
7544: 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 */
7545: exit(1);
7546: }
7547: } /* End loop on variable of this resultline */
7548: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]); */
1.222 brouard 7549: fprintf(ficresprob, "**********\n#\n");
7550: fprintf(ficresprobcov, "**********\n#\n");
7551: fprintf(ficgp, "**********\n#\n");
7552: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
7553: fprintf(ficresprobcor, "**********\n#");
7554: if(invalidvarcomb[j1]){
7555: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
7556: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
7557: continue;
7558: }
7559: }
7560: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
7561: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7562: gp=vector(1,(nlstate)*(nlstate+ndeath));
7563: gm=vector(1,(nlstate)*(nlstate+ndeath));
1.334 brouard 7564: for (age=bage; age<=fage; age ++){ /* Fo each age we feed the model equation with covariates, using precov as in hpxij() ? */
1.222 brouard 7565: cov[2]=age;
7566: if(nagesqr==1)
7567: cov[3]= age*age;
1.334 brouard 7568: /* New code end of combination but for each resultline */
7569: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
7570: if(Typevar[k1]==1){ /* A product with age */
7571: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.326 brouard 7572: }else{
1.334 brouard 7573: cov[2+nagesqr+k1]=precov[nres][k1];
1.326 brouard 7574: }
1.334 brouard 7575: }/* End of loop on model equation */
7576: /* Old code */
7577: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
7578: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; *\/ */
7579: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
7580: /* /\* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates *\/ */
7581: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TnsdVar[TvarsD[k]])]; */
7582: /* /\*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*\//\* j1 1 2 3 4 */
7583: /* * 1 1 1 1 1 */
7584: /* * 2 2 1 1 1 */
7585: /* * 3 1 2 1 1 */
7586: /* *\/ */
7587: /* /\* nbcode[1][1]=0 nbcode[1][2]=1;*\/ */
7588: /* } */
7589: /* /\* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
7590: /* /\* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] *\/ */
7591: /* /\*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; *\/ */
7592: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
7593: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
7594: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,TnsdVar[Tvar[Tage[k]]])]*cov[2]; */
7595: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7596: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
7597: /* 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]); */
7598: /* /\* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\\* Using the mean of quantitative variable Tvar[Tage[k]] /\\* Tqresult[nres][k]; *\\/ *\/ */
7599: /* /\* exit(1); *\/ */
7600: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
7601: /* } */
7602: /* /\* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7603: /* } */
7604: /* for (k=1; k<=cptcovprod;k++){/\* For product without age *\/ */
7605: /* if(Dummy[Tvard[k][1]]==0){ */
7606: /* if(Dummy[Tvard[k][2]]==0){ */
7607: /* 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]])]; */
7608: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7609: /* }else{ /\* Should we use the mean of the quantitative variables? *\/ */
7610: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * Tqresult[nres][resultmodel[nres][k]]; */
7611: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
7612: /* } */
7613: /* }else{ */
7614: /* if(Dummy[Tvard[k][2]]==0){ */
7615: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][TnsdVar[Tvard[k][1]]]; */
7616: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
7617: /* }else{ */
7618: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][TnsdVar[Tvard[k][1]]]* Tqinvresult[nres][TnsdVar[Tvard[k][2]]]; */
7619: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
7620: /* } */
7621: /* } */
7622: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7623: /* } */
1.326 brouard 7624: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 7625: for(theta=1; theta <=npar; theta++){
7626: for(i=1; i<=npar; i++)
7627: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7628:
1.222 brouard 7629: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7630:
1.222 brouard 7631: k=0;
7632: for(i=1; i<= (nlstate); i++){
7633: for(j=1; j<=(nlstate+ndeath);j++){
7634: k=k+1;
7635: gp[k]=pmmij[i][j];
7636: }
7637: }
1.220 brouard 7638:
1.222 brouard 7639: for(i=1; i<=npar; i++)
7640: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7641:
1.222 brouard 7642: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7643: k=0;
7644: for(i=1; i<=(nlstate); i++){
7645: for(j=1; j<=(nlstate+ndeath);j++){
7646: k=k+1;
7647: gm[k]=pmmij[i][j];
7648: }
7649: }
1.220 brouard 7650:
1.222 brouard 7651: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7652: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7653: }
1.126 brouard 7654:
1.222 brouard 7655: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7656: for(theta=1; theta <=npar; theta++)
7657: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7658:
1.222 brouard 7659: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7660: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7661:
1.222 brouard 7662: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7663:
1.222 brouard 7664: k=0;
7665: for(i=1; i<=(nlstate); i++){
7666: for(j=1; j<=(nlstate+ndeath);j++){
7667: k=k+1;
7668: mu[k][(int) age]=pmmij[i][j];
7669: }
7670: }
7671: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7672: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7673: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7674:
1.222 brouard 7675: /*printf("\n%d ",(int)age);
7676: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7677: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7678: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7679: }*/
1.220 brouard 7680:
1.222 brouard 7681: fprintf(ficresprob,"\n%d ",(int)age);
7682: fprintf(ficresprobcov,"\n%d ",(int)age);
7683: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7684:
1.222 brouard 7685: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7686: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7687: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7688: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7689: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7690: }
7691: i=0;
7692: for (k=1; k<=(nlstate);k++){
7693: for (l=1; l<=(nlstate+ndeath);l++){
7694: i++;
7695: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7696: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7697: for (j=1; j<=i;j++){
7698: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7699: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7700: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7701: }
7702: }
7703: }/* end of loop for state */
7704: } /* end of loop for age */
7705: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7706: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7707: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7708: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7709:
7710: /* Confidence intervalle of pij */
7711: /*
7712: fprintf(ficgp,"\nunset parametric;unset label");
7713: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7714: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7715: 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);
7716: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7717: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7718: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7719: */
7720:
7721: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7722: first1=1;first2=2;
7723: for (k2=1; k2<=(nlstate);k2++){
7724: for (l2=1; l2<=(nlstate+ndeath);l2++){
7725: if(l2==k2) continue;
7726: j=(k2-1)*(nlstate+ndeath)+l2;
7727: for (k1=1; k1<=(nlstate);k1++){
7728: for (l1=1; l1<=(nlstate+ndeath);l1++){
7729: if(l1==k1) continue;
7730: i=(k1-1)*(nlstate+ndeath)+l1;
7731: if(i<=j) continue;
7732: for (age=bage; age<=fage; age ++){
7733: if ((int)age %5==0){
7734: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7735: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7736: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7737: mu1=mu[i][(int) age]/stepm*YEARM ;
7738: mu2=mu[j][(int) age]/stepm*YEARM;
7739: c12=cv12/sqrt(v1*v2);
7740: /* Computing eigen value of matrix of covariance */
7741: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7742: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7743: if ((lc2 <0) || (lc1 <0) ){
7744: if(first2==1){
7745: first1=0;
7746: 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);
7747: }
7748: 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);
7749: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7750: /* lc2=fabs(lc2); */
7751: }
1.220 brouard 7752:
1.222 brouard 7753: /* Eigen vectors */
1.280 brouard 7754: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7755: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7756: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7757: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7758: }else
7759: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7760: /*v21=sqrt(1.-v11*v11); *//* error */
7761: v21=(lc1-v1)/cv12*v11;
7762: v12=-v21;
7763: v22=v11;
7764: tnalp=v21/v11;
7765: if(first1==1){
7766: first1=0;
7767: 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);
7768: }
7769: 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);
7770: /*printf(fignu*/
7771: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7772: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7773: if(first==1){
7774: first=0;
7775: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7776: fprintf(ficgp,"\nset parametric;unset label");
7777: 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);
7778: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7779: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7780: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7781: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7782: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7783: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7784: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7785: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7786: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7787: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7788: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7789: 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 7790: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7791: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7792: }else{
7793: first=0;
7794: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7795: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7796: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7797: 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 7798: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7799: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7800: }/* if first */
7801: } /* age mod 5 */
7802: } /* end loop age */
7803: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7804: first=1;
7805: } /*l12 */
7806: } /* k12 */
7807: } /*l1 */
7808: }/* k1 */
1.332 brouard 7809: } /* loop on combination of covariates j1 */
1.326 brouard 7810: } /* loop on nres */
1.222 brouard 7811: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7812: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7813: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7814: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7815: free_vector(xp,1,npar);
7816: fclose(ficresprob);
7817: fclose(ficresprobcov);
7818: fclose(ficresprobcor);
7819: fflush(ficgp);
7820: fflush(fichtmcov);
7821: }
1.126 brouard 7822:
7823:
7824: /******************* Printing html file ***********/
1.201 brouard 7825: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7826: int lastpass, int stepm, int weightopt, char model[],\
7827: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7828: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7829: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7830: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7831: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7832: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7833: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7834: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7835: </ul>");
1.319 brouard 7836: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7837: /* </ul>", model); */
1.214 brouard 7838: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7839: 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",
7840: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
1.332 brouard 7841: 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 7842: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7843: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7844: fprintf(fichtm,"\
7845: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7846: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7847: fprintf(fichtm,"\
1.217 brouard 7848: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7849: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7850: fprintf(fichtm,"\
1.288 brouard 7851: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7852: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7853: fprintf(fichtm,"\
1.288 brouard 7854: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7855: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7856: fprintf(fichtm,"\
1.211 brouard 7857: - (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 7858: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7859: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7860: if(prevfcast==1){
7861: fprintf(fichtm,"\
7862: - Prevalence projections by age and states: \
1.201 brouard 7863: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7864: }
1.126 brouard 7865:
7866:
1.225 brouard 7867: m=pow(2,cptcoveff);
1.222 brouard 7868: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7869:
1.317 brouard 7870: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7871:
7872: jj1=0;
7873:
7874: fprintf(fichtm," \n<ul>");
1.337 brouard 7875: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7876: /* k1=nres; */
1.338 brouard 7877: k1=TKresult[nres];
7878: if(TKresult[nres]==0)k1=1; /* To be checked for no result */
1.337 brouard 7879: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
7880: /* if(m != 1 && TKresult[nres]!= k1) */
7881: /* continue; */
1.264 brouard 7882: jj1++;
7883: if (cptcovn > 0) {
7884: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
1.337 brouard 7885: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
7886: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7887: }
1.337 brouard 7888: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
7889: /* fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
7890: /* } */
7891: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7892: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7893: /* } */
1.264 brouard 7894: fprintf(fichtm,"\">");
7895:
7896: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7897: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 7898: for (cpt=1; cpt<=cptcovs;cpt++){
7899: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7900: }
1.337 brouard 7901: /* fprintf(fichtm,"************ Results for covariates"); */
7902: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
7903: /* fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
7904: /* } */
7905: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7906: /* fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7907: /* } */
1.264 brouard 7908: if(invalidvarcomb[k1]){
7909: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7910: continue;
7911: }
7912: fprintf(fichtm,"</a></li>");
7913: } /* cptcovn >0 */
7914: }
1.317 brouard 7915: fprintf(fichtm," \n</ul>");
1.264 brouard 7916:
1.222 brouard 7917: jj1=0;
1.237 brouard 7918:
1.337 brouard 7919: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7920: /* k1=nres; */
1.338 brouard 7921: k1=TKresult[nres];
7922: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 7923: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
7924: /* if(m != 1 && TKresult[nres]!= k1) */
7925: /* continue; */
1.220 brouard 7926:
1.222 brouard 7927: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7928: jj1++;
7929: if (cptcovn > 0) {
1.264 brouard 7930: fprintf(fichtm,"\n<p><a name=\"rescov");
1.337 brouard 7931: for (cpt=1; cpt<=cptcovs;cpt++){
7932: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7933: }
1.337 brouard 7934: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7935: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7936: /* } */
1.264 brouard 7937: fprintf(fichtm,"\"</a>");
7938:
1.222 brouard 7939: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 7940: for (cpt=1; cpt<=cptcovs;cpt++){
7941: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
7942: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 7943: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7944: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7945: }
1.230 brouard 7946: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.338 brouard 7947: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7948: if(invalidvarcomb[k1]){
7949: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7950: printf("\nCombination (%d) ignored because no cases \n",k1);
7951: continue;
7952: }
7953: }
7954: /* aij, bij */
1.259 brouard 7955: 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 7956: <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 7957: /* Pij */
1.241 brouard 7958: 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> \
7959: <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 7960: /* Quasi-incidences */
7961: 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 7962: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7963: 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 7964: 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> \
7965: <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 7966: /* Survival functions (period) in state j */
7967: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7968: 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. <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);
7969: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7970: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7971: }
7972: /* State specific survival functions (period) */
7973: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7974: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7975: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.329 brouard 7976: <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);
7977: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7978: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7979: }
1.288 brouard 7980: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7981: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7982: 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 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 7983: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
1.329 brouard 7984: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7985: }
1.296 brouard 7986: if(prevbcast==1){
1.288 brouard 7987: /* Backward prevalence in each health state */
1.222 brouard 7988: for(cpt=1; cpt<=nlstate;cpt++){
1.338 brouard 7989: 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);
7990: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJB_"),subdirf2(optionfilefiname,"PIJB_"));
7991: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.222 brouard 7992: }
1.217 brouard 7993: }
1.222 brouard 7994: if(prevfcast==1){
1.288 brouard 7995: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7996: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7997: 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);
7998: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7999: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
8000: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 8001: }
8002: }
1.296 brouard 8003: if(prevbcast==1){
1.268 brouard 8004: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
8005: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 8006: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
8007: 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 \
8008: 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 8009: 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);
8010: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
8011: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 8012: }
8013: }
1.220 brouard 8014:
1.222 brouard 8015: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 8016: 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);
8017: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
8018: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 8019: }
8020: /* } /\* end i1 *\/ */
1.337 brouard 8021: }/* End k1=nres */
1.222 brouard 8022: fprintf(fichtm,"</ul>");
1.126 brouard 8023:
1.222 brouard 8024: fprintf(fichtm,"\
1.126 brouard 8025: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 8026: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 8027: - 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 8028: But because parameters are usually highly correlated (a higher incidence of disability \
8029: and a higher incidence of recovery can give very close observed transition) it might \
8030: be very useful to look not only at linear confidence intervals estimated from the \
8031: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
8032: (parameters) of the logistic regression, it might be more meaningful to visualize the \
8033: covariance matrix of the one-step probabilities. \
8034: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 8035:
1.222 brouard 8036: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
8037: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
8038: fprintf(fichtm,"\
1.126 brouard 8039: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 8040: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 8041:
1.222 brouard 8042: fprintf(fichtm,"\
1.126 brouard 8043: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 8044: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
8045: fprintf(fichtm,"\
1.126 brouard 8046: - 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): \
8047: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 8048: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 8049: fprintf(fichtm,"\
1.126 brouard 8050: - (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): \
8051: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 8052: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 8053: fprintf(fichtm,"\
1.288 brouard 8054: - 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 8055: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
8056: fprintf(fichtm,"\
1.128 brouard 8057: - 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 8058: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
8059: fprintf(fichtm,"\
1.288 brouard 8060: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 8061: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 8062:
8063: /* if(popforecast==1) fprintf(fichtm,"\n */
8064: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
8065: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
8066: /* <br>",fileres,fileres,fileres,fileres); */
8067: /* else */
1.338 brouard 8068: /* 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 8069: fflush(fichtm);
1.126 brouard 8070:
1.225 brouard 8071: m=pow(2,cptcoveff);
1.222 brouard 8072: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 8073:
1.317 brouard 8074: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
8075:
8076: jj1=0;
8077:
8078: fprintf(fichtm," \n<ul>");
1.337 brouard 8079: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8080: /* k1=nres; */
1.338 brouard 8081: k1=TKresult[nres];
1.337 brouard 8082: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
8083: /* if(m != 1 && TKresult[nres]!= k1) */
8084: /* continue; */
1.317 brouard 8085: jj1++;
8086: if (cptcovn > 0) {
8087: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
1.337 brouard 8088: for (cpt=1; cpt<=cptcovs;cpt++){
8089: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8090: }
8091: fprintf(fichtm,"\">");
8092:
8093: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
8094: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 8095: for (cpt=1; cpt<=cptcovs;cpt++){
8096: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8097: }
8098: if(invalidvarcomb[k1]){
8099: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
8100: continue;
8101: }
8102: fprintf(fichtm,"</a></li>");
8103: } /* cptcovn >0 */
1.337 brouard 8104: } /* End nres */
1.317 brouard 8105: fprintf(fichtm," \n</ul>");
8106:
1.222 brouard 8107: jj1=0;
1.237 brouard 8108:
1.241 brouard 8109: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8110: /* k1=nres; */
1.338 brouard 8111: k1=TKresult[nres];
8112: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8113: /* for(k1=1; k1<=m;k1++){ */
8114: /* if(m != 1 && TKresult[nres]!= k1) */
8115: /* continue; */
1.222 brouard 8116: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
8117: jj1++;
1.126 brouard 8118: if (cptcovn > 0) {
1.317 brouard 8119: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
1.337 brouard 8120: for (cpt=1; cpt<=cptcovs;cpt++){
8121: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8122: }
8123: fprintf(fichtm,"\"</a>");
8124:
1.126 brouard 8125: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 8126: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcoveff number of variables */
8127: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
8128: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 8129: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 8130: }
1.237 brouard 8131:
1.338 brouard 8132: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 8133:
1.222 brouard 8134: if(invalidvarcomb[k1]){
8135: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
8136: continue;
8137: }
1.337 brouard 8138: } /* If cptcovn >0 */
1.126 brouard 8139: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 8140: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 8141: 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);
8142: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
8143: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 8144: }
8145: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 8146: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 8147: true period expectancies (those weighted with period prevalences are also\
8148: drawn in addition to the population based expectancies computed using\
1.314 brouard 8149: 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);
8150: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
8151: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 8152: /* } /\* end i1 *\/ */
1.241 brouard 8153: }/* End nres */
1.222 brouard 8154: fprintf(fichtm,"</ul>");
8155: fflush(fichtm);
1.126 brouard 8156: }
8157:
8158: /******************* Gnuplot file **************/
1.296 brouard 8159: 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 8160:
8161: char dirfileres[132],optfileres[132];
1.264 brouard 8162: char gplotcondition[132], gplotlabel[132];
1.343 brouard 8163: 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 8164: int lv=0, vlv=0, kl=0;
1.130 brouard 8165: int ng=0;
1.201 brouard 8166: int vpopbased;
1.223 brouard 8167: int ioffset; /* variable offset for columns */
1.270 brouard 8168: int iyearc=1; /* variable column for year of projection */
8169: int iagec=1; /* variable column for age of projection */
1.235 brouard 8170: int nres=0; /* Index of resultline */
1.266 brouard 8171: int istart=1; /* For starting graphs in projections */
1.219 brouard 8172:
1.126 brouard 8173: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
8174: /* printf("Problem with file %s",optionfilegnuplot); */
8175: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
8176: /* } */
8177:
8178: /*#ifdef windows */
8179: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 8180: /*#endif */
1.225 brouard 8181: m=pow(2,cptcoveff);
1.126 brouard 8182:
1.274 brouard 8183: /* diagram of the model */
8184: fprintf(ficgp,"\n#Diagram of the model \n");
8185: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
8186: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
8187: 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);
8188:
1.343 brouard 8189: 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 8190: fprintf(ficgp,"\n#show arrow\nunset label\n");
8191: 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);
8192: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
8193: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
8194: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
8195: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
8196:
1.202 brouard 8197: /* Contribution to likelihood */
8198: /* Plot the probability implied in the likelihood */
1.223 brouard 8199: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
8200: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
8201: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
8202: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 8203: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 8204: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
8205: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 8206: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
8207: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
8208: 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));
8209: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
8210: 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));
8211: for (i=1; i<= nlstate ; i ++) {
8212: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
8213: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
8214: 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);
8215: for (j=2; j<= nlstate+ndeath ; j ++) {
8216: 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);
8217: }
8218: fprintf(ficgp,";\nset out; unset ylabel;\n");
8219: }
8220: /* 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 */
8221: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
8222: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
8223: fprintf(ficgp,"\nset out;unset log\n");
8224: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 8225:
1.343 brouard 8226: /* Plot the probability implied in the likelihood by covariate value */
8227: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
8228: /* if(debugILK==1){ */
8229: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
1.347 brouard 8230: kvar=Tvar[TvarFind[kf]]; /* variable name */
8231: /* k=18+Tvar[TvarFind[kf]];/\*offset because there are 18 columns in the ILK_ file but could be placed else where *\/ */
8232: k=18+kf;/*offset because there are 18 columns in the ILK_ file */
1.343 brouard 8233: for (i=1; i<= nlstate ; i ++) {
8234: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
8235: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
1.348 ! brouard 8236: if(gnuplotversion >=5.2){ /* Former gnuplot versions do not have variable pointsize!! */
! 8237: 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);
! 8238: for (j=2; j<= nlstate+ndeath ; j ++) {
! 8239: 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);
! 8240: }
! 8241: }else{
! 8242: 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);
! 8243: for (j=2; j<= nlstate+ndeath ; j ++) {
! 8244: 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);
! 8245: }
1.343 brouard 8246: }
8247: fprintf(ficgp,";\nset out; unset ylabel;\n");
8248: }
8249: } /* End of each covariate dummy */
8250: for(ncovv=1, iposold=0, kk=0; ncovv <= ncovvt ; ncovv++){
8251: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
8252: * kmodel = 1 2 3 4 5 6 7 8 9
8253: * varying 1 2 3 4 5
8254: * ncovv 1 2 3 4 5 6 7 8
8255: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
8256: * TvarVVind[ncovv]=kmodel 2 3 7 7 8 8 9 9
8257: * TvarFind[kmodel] 1 0 0 0 0 0 0 0 0
8258: * kdata ncovcol=[V1 V2] nqv=0 ntv=[V3 V4] nqtv=V5
8259: * Dummy[kmodel] 0 0 1 2 2 3 1 1 1
8260: */
8261: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
8262: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
8263: /* 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]); */
8264: if(ipos!=iposold){ /* Not a product or first of a product */
8265: /* printf(" %d",ipos); */
8266: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
8267: /* printf(" DebugILK ficgp suite ipos=%d != iposold=%d\n", ipos, iposold); */
8268: kk++; /* Position of the ncovv column in ILK_ */
8269: k=18+ncovf+kk; /*offset because there are 18 columns in the ILK_ file plus ncovf fixed covariate */
8270: 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) */
8271: for (i=1; i<= nlstate ; i ++) {
8272: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
8273: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
8274:
1.348 ! brouard 8275: /* printf("Before DebugILK gnuplotversion=%g >=5.2\n",gnuplotversion); */
1.343 brouard 8276: if(gnuplotversion >=5.2){ /* Former gnuplot versions do not have variable pointsize!! */
8277: /* printf("DebugILK gnuplotversion=%g >=5.2\n",gnuplotversion); */
8278: 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);
8279: for (j=2; j<= nlstate+ndeath ; j ++) {
8280: 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);
8281: }
8282: }else{
8283: /* printf("DebugILK gnuplotversion=%g <5.2\n",gnuplotversion); */
8284: 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);
8285: for (j=2; j<= nlstate+ndeath ; j ++) {
8286: 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);
8287: }
8288: }
8289: fprintf(ficgp,";\nset out; unset ylabel;\n");
8290: }
8291: }/* End if dummy varying */
8292: }else{ /*Product */
8293: /* printf("*"); */
8294: /* fprintf(ficresilk,"*"); */
8295: }
8296: iposold=ipos;
8297: } /* For each time varying covariate */
8298: /* } /\* debugILK==1 *\/ */
8299: /* 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 */
8300: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
8301: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
8302: fprintf(ficgp,"\nset out;unset log\n");
8303: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
8304:
8305:
8306:
1.126 brouard 8307: strcpy(dirfileres,optionfilefiname);
8308: strcpy(optfileres,"vpl");
1.223 brouard 8309: /* 1eme*/
1.238 brouard 8310: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
1.337 brouard 8311: /* for (k1=1; k1<= m ; k1 ++){ /\* For each valid combination of covariate *\/ */
1.236 brouard 8312: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8313: k1=TKresult[nres];
1.338 brouard 8314: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.238 brouard 8315: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.337 brouard 8316: /* if(m != 1 && TKresult[nres]!= k1) */
8317: /* continue; */
1.238 brouard 8318: /* We are interested in selected combination by the resultline */
1.246 brouard 8319: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 8320: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 8321: strcpy(gplotlabel,"(");
1.337 brouard 8322: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8323: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8324: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8325:
8326: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate k get corresponding value lv for combination k1 *\/ */
8327: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the value of the covariate corresponding to k1 combination *\\/ *\/ */
8328: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8329: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8330: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8331: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8332: /* vlv= nbcode[Tvaraff[k]][lv]; /\* vlv is the value of the covariate lv, 0 or 1 *\/ */
8333: /* /\* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv *\/ */
8334: /* /\* printf(" V%d=%d ",Tvaraff[k],vlv); *\/ */
8335: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8336: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8337: /* } */
8338: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8339: /* /\* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); *\/ */
8340: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8341: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.264 brouard 8342: }
8343: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 8344: /* printf("\n#\n"); */
1.238 brouard 8345: fprintf(ficgp,"\n#\n");
8346: if(invalidvarcomb[k1]){
1.260 brouard 8347: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 8348: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8349: continue;
8350: }
1.235 brouard 8351:
1.241 brouard 8352: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
8353: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 8354: /* 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 8355: fprintf(ficgp,"set title \"Alive state %d %s model=1+age+%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 8356: 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);
8357: /* 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); */
8358: /* k1-1 error should be nres-1*/
1.238 brouard 8359: for (i=1; i<= nlstate ; i ++) {
8360: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8361: else fprintf(ficgp," %%*lf (%%*lf)");
8362: }
1.288 brouard 8363: 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 8364: for (i=1; i<= nlstate ; i ++) {
8365: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8366: else fprintf(ficgp," %%*lf (%%*lf)");
8367: }
1.260 brouard 8368: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 8369: for (i=1; i<= nlstate ; i ++) {
8370: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8371: else fprintf(ficgp," %%*lf (%%*lf)");
8372: }
1.265 brouard 8373: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
8374:
8375: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
8376: if(cptcoveff ==0){
1.271 brouard 8377: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 8378: }else{
8379: kl=0;
8380: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8381: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8382: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.265 brouard 8383: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8384: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8385: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8386: vlv= nbcode[Tvaraff[k]][lv];
8387: kl++;
8388: /* 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 *\/ */
8389: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8390: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8391: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8392: if(k==cptcoveff){
8393: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
8394: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
8395: }else{
8396: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
8397: kl++;
8398: }
8399: } /* end covariate */
8400: } /* end if no covariate */
8401:
1.296 brouard 8402: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 8403: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 8404: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 8405: if(cptcoveff ==0){
1.245 brouard 8406: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 8407: }else{
8408: kl=0;
8409: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8410: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8411: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 8412: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8413: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8414: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8415: /* vlv= nbcode[Tvaraff[k]][lv]; */
8416: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 8417: kl++;
1.238 brouard 8418: /* 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 *\/ */
8419: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8420: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8421: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8422: if(k==cptcoveff){
1.245 brouard 8423: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 8424: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 8425: }else{
1.332 brouard 8426: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]);
1.238 brouard 8427: kl++;
8428: }
8429: } /* end covariate */
8430: } /* end if no covariate */
1.296 brouard 8431: if(prevbcast == 1){
1.268 brouard 8432: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
8433: /* k1-1 error should be nres-1*/
8434: for (i=1; i<= nlstate ; i ++) {
8435: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8436: else fprintf(ficgp," %%*lf (%%*lf)");
8437: }
1.271 brouard 8438: 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 8439: for (i=1; i<= nlstate ; i ++) {
8440: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8441: else fprintf(ficgp," %%*lf (%%*lf)");
8442: }
1.276 brouard 8443: 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 8444: for (i=1; i<= nlstate ; i ++) {
8445: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8446: else fprintf(ficgp," %%*lf (%%*lf)");
8447: }
1.274 brouard 8448: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 8449: } /* end if backprojcast */
1.296 brouard 8450: } /* end if prevbcast */
1.276 brouard 8451: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
8452: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 8453: } /* nres */
1.337 brouard 8454: /* } /\* k1 *\/ */
1.201 brouard 8455: } /* cpt */
1.235 brouard 8456:
8457:
1.126 brouard 8458: /*2 eme*/
1.337 brouard 8459: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8460: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8461: k1=TKresult[nres];
1.338 brouard 8462: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8463: /* if(m != 1 && TKresult[nres]!= k1) */
8464: /* continue; */
1.238 brouard 8465: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 8466: strcpy(gplotlabel,"(");
1.337 brouard 8467: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8468: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8469: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8470: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8471: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8472: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8473: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8474: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8475: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8476: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8477: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8478: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8479: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8480: /* } */
8481: /* /\* for(k=1; k <= ncovds; k++){ *\/ */
8482: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8483: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8484: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8485: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 8486: }
1.264 brouard 8487: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8488: fprintf(ficgp,"\n#\n");
1.223 brouard 8489: if(invalidvarcomb[k1]){
8490: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8491: continue;
8492: }
1.219 brouard 8493:
1.241 brouard 8494: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 8495: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 8496: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
8497: if(vpopbased==0){
1.238 brouard 8498: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 8499: }else
1.238 brouard 8500: fprintf(ficgp,"\nreplot ");
8501: for (i=1; i<= nlstate+1 ; i ++) {
8502: k=2*i;
1.261 brouard 8503: 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);
1.238 brouard 8504: for (j=1; j<= nlstate+1 ; j ++) {
8505: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8506: else fprintf(ficgp," %%*lf (%%*lf)");
8507: }
8508: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
8509: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 8510: 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 8511: for (j=1; j<= nlstate+1 ; j ++) {
8512: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8513: else fprintf(ficgp," %%*lf (%%*lf)");
8514: }
8515: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 8516: 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 8517: for (j=1; j<= nlstate+1 ; j ++) {
8518: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8519: else fprintf(ficgp," %%*lf (%%*lf)");
8520: }
8521: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
8522: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
8523: } /* state */
8524: } /* vpopbased */
1.264 brouard 8525: 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 8526: } /* end nres */
1.337 brouard 8527: /* } /\* k1 end 2 eme*\/ */
1.238 brouard 8528:
8529:
8530: /*3eme*/
1.337 brouard 8531: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8532: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8533: k1=TKresult[nres];
1.338 brouard 8534: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8535: /* if(m != 1 && TKresult[nres]!= k1) */
8536: /* continue; */
1.238 brouard 8537:
1.332 brouard 8538: for (cpt=1; cpt<= nlstate ; cpt ++) { /* Fragile no verification of covariate values */
1.261 brouard 8539: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 8540: strcpy(gplotlabel,"(");
1.337 brouard 8541: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8542: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8543: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8544: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8545: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8546: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8547: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8548: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8549: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8550: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8551: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8552: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8553: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8554: /* } */
8555: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8556: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
8557: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
8558: }
1.264 brouard 8559: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8560: fprintf(ficgp,"\n#\n");
8561: if(invalidvarcomb[k1]){
8562: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8563: continue;
8564: }
8565:
8566: /* k=2+nlstate*(2*cpt-2); */
8567: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 8568: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 8569: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 8570: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 8571: 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 8572: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8573: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8574: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
8575: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8576: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8577: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 8578:
1.238 brouard 8579: */
8580: for (i=1; i< nlstate ; i ++) {
1.261 brouard 8581: 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 8582: /* 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 8583:
1.238 brouard 8584: }
1.261 brouard 8585: 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 8586: }
1.264 brouard 8587: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 8588: } /* end nres */
1.337 brouard 8589: /* } /\* end kl 3eme *\/ */
1.126 brouard 8590:
1.223 brouard 8591: /* 4eme */
1.201 brouard 8592: /* Survival functions (period) from state i in state j by initial state i */
1.337 brouard 8593: /* for (k1=1; k1<=m; k1++){ /\* For each covariate and each value *\/ */
1.238 brouard 8594: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8595: k1=TKresult[nres];
1.338 brouard 8596: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8597: /* if(m != 1 && TKresult[nres]!= k1) */
8598: /* continue; */
1.238 brouard 8599: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 8600: strcpy(gplotlabel,"(");
1.337 brouard 8601: fprintf(ficgp,"\n#\n#\n# Survival functions in state %d : 'LIJ_' files, cov=%d state=%d", cpt, k1, cpt);
8602: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8603: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8604: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8605: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8606: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8607: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8608: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8609: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8610: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8611: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8612: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8613: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8614: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8615: /* } */
8616: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8617: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8618: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8619: }
1.264 brouard 8620: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8621: fprintf(ficgp,"\n#\n");
8622: if(invalidvarcomb[k1]){
8623: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8624: continue;
1.223 brouard 8625: }
1.238 brouard 8626:
1.241 brouard 8627: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 8628: 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 8629: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8630: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8631: k=3;
8632: for (i=1; i<= nlstate ; i ++){
8633: if(i==1){
8634: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8635: }else{
8636: fprintf(ficgp,", '' ");
8637: }
8638: l=(nlstate+ndeath)*(i-1)+1;
8639: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8640: for (j=2; j<= nlstate+ndeath ; j ++)
8641: fprintf(ficgp,"+$%d",k+l+j-1);
8642: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
8643: } /* nlstate */
1.264 brouard 8644: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8645: } /* end cpt state*/
8646: } /* end nres */
1.337 brouard 8647: /* } /\* end covariate k1 *\/ */
1.238 brouard 8648:
1.220 brouard 8649: /* 5eme */
1.201 brouard 8650: /* Survival functions (period) from state i in state j by final state j */
1.337 brouard 8651: /* for (k1=1; k1<= m ; k1++){ /\* For each covariate combination if any *\/ */
1.238 brouard 8652: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8653: k1=TKresult[nres];
1.338 brouard 8654: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8655: /* if(m != 1 && TKresult[nres]!= k1) */
8656: /* continue; */
1.238 brouard 8657: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 8658: strcpy(gplotlabel,"(");
1.238 brouard 8659: 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 8660: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8661: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8662: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8663: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8664: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8665: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8666: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8667: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8668: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8669: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8670: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8671: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8672: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8673: /* } */
8674: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8675: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8676: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8677: }
1.264 brouard 8678: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8679: fprintf(ficgp,"\n#\n");
8680: if(invalidvarcomb[k1]){
8681: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8682: continue;
8683: }
1.227 brouard 8684:
1.241 brouard 8685: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 8686: 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 8687: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8688: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8689: k=3;
8690: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8691: if(j==1)
8692: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8693: else
8694: fprintf(ficgp,", '' ");
8695: l=(nlstate+ndeath)*(cpt-1) +j;
8696: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
8697: /* for (i=2; i<= nlstate+ndeath ; i ++) */
8698: /* fprintf(ficgp,"+$%d",k+l+i-1); */
8699: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
8700: } /* nlstate */
8701: fprintf(ficgp,", '' ");
8702: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
8703: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8704: l=(nlstate+ndeath)*(cpt-1) +j;
8705: if(j < nlstate)
8706: fprintf(ficgp,"$%d +",k+l);
8707: else
8708: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
8709: }
1.264 brouard 8710: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8711: } /* end cpt state*/
1.337 brouard 8712: /* } /\* end covariate *\/ */
1.238 brouard 8713: } /* end nres */
1.227 brouard 8714:
1.220 brouard 8715: /* 6eme */
1.202 brouard 8716: /* CV preval stable (period) for each covariate */
1.337 brouard 8717: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8718: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8719: k1=TKresult[nres];
1.338 brouard 8720: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8721: /* if(m != 1 && TKresult[nres]!= k1) */
8722: /* continue; */
1.255 brouard 8723: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 8724: strcpy(gplotlabel,"(");
1.288 brouard 8725: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8726: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8727: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8728: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8729: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8730: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8731: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8732: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8733: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8734: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8735: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8736: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8737: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8738: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8739: /* } */
8740: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8741: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8742: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8743: }
1.264 brouard 8744: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8745: fprintf(ficgp,"\n#\n");
1.223 brouard 8746: if(invalidvarcomb[k1]){
1.227 brouard 8747: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8748: continue;
1.223 brouard 8749: }
1.227 brouard 8750:
1.241 brouard 8751: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 8752: 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 8753: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8754: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 8755: k=3; /* Offset */
1.255 brouard 8756: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8757: if(i==1)
8758: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8759: else
8760: fprintf(ficgp,", '' ");
1.255 brouard 8761: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8762: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8763: for (j=2; j<= nlstate ; j ++)
8764: fprintf(ficgp,"+$%d",k+l+j-1);
8765: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8766: } /* nlstate */
1.264 brouard 8767: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8768: } /* end cpt state*/
8769: } /* end covariate */
1.227 brouard 8770:
8771:
1.220 brouard 8772: /* 7eme */
1.296 brouard 8773: if(prevbcast == 1){
1.288 brouard 8774: /* CV backward prevalence for each covariate */
1.337 brouard 8775: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8776: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8777: k1=TKresult[nres];
1.338 brouard 8778: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8779: /* if(m != 1 && TKresult[nres]!= k1) */
8780: /* continue; */
1.268 brouard 8781: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8782: strcpy(gplotlabel,"(");
1.288 brouard 8783: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8784: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8785: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8786: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8787: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8788: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8789: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8790: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8791: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8792: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8793: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8794: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8795: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8796: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8797: /* } */
8798: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8799: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8800: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8801: }
1.264 brouard 8802: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8803: fprintf(ficgp,"\n#\n");
8804: if(invalidvarcomb[k1]){
8805: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8806: continue;
8807: }
8808:
1.241 brouard 8809: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8810: 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 8811: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8812: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8813: k=3; /* Offset */
1.268 brouard 8814: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8815: if(i==1)
8816: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8817: else
8818: fprintf(ficgp,", '' ");
8819: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8820: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8821: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8822: /* 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 8823: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8824: /* for (j=2; j<= nlstate ; j ++) */
8825: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8826: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8827: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8828: } /* nlstate */
1.264 brouard 8829: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8830: } /* end cpt state*/
8831: } /* end covariate */
1.296 brouard 8832: } /* End if prevbcast */
1.218 brouard 8833:
1.223 brouard 8834: /* 8eme */
1.218 brouard 8835: if(prevfcast==1){
1.288 brouard 8836: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8837:
1.337 brouard 8838: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8839: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8840: k1=TKresult[nres];
1.338 brouard 8841: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8842: /* if(m != 1 && TKresult[nres]!= k1) */
8843: /* continue; */
1.211 brouard 8844: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8845: strcpy(gplotlabel,"(");
1.288 brouard 8846: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8847: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8848: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8849: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8850: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
8851: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
8852: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8853: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8854: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8855: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8856: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8857: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8858: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8859: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8860: /* } */
8861: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8862: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8863: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8864: }
1.264 brouard 8865: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8866: fprintf(ficgp,"\n#\n");
8867: if(invalidvarcomb[k1]){
8868: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8869: continue;
8870: }
8871:
8872: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8873: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8874: 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 8875: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8876: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8877:
8878: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8879: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8880: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8881: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8882: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8883: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8884: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8885: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8886: if(i==istart){
1.227 brouard 8887: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8888: }else{
8889: fprintf(ficgp,",\\\n '' ");
8890: }
8891: if(cptcoveff ==0){ /* No covariate */
8892: ioffset=2; /* Age is in 2 */
8893: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8894: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8895: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8896: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8897: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8898: if(i==nlstate+1){
1.270 brouard 8899: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8900: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8901: fprintf(ficgp,",\\\n '' ");
8902: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8903: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8904: offyear, \
1.268 brouard 8905: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8906: }else
1.227 brouard 8907: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8908: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8909: }else{ /* more than 2 covariates */
1.270 brouard 8910: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8911: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8912: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8913: iyearc=ioffset-1;
8914: iagec=ioffset;
1.227 brouard 8915: fprintf(ficgp," u %d:(",ioffset);
8916: kl=0;
8917: strcpy(gplotcondition,"(");
8918: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8919: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8920: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8921: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8922: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8923: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8924: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8925: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8926: kl++;
8927: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8928: kl++;
8929: if(k <cptcoveff && cptcoveff>1)
8930: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8931: }
8932: strcpy(gplotcondition+strlen(gplotcondition),")");
8933: /* 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 *\/ */
8934: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8935: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8936: /* '' 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*/
8937: if(i==nlstate+1){
1.270 brouard 8938: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8939: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8940: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8941: fprintf(ficgp," u %d:(",iagec);
8942: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8943: iyearc, iagec, offyear, \
8944: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8945: /* '' 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 8946: }else{
8947: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8948: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8949: }
8950: } /* end if covariate */
8951: } /* nlstate */
1.264 brouard 8952: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8953: } /* end cpt state*/
8954: } /* end covariate */
8955: } /* End if prevfcast */
1.227 brouard 8956:
1.296 brouard 8957: if(prevbcast==1){
1.268 brouard 8958: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8959:
1.337 brouard 8960: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.268 brouard 8961: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8962: k1=TKresult[nres];
1.338 brouard 8963: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8964: /* if(m != 1 && TKresult[nres]!= k1) */
8965: /* continue; */
1.268 brouard 8966: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8967: strcpy(gplotlabel,"(");
8968: 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 8969: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8970: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8971: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8972: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
8973: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
8974: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8975: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8976: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8977: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8978: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8979: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8980: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8981: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8982: /* } */
8983: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8984: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8985: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.268 brouard 8986: }
8987: strcpy(gplotlabel+strlen(gplotlabel),")");
8988: fprintf(ficgp,"\n#\n");
8989: if(invalidvarcomb[k1]){
8990: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8991: continue;
8992: }
8993:
8994: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8995: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8996: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8997: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8998: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8999:
9000: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
9001: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
9002: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
9003: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
9004: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
9005: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
9006: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
9007: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
9008: if(i==istart){
9009: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
9010: }else{
9011: fprintf(ficgp,",\\\n '' ");
9012: }
9013: if(cptcoveff ==0){ /* No covariate */
9014: ioffset=2; /* Age is in 2 */
9015: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
9016: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
9017: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
9018: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
9019: fprintf(ficgp," u %d:(", ioffset);
9020: if(i==nlstate+1){
1.270 brouard 9021: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 9022: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
9023: fprintf(ficgp,",\\\n '' ");
9024: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 9025: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 9026: offbyear, \
9027: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
9028: }else
9029: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
9030: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
9031: }else{ /* more than 2 covariates */
1.270 brouard 9032: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
9033: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
9034: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
9035: iyearc=ioffset-1;
9036: iagec=ioffset;
1.268 brouard 9037: fprintf(ficgp," u %d:(",ioffset);
9038: kl=0;
9039: strcpy(gplotcondition,"(");
1.337 brouard 9040: for (k=1; k<=cptcovs; k++){ /* For each covariate k of the resultline, get corresponding value lv for combination k1 */
1.338 brouard 9041: if(Dummy[modelresult[nres][k]]==0){ /* To be verified */
1.337 brouard 9042: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate writing the chain of conditions *\/ */
9043: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
9044: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
9045: lv=Tvresult[nres][k];
9046: vlv=TinvDoQresult[nres][Tvresult[nres][k]];
9047: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
9048: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
9049: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
9050: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
9051: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
9052: kl++;
9053: /* sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]); */
9054: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%lg " ,kl,Tvresult[nres][k], kl+1,TinvDoQresult[nres][Tvresult[nres][k]]);
9055: kl++;
1.338 brouard 9056: if(k <cptcovs && cptcovs>1)
1.337 brouard 9057: sprintf(gplotcondition+strlen(gplotcondition)," && ");
9058: }
1.268 brouard 9059: }
9060: strcpy(gplotcondition+strlen(gplotcondition),")");
9061: /* 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 *\/ */
9062: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
9063: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
9064: /* '' 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*/
9065: if(i==nlstate+1){
1.270 brouard 9066: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
9067: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 9068: fprintf(ficgp,",\\\n '' ");
1.270 brouard 9069: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 9070: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 9071: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
9072: iyearc,iagec,offbyear, \
9073: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 9074: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
9075: }else{
9076: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
9077: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
9078: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
9079: }
9080: } /* end if covariate */
9081: } /* nlstate */
9082: fprintf(ficgp,"\nset out; unset label;\n");
9083: } /* end cpt state*/
9084: } /* end covariate */
1.296 brouard 9085: } /* End if prevbcast */
1.268 brouard 9086:
1.227 brouard 9087:
1.238 brouard 9088: /* 9eme writing MLE parameters */
9089: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 9090: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 9091: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 9092: for(k=1; k <=(nlstate+ndeath); k++){
9093: if (k != i) {
1.227 brouard 9094: fprintf(ficgp,"# current state %d\n",k);
9095: for(j=1; j <=ncovmodel; j++){
9096: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
9097: jk++;
9098: }
9099: fprintf(ficgp,"\n");
1.126 brouard 9100: }
9101: }
1.223 brouard 9102: }
1.187 brouard 9103: fprintf(ficgp,"##############\n#\n");
1.227 brouard 9104:
1.145 brouard 9105: /*goto avoid;*/
1.238 brouard 9106: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
9107: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 9108: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
9109: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
9110: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
9111: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
9112: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9113: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
9114: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9115: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
9116: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
9117: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9118: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
9119: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
9120: fprintf(ficgp,"#\n");
1.223 brouard 9121: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 9122: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.338 brouard 9123: fprintf(ficgp,"#model=1+age+%s \n",model);
1.238 brouard 9124: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 9125: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
1.337 brouard 9126: /* for(k1=1; k1 <=m; k1++) /\* For each combination of covariate *\/ */
1.237 brouard 9127: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 9128: /* k1=nres; */
1.338 brouard 9129: k1=TKresult[nres];
9130: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 9131: fprintf(ficgp,"\n\n# Resultline k1=%d ",k1);
1.264 brouard 9132: strcpy(gplotlabel,"(");
1.276 brouard 9133: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.337 brouard 9134: for (k=1; k<=cptcovs; k++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
9135: /* for each resultline nres, and position k, Tvresult[nres][k] gives the name of the variable and
9136: TinvDoQresult[nres][Tvresult[nres][k]] gives its value double or integer) */
9137: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
9138: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
9139: }
9140: /* if(m != 1 && TKresult[nres]!= k1) */
9141: /* continue; */
9142: /* fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1); */
9143: /* strcpy(gplotlabel,"("); */
9144: /* /\*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*\/ */
9145: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
9146: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
9147: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
9148: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
9149: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
9150: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
9151: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
9152: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
9153: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
9154: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
9155: /* } */
9156: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
9157: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
9158: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
9159: /* } */
1.264 brouard 9160: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 9161: fprintf(ficgp,"\n#\n");
1.264 brouard 9162: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 9163: fprintf(ficgp,"\nset key outside ");
9164: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
9165: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 9166: fprintf(ficgp,"\nset ter svg size 640, 480 ");
9167: if (ng==1){
9168: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
9169: fprintf(ficgp,"\nunset log y");
9170: }else if (ng==2){
9171: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
9172: fprintf(ficgp,"\nset log y");
9173: }else if (ng==3){
9174: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
9175: fprintf(ficgp,"\nset log y");
9176: }else
9177: fprintf(ficgp,"\nunset title ");
9178: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
9179: i=1;
9180: for(k2=1; k2<=nlstate; k2++) {
9181: k3=i;
9182: for(k=1; k<=(nlstate+ndeath); k++) {
9183: if (k != k2){
9184: switch( ng) {
9185: case 1:
9186: if(nagesqr==0)
9187: fprintf(ficgp," p%d+p%d*x",i,i+1);
9188: else /* nagesqr =1 */
9189: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
9190: break;
9191: case 2: /* ng=2 */
9192: if(nagesqr==0)
9193: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
9194: else /* nagesqr =1 */
9195: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
9196: break;
9197: case 3:
9198: if(nagesqr==0)
9199: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
9200: else /* nagesqr =1 */
9201: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
9202: break;
9203: }
9204: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 9205: ijp=1; /* product no age */
9206: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
9207: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 9208: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 9209: switch(Typevar[j]){
9210: case 1:
9211: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
9212: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
9213: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
9214: if(DummyV[j]==0){/* Bug valgrind */
9215: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
9216: }else{ /* quantitative */
9217: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
9218: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9219: }
9220: ij++;
1.268 brouard 9221: }
1.237 brouard 9222: }
1.329 brouard 9223: }
9224: break;
9225: case 2:
9226: if(cptcovprod >0){
9227: if(j==Tprod[ijp]) { /* */
9228: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
9229: if(ijp <=cptcovprod) { /* Product */
9230: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
9231: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
9232: /* 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)]); */
9233: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
9234: }else{ /* Vn is dummy and Vm is quanti */
9235: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
9236: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9237: }
9238: }else{ /* Vn*Vm Vn is quanti */
9239: if(DummyV[Tvard[ijp][2]]==0){
9240: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
9241: }else{ /* Both quanti */
9242: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9243: }
1.268 brouard 9244: }
1.329 brouard 9245: ijp++;
1.237 brouard 9246: }
1.329 brouard 9247: } /* end Tprod */
9248: }
9249: break;
9250: case 0:
9251: /* simple covariate */
1.264 brouard 9252: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 9253: if(Dummy[j]==0){
9254: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
9255: }else{ /* quantitative */
9256: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 9257: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 9258: }
1.329 brouard 9259: /* end simple */
9260: break;
9261: default:
9262: break;
9263: } /* end switch */
1.237 brouard 9264: } /* end j */
1.329 brouard 9265: }else{ /* k=k2 */
9266: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
9267: fprintf(ficgp," (1.");i=i-ncovmodel;
9268: }else
9269: i=i-ncovmodel;
1.223 brouard 9270: }
1.227 brouard 9271:
1.223 brouard 9272: if(ng != 1){
9273: fprintf(ficgp,")/(1");
1.227 brouard 9274:
1.264 brouard 9275: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 9276: if(nagesqr==0)
1.264 brouard 9277: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 9278: else /* nagesqr =1 */
1.264 brouard 9279: 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 9280:
1.223 brouard 9281: ij=1;
1.329 brouard 9282: ijp=1;
9283: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
9284: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
9285: switch(Typevar[j]){
9286: case 1:
9287: if(cptcovage >0){
9288: if(j==Tage[ij]) { /* Bug valgrind */
9289: if(ij <=cptcovage) { /* Bug valgrind */
9290: if(DummyV[j]==0){/* Bug valgrind */
9291: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
9292: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
9293: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
9294: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
9295: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9296: }else{ /* quantitative */
9297: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
9298: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
9299: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
9300: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9301: }
9302: ij++;
9303: }
9304: }
9305: }
9306: break;
9307: case 2:
9308: if(cptcovprod >0){
9309: if(j==Tprod[ijp]) { /* */
9310: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
9311: if(ijp <=cptcovprod) { /* Product */
9312: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
9313: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
9314: /* 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)]); */
9315: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
9316: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
9317: }else{ /* Vn is dummy and Vm is quanti */
9318: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
9319: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9320: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
9321: }
9322: }else{ /* Vn*Vm Vn is quanti */
9323: if(DummyV[Tvard[ijp][2]]==0){
9324: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
9325: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
9326: }else{ /* Both quanti */
9327: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9328: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
9329: }
9330: }
9331: ijp++;
9332: }
9333: } /* end Tprod */
9334: } /* end if */
9335: break;
9336: case 0:
9337: /* simple covariate */
9338: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
9339: if(Dummy[j]==0){
9340: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
9341: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
9342: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
9343: }else{ /* quantitative */
9344: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
9345: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
9346: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9347: }
9348: /* end simple */
9349: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
9350: break;
9351: default:
9352: break;
9353: } /* end switch */
1.223 brouard 9354: }
9355: fprintf(ficgp,")");
9356: }
9357: fprintf(ficgp,")");
9358: if(ng ==2)
1.276 brouard 9359: 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 9360: else /* ng= 3 */
1.276 brouard 9361: 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 9362: }else{ /* end ng <> 1 */
1.223 brouard 9363: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 9364: 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 9365: }
9366: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
9367: fprintf(ficgp,",");
9368: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
9369: fprintf(ficgp,",");
9370: i=i+ncovmodel;
9371: } /* end k */
9372: } /* end k2 */
1.276 brouard 9373: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
9374: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.337 brouard 9375: } /* end resultline */
1.223 brouard 9376: } /* end ng */
9377: /* avoid: */
9378: fflush(ficgp);
1.126 brouard 9379: } /* end gnuplot */
9380:
9381:
9382: /*************** Moving average **************/
1.219 brouard 9383: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 9384: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 9385:
1.222 brouard 9386: int i, cpt, cptcod;
9387: int modcovmax =1;
9388: int mobilavrange, mob;
9389: int iage=0;
1.288 brouard 9390: int firstA1=0, firstA2=0;
1.222 brouard 9391:
1.266 brouard 9392: double sum=0., sumr=0.;
1.222 brouard 9393: double age;
1.266 brouard 9394: double *sumnewp, *sumnewm, *sumnewmr;
9395: double *agemingood, *agemaxgood;
9396: double *agemingoodr, *agemaxgoodr;
1.222 brouard 9397:
9398:
1.278 brouard 9399: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
9400: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 9401:
9402: sumnewp = vector(1,ncovcombmax);
9403: sumnewm = vector(1,ncovcombmax);
1.266 brouard 9404: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 9405: agemingood = vector(1,ncovcombmax);
1.266 brouard 9406: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 9407: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 9408: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 9409:
9410: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 9411: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 9412: sumnewp[cptcod]=0.;
1.266 brouard 9413: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
9414: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 9415: }
9416: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
9417:
1.266 brouard 9418: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
9419: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 9420: else mobilavrange=mobilav;
9421: for (age=bage; age<=fage; age++)
9422: for (i=1; i<=nlstate;i++)
9423: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
9424: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9425: /* We keep the original values on the extreme ages bage, fage and for
9426: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
9427: we use a 5 terms etc. until the borders are no more concerned.
9428: */
9429: for (mob=3;mob <=mobilavrange;mob=mob+2){
9430: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 9431: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
9432: sumnewm[cptcod]=0.;
9433: for (i=1; i<=nlstate;i++){
1.222 brouard 9434: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
9435: for (cpt=1;cpt<=(mob-1)/2;cpt++){
9436: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
9437: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
9438: }
9439: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 9440: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9441: } /* end i */
9442: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
9443: } /* end cptcod */
1.222 brouard 9444: }/* end age */
9445: }/* end mob */
1.266 brouard 9446: }else{
9447: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 9448: return -1;
1.266 brouard 9449: }
9450:
9451: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 9452: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
9453: if(invalidvarcomb[cptcod]){
9454: printf("\nCombination (%d) ignored because no cases \n",cptcod);
9455: continue;
9456: }
1.219 brouard 9457:
1.266 brouard 9458: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
9459: sumnewm[cptcod]=0.;
9460: sumnewmr[cptcod]=0.;
9461: for (i=1; i<=nlstate;i++){
9462: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9463: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9464: }
9465: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9466: agemingoodr[cptcod]=age;
9467: }
9468: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9469: agemingood[cptcod]=age;
9470: }
9471: } /* age */
9472: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 9473: sumnewm[cptcod]=0.;
1.266 brouard 9474: sumnewmr[cptcod]=0.;
1.222 brouard 9475: for (i=1; i<=nlstate;i++){
9476: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9477: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9478: }
9479: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9480: agemaxgoodr[cptcod]=age;
1.222 brouard 9481: }
9482: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 9483: agemaxgood[cptcod]=age;
9484: }
9485: } /* age */
9486: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
9487: /* but they will change */
1.288 brouard 9488: firstA1=0;firstA2=0;
1.266 brouard 9489: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
9490: sumnewm[cptcod]=0.;
9491: sumnewmr[cptcod]=0.;
9492: for (i=1; i<=nlstate;i++){
9493: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9494: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9495: }
9496: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9497: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9498: agemaxgoodr[cptcod]=age; /* age min */
9499: for (i=1; i<=nlstate;i++)
9500: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9501: }else{ /* bad we change the value with the values of good ages */
9502: for (i=1; i<=nlstate;i++){
9503: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
9504: } /* i */
9505: } /* end bad */
9506: }else{
9507: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9508: agemaxgood[cptcod]=age;
9509: }else{ /* bad we change the value with the values of good ages */
9510: for (i=1; i<=nlstate;i++){
9511: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
9512: } /* i */
9513: } /* end bad */
9514: }/* end else */
9515: sum=0.;sumr=0.;
9516: for (i=1; i<=nlstate;i++){
9517: sum+=mobaverage[(int)age][i][cptcod];
9518: sumr+=probs[(int)age][i][cptcod];
9519: }
9520: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 9521: if(!firstA1){
9522: firstA1=1;
9523: 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);
9524: }
9525: 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 9526: } /* end bad */
9527: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9528: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 9529: if(!firstA2){
9530: firstA2=1;
9531: 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);
9532: }
9533: 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 9534: } /* end bad */
9535: }/* age */
1.266 brouard 9536:
9537: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 9538: sumnewm[cptcod]=0.;
1.266 brouard 9539: sumnewmr[cptcod]=0.;
1.222 brouard 9540: for (i=1; i<=nlstate;i++){
9541: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9542: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9543: }
9544: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9545: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
9546: agemingoodr[cptcod]=age;
9547: for (i=1; i<=nlstate;i++)
9548: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9549: }else{ /* bad we change the value with the values of good ages */
9550: for (i=1; i<=nlstate;i++){
9551: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
9552: } /* i */
9553: } /* end bad */
9554: }else{
9555: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9556: agemingood[cptcod]=age;
9557: }else{ /* bad */
9558: for (i=1; i<=nlstate;i++){
9559: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
9560: } /* i */
9561: } /* end bad */
9562: }/* end else */
9563: sum=0.;sumr=0.;
9564: for (i=1; i<=nlstate;i++){
9565: sum+=mobaverage[(int)age][i][cptcod];
9566: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 9567: }
1.266 brouard 9568: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 9569: 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 9570: } /* end bad */
9571: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9572: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 9573: 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 9574: } /* end bad */
9575: }/* age */
1.266 brouard 9576:
1.222 brouard 9577:
9578: for (age=bage; age<=fage; age++){
1.235 brouard 9579: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 9580: sumnewp[cptcod]=0.;
9581: sumnewm[cptcod]=0.;
9582: for (i=1; i<=nlstate;i++){
9583: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
9584: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9585: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
9586: }
9587: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
9588: }
9589: /* printf("\n"); */
9590: /* } */
1.266 brouard 9591:
1.222 brouard 9592: /* brutal averaging */
1.266 brouard 9593: /* for (i=1; i<=nlstate;i++){ */
9594: /* for (age=1; age<=bage; age++){ */
9595: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
9596: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9597: /* } */
9598: /* for (age=fage; age<=AGESUP; age++){ */
9599: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
9600: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9601: /* } */
9602: /* } /\* end i status *\/ */
9603: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
9604: /* for (age=1; age<=AGESUP; age++){ */
9605: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
9606: /* mobaverage[(int)age][i][cptcod]=0.; */
9607: /* } */
9608: /* } */
1.222 brouard 9609: }/* end cptcod */
1.266 brouard 9610: free_vector(agemaxgoodr,1, ncovcombmax);
9611: free_vector(agemaxgood,1, ncovcombmax);
9612: free_vector(agemingood,1, ncovcombmax);
9613: free_vector(agemingoodr,1, ncovcombmax);
9614: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 9615: free_vector(sumnewm,1, ncovcombmax);
9616: free_vector(sumnewp,1, ncovcombmax);
9617: return 0;
9618: }/* End movingaverage */
1.218 brouard 9619:
1.126 brouard 9620:
1.296 brouard 9621:
1.126 brouard 9622: /************** Forecasting ******************/
1.296 brouard 9623: /* 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)*/
9624: 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){
9625: /* dateintemean, mean date of interviews
9626: dateprojd, year, month, day of starting projection
9627: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 9628: agemin, agemax range of age
9629: dateprev1 dateprev2 range of dates during which prevalence is computed
9630: */
1.296 brouard 9631: /* double anprojd, mprojd, jprojd; */
9632: /* double anprojf, mprojf, jprojf; */
1.267 brouard 9633: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 9634: double agec; /* generic age */
1.296 brouard 9635: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 9636: double *popeffectif,*popcount;
9637: double ***p3mat;
1.218 brouard 9638: /* double ***mobaverage; */
1.126 brouard 9639: char fileresf[FILENAMELENGTH];
9640:
9641: agelim=AGESUP;
1.211 brouard 9642: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9643: in each health status at the date of interview (if between dateprev1 and dateprev2).
9644: We still use firstpass and lastpass as another selection.
9645: */
1.214 brouard 9646: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9647: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 9648:
1.201 brouard 9649: strcpy(fileresf,"F_");
9650: strcat(fileresf,fileresu);
1.126 brouard 9651: if((ficresf=fopen(fileresf,"w"))==NULL) {
9652: printf("Problem with forecast resultfile: %s\n", fileresf);
9653: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
9654: }
1.235 brouard 9655: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
9656: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 9657:
1.225 brouard 9658: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 9659:
9660:
9661: stepsize=(int) (stepm+YEARM-1)/YEARM;
9662: if (stepm<=12) stepsize=1;
9663: if(estepm < stepm){
9664: printf ("Problem %d lower than %d\n",estepm, stepm);
9665: }
1.270 brouard 9666: else{
9667: hstepm=estepm;
9668: }
9669: if(estepm > stepm){ /* Yes every two year */
9670: stepsize=2;
9671: }
1.296 brouard 9672: hstepm=hstepm/stepm;
1.126 brouard 9673:
1.296 brouard 9674:
9675: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9676: /* fractional in yp1 *\/ */
9677: /* aintmean=yp; */
9678: /* yp2=modf((yp1*12),&yp); */
9679: /* mintmean=yp; */
9680: /* yp1=modf((yp2*30.5),&yp); */
9681: /* jintmean=yp; */
9682: /* if(jintmean==0) jintmean=1; */
9683: /* if(mintmean==0) mintmean=1; */
1.126 brouard 9684:
1.296 brouard 9685:
9686: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
9687: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
9688: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 9689: i1=pow(2,cptcoveff);
1.126 brouard 9690: if (cptcovn < 1){i1=1;}
9691:
1.296 brouard 9692: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 9693:
9694: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 9695:
1.126 brouard 9696: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 9697: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9698: 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) */
1.253 brouard 9699: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9700: continue;
1.227 brouard 9701: if(invalidvarcomb[k]){
9702: printf("\nCombination (%d) projection ignored because no cases \n",k);
9703: continue;
9704: }
9705: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
9706: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9707: /* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); */
9708: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.227 brouard 9709: }
1.235 brouard 9710: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 9711: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 9712: }
1.227 brouard 9713: fprintf(ficresf," yearproj age");
9714: for(j=1; j<=nlstate+ndeath;j++){
9715: for(i=1; i<=nlstate;i++)
9716: fprintf(ficresf," p%d%d",i,j);
9717: fprintf(ficresf," wp.%d",j);
9718: }
1.296 brouard 9719: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 9720: fprintf(ficresf,"\n");
1.296 brouard 9721: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 9722: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
9723: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 9724: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
9725: nhstepm = nhstepm/hstepm;
9726: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9727: oldm=oldms;savm=savms;
1.268 brouard 9728: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 9729: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 9730: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 9731: for (h=0; h<=nhstepm; h++){
9732: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 9733: break;
9734: }
9735: }
9736: fprintf(ficresf,"\n");
9737: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9738: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Tvaraff not correct *\/ */
9739: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* TnsdVar[Tvaraff] correct */
1.296 brouard 9740: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 9741:
9742: for(j=1; j<=nlstate+ndeath;j++) {
9743: ppij=0.;
9744: for(i=1; i<=nlstate;i++) {
1.278 brouard 9745: if (mobilav>=1)
9746: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
9747: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
9748: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
9749: }
1.268 brouard 9750: fprintf(ficresf," %.3f", p3mat[i][j][h]);
9751: } /* end i */
9752: fprintf(ficresf," %.3f", ppij);
9753: }/* end j */
1.227 brouard 9754: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9755: } /* end agec */
1.266 brouard 9756: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
9757: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 9758: } /* end yearp */
9759: } /* end k */
1.219 brouard 9760:
1.126 brouard 9761: fclose(ficresf);
1.215 brouard 9762: printf("End of Computing forecasting \n");
9763: fprintf(ficlog,"End of Computing forecasting\n");
9764:
1.126 brouard 9765: }
9766:
1.269 brouard 9767: /************** Back Forecasting ******************/
1.296 brouard 9768: /* 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){ */
9769: 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){
9770: /* back1, year, month, day of starting backprojection
1.267 brouard 9771: agemin, agemax range of age
9772: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 9773: anback2 year of end of backprojection (same day and month as back1).
9774: prevacurrent and prev are prevalences.
1.267 brouard 9775: */
9776: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
9777: double agec; /* generic age */
1.302 brouard 9778: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 9779: double *popeffectif,*popcount;
9780: double ***p3mat;
9781: /* double ***mobaverage; */
9782: char fileresfb[FILENAMELENGTH];
9783:
1.268 brouard 9784: agelim=AGEINF;
1.267 brouard 9785: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9786: in each health status at the date of interview (if between dateprev1 and dateprev2).
9787: We still use firstpass and lastpass as another selection.
9788: */
9789: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9790: /* firstpass, lastpass, stepm, weightopt, model); */
9791:
9792: /*Do we need to compute prevalence again?*/
9793:
9794: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
9795:
9796: strcpy(fileresfb,"FB_");
9797: strcat(fileresfb,fileresu);
9798: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
9799: printf("Problem with back forecast resultfile: %s\n", fileresfb);
9800: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
9801: }
9802: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9803: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9804:
9805: if (cptcoveff==0) ncodemax[cptcoveff]=1;
9806:
9807:
9808: stepsize=(int) (stepm+YEARM-1)/YEARM;
9809: if (stepm<=12) stepsize=1;
9810: if(estepm < stepm){
9811: printf ("Problem %d lower than %d\n",estepm, stepm);
9812: }
1.270 brouard 9813: else{
9814: hstepm=estepm;
9815: }
9816: if(estepm >= stepm){ /* Yes every two year */
9817: stepsize=2;
9818: }
1.267 brouard 9819:
9820: hstepm=hstepm/stepm;
1.296 brouard 9821: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9822: /* fractional in yp1 *\/ */
9823: /* aintmean=yp; */
9824: /* yp2=modf((yp1*12),&yp); */
9825: /* mintmean=yp; */
9826: /* yp1=modf((yp2*30.5),&yp); */
9827: /* jintmean=yp; */
9828: /* if(jintmean==0) jintmean=1; */
9829: /* if(mintmean==0) jintmean=1; */
1.267 brouard 9830:
9831: i1=pow(2,cptcoveff);
9832: if (cptcovn < 1){i1=1;}
9833:
1.296 brouard 9834: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
9835: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 9836:
9837: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
9838:
9839: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9840: for(k=1; k<=i1;k++){
9841: if(i1 != 1 && TKresult[nres]!= k)
9842: continue;
9843: if(invalidvarcomb[k]){
9844: printf("\nCombination (%d) projection ignored because no cases \n",k);
9845: continue;
9846: }
1.268 brouard 9847: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 9848: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9849: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.267 brouard 9850: }
9851: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9852: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9853: }
9854: fprintf(ficresfb," yearbproj age");
9855: for(j=1; j<=nlstate+ndeath;j++){
9856: for(i=1; i<=nlstate;i++)
1.268 brouard 9857: fprintf(ficresfb," b%d%d",i,j);
9858: fprintf(ficresfb," b.%d",j);
1.267 brouard 9859: }
1.296 brouard 9860: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 9861: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
9862: fprintf(ficresfb,"\n");
1.296 brouard 9863: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 9864: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 9865: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
9866: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 9867: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 9868: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 9869: nhstepm = nhstepm/hstepm;
9870: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9871: oldm=oldms;savm=savms;
1.268 brouard 9872: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 9873: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 9874: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 9875: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
9876: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
9877: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9878: for (h=0; h<=nhstepm; h++){
1.268 brouard 9879: if (h*hstepm/YEARM*stepm ==-yearp) {
9880: break;
9881: }
9882: }
9883: fprintf(ficresfb,"\n");
9884: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9885: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.296 brouard 9886: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9887: for(i=1; i<=nlstate+ndeath;i++) {
9888: ppij=0.;ppi=0.;
9889: for(j=1; j<=nlstate;j++) {
9890: /* if (mobilav==1) */
1.269 brouard 9891: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9892: ppi=ppi+prevacurrent[(int)agec][j][k];
9893: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9894: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9895: /* else { */
9896: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9897: /* } */
1.268 brouard 9898: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9899: } /* end j */
9900: if(ppi <0.99){
9901: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9902: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9903: }
9904: fprintf(ficresfb," %.3f", ppij);
9905: }/* end j */
1.267 brouard 9906: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9907: } /* end agec */
9908: } /* end yearp */
9909: } /* end k */
1.217 brouard 9910:
1.267 brouard 9911: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9912:
1.267 brouard 9913: fclose(ficresfb);
9914: printf("End of Computing Back forecasting \n");
9915: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9916:
1.267 brouard 9917: }
1.217 brouard 9918:
1.269 brouard 9919: /* Variance of prevalence limit: varprlim */
9920: 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 9921: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9922:
9923: char fileresvpl[FILENAMELENGTH];
9924: FILE *ficresvpl;
9925: double **oldm, **savm;
9926: double **varpl; /* Variances of prevalence limits by age */
9927: int i1, k, nres, j ;
9928:
9929: strcpy(fileresvpl,"VPL_");
9930: strcat(fileresvpl,fileresu);
9931: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9932: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9933: exit(0);
9934: }
1.288 brouard 9935: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9936: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9937:
9938: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9939: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9940:
9941: i1=pow(2,cptcoveff);
9942: if (cptcovn < 1){i1=1;}
9943:
1.337 brouard 9944: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9945: k=TKresult[nres];
1.338 brouard 9946: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 9947: /* for(k=1; k<=i1;k++){ /\* We find the combination equivalent to result line values of dummies *\/ */
1.269 brouard 9948: if(i1 != 1 && TKresult[nres]!= k)
9949: continue;
9950: fprintf(ficresvpl,"\n#****** ");
9951: printf("\n#****** ");
9952: fprintf(ficlog,"\n#****** ");
1.337 brouard 9953: for(j=1;j<=cptcovs;j++) {
9954: fprintf(ficresvpl,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9955: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9956: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9957: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9958: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.269 brouard 9959: }
1.337 brouard 9960: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
9961: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9962: /* fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9963: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9964: /* } */
1.269 brouard 9965: fprintf(ficresvpl,"******\n");
9966: printf("******\n");
9967: fprintf(ficlog,"******\n");
9968:
9969: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9970: oldm=oldms;savm=savms;
9971: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9972: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9973: /*}*/
9974: }
9975:
9976: fclose(ficresvpl);
1.288 brouard 9977: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9978: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9979:
9980: }
9981: /* Variance of back prevalence: varbprlim */
9982: 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){
9983: /*------- Variance of back (stable) prevalence------*/
9984:
9985: char fileresvbl[FILENAMELENGTH];
9986: FILE *ficresvbl;
9987:
9988: double **oldm, **savm;
9989: double **varbpl; /* Variances of back prevalence limits by age */
9990: int i1, k, nres, j ;
9991:
9992: strcpy(fileresvbl,"VBL_");
9993: strcat(fileresvbl,fileresu);
9994: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9995: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9996: exit(0);
9997: }
9998: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9999: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
10000:
10001:
10002: i1=pow(2,cptcoveff);
10003: if (cptcovn < 1){i1=1;}
10004:
1.337 brouard 10005: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10006: k=TKresult[nres];
1.338 brouard 10007: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 10008: /* for(k=1; k<=i1;k++){ */
10009: /* if(i1 != 1 && TKresult[nres]!= k) */
10010: /* continue; */
1.269 brouard 10011: fprintf(ficresvbl,"\n#****** ");
10012: printf("\n#****** ");
10013: fprintf(ficlog,"\n#****** ");
1.337 brouard 10014: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */
1.338 brouard 10015: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
10016: fprintf(ficresvbl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
10017: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
1.337 brouard 10018: /* for(j=1;j<=cptcoveff;j++) { */
10019: /* fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
10020: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
10021: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
10022: /* } */
10023: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
10024: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
10025: /* fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
10026: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
1.269 brouard 10027: }
10028: fprintf(ficresvbl,"******\n");
10029: printf("******\n");
10030: fprintf(ficlog,"******\n");
10031:
10032: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
10033: oldm=oldms;savm=savms;
10034:
10035: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
10036: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
10037: /*}*/
10038: }
10039:
10040: fclose(ficresvbl);
10041: printf("done variance-covariance of back prevalence\n");fflush(stdout);
10042: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
10043:
10044: } /* End of varbprlim */
10045:
1.126 brouard 10046: /************** Forecasting *****not tested NB*************/
1.227 brouard 10047: /* 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 10048:
1.227 brouard 10049: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
10050: /* int *popage; */
10051: /* double calagedatem, agelim, kk1, kk2; */
10052: /* double *popeffectif,*popcount; */
10053: /* double ***p3mat,***tabpop,***tabpopprev; */
10054: /* /\* double ***mobaverage; *\/ */
10055: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 10056:
1.227 brouard 10057: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10058: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10059: /* agelim=AGESUP; */
10060: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 10061:
1.227 brouard 10062: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 10063:
10064:
1.227 brouard 10065: /* strcpy(filerespop,"POP_"); */
10066: /* strcat(filerespop,fileresu); */
10067: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
10068: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
10069: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
10070: /* } */
10071: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
10072: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 10073:
1.227 brouard 10074: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 10075:
1.227 brouard 10076: /* /\* if (mobilav!=0) { *\/ */
10077: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
10078: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
10079: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
10080: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
10081: /* /\* } *\/ */
10082: /* /\* } *\/ */
1.126 brouard 10083:
1.227 brouard 10084: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
10085: /* if (stepm<=12) stepsize=1; */
1.126 brouard 10086:
1.227 brouard 10087: /* agelim=AGESUP; */
1.126 brouard 10088:
1.227 brouard 10089: /* hstepm=1; */
10090: /* hstepm=hstepm/stepm; */
1.218 brouard 10091:
1.227 brouard 10092: /* if (popforecast==1) { */
10093: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
10094: /* printf("Problem with population file : %s\n",popfile);exit(0); */
10095: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
10096: /* } */
10097: /* popage=ivector(0,AGESUP); */
10098: /* popeffectif=vector(0,AGESUP); */
10099: /* popcount=vector(0,AGESUP); */
1.126 brouard 10100:
1.227 brouard 10101: /* i=1; */
10102: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 10103:
1.227 brouard 10104: /* imx=i; */
10105: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
10106: /* } */
1.218 brouard 10107:
1.227 brouard 10108: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
10109: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
10110: /* k=k+1; */
10111: /* fprintf(ficrespop,"\n#******"); */
10112: /* for(j=1;j<=cptcoveff;j++) { */
10113: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
10114: /* } */
10115: /* fprintf(ficrespop,"******\n"); */
10116: /* fprintf(ficrespop,"# Age"); */
10117: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
10118: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 10119:
1.227 brouard 10120: /* for (cpt=0; cpt<=0;cpt++) { */
10121: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 10122:
1.227 brouard 10123: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
10124: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
10125: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 10126:
1.227 brouard 10127: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10128: /* oldm=oldms;savm=savms; */
10129: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 10130:
1.227 brouard 10131: /* for (h=0; h<=nhstepm; h++){ */
10132: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
10133: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
10134: /* } */
10135: /* for(j=1; j<=nlstate+ndeath;j++) { */
10136: /* kk1=0.;kk2=0; */
10137: /* for(i=1; i<=nlstate;i++) { */
10138: /* if (mobilav==1) */
10139: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
10140: /* else { */
10141: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
10142: /* } */
10143: /* } */
10144: /* if (h==(int)(calagedatem+12*cpt)){ */
10145: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
10146: /* /\*fprintf(ficrespop," %.3f", kk1); */
10147: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
10148: /* } */
10149: /* } */
10150: /* for(i=1; i<=nlstate;i++){ */
10151: /* kk1=0.; */
10152: /* for(j=1; j<=nlstate;j++){ */
10153: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
10154: /* } */
10155: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
10156: /* } */
1.218 brouard 10157:
1.227 brouard 10158: /* if (h==(int)(calagedatem+12*cpt)) */
10159: /* for(j=1; j<=nlstate;j++) */
10160: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
10161: /* } */
10162: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10163: /* } */
10164: /* } */
1.218 brouard 10165:
1.227 brouard 10166: /* /\******\/ */
1.218 brouard 10167:
1.227 brouard 10168: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
10169: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
10170: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
10171: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
10172: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 10173:
1.227 brouard 10174: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10175: /* oldm=oldms;savm=savms; */
10176: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
10177: /* for (h=0; h<=nhstepm; h++){ */
10178: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
10179: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
10180: /* } */
10181: /* for(j=1; j<=nlstate+ndeath;j++) { */
10182: /* kk1=0.;kk2=0; */
10183: /* for(i=1; i<=nlstate;i++) { */
10184: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
10185: /* } */
10186: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
10187: /* } */
10188: /* } */
10189: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10190: /* } */
10191: /* } */
10192: /* } */
10193: /* } */
1.218 brouard 10194:
1.227 brouard 10195: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 10196:
1.227 brouard 10197: /* if (popforecast==1) { */
10198: /* free_ivector(popage,0,AGESUP); */
10199: /* free_vector(popeffectif,0,AGESUP); */
10200: /* free_vector(popcount,0,AGESUP); */
10201: /* } */
10202: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10203: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10204: /* fclose(ficrespop); */
10205: /* } /\* End of popforecast *\/ */
1.218 brouard 10206:
1.126 brouard 10207: int fileappend(FILE *fichier, char *optionfich)
10208: {
10209: if((fichier=fopen(optionfich,"a"))==NULL) {
10210: printf("Problem with file: %s\n", optionfich);
10211: fprintf(ficlog,"Problem with file: %s\n", optionfich);
10212: return (0);
10213: }
10214: fflush(fichier);
10215: return (1);
10216: }
10217:
10218:
10219: /**************** function prwizard **********************/
10220: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
10221: {
10222:
10223: /* Wizard to print covariance matrix template */
10224:
1.164 brouard 10225: char ca[32], cb[32];
10226: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 10227: int numlinepar;
10228:
10229: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10230: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10231: for(i=1; i <=nlstate; i++){
10232: jj=0;
10233: for(j=1; j <=nlstate+ndeath; j++){
10234: if(j==i) continue;
10235: jj++;
10236: /*ca[0]= k+'a'-1;ca[1]='\0';*/
10237: printf("%1d%1d",i,j);
10238: fprintf(ficparo,"%1d%1d",i,j);
10239: for(k=1; k<=ncovmodel;k++){
10240: /* printf(" %lf",param[i][j][k]); */
10241: /* fprintf(ficparo," %lf",param[i][j][k]); */
10242: printf(" 0.");
10243: fprintf(ficparo," 0.");
10244: }
10245: printf("\n");
10246: fprintf(ficparo,"\n");
10247: }
10248: }
10249: printf("# Scales (for hessian or gradient estimation)\n");
10250: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
10251: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
10252: for(i=1; i <=nlstate; i++){
10253: jj=0;
10254: for(j=1; j <=nlstate+ndeath; j++){
10255: if(j==i) continue;
10256: jj++;
10257: fprintf(ficparo,"%1d%1d",i,j);
10258: printf("%1d%1d",i,j);
10259: fflush(stdout);
10260: for(k=1; k<=ncovmodel;k++){
10261: /* printf(" %le",delti3[i][j][k]); */
10262: /* fprintf(ficparo," %le",delti3[i][j][k]); */
10263: printf(" 0.");
10264: fprintf(ficparo," 0.");
10265: }
10266: numlinepar++;
10267: printf("\n");
10268: fprintf(ficparo,"\n");
10269: }
10270: }
10271: printf("# Covariance matrix\n");
10272: /* # 121 Var(a12)\n\ */
10273: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10274: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10275: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10276: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10277: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10278: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10279: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10280: fflush(stdout);
10281: fprintf(ficparo,"# Covariance matrix\n");
10282: /* # 121 Var(a12)\n\ */
10283: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10284: /* # ...\n\ */
10285: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10286:
10287: for(itimes=1;itimes<=2;itimes++){
10288: jj=0;
10289: for(i=1; i <=nlstate; i++){
10290: for(j=1; j <=nlstate+ndeath; j++){
10291: if(j==i) continue;
10292: for(k=1; k<=ncovmodel;k++){
10293: jj++;
10294: ca[0]= k+'a'-1;ca[1]='\0';
10295: if(itimes==1){
10296: printf("#%1d%1d%d",i,j,k);
10297: fprintf(ficparo,"#%1d%1d%d",i,j,k);
10298: }else{
10299: printf("%1d%1d%d",i,j,k);
10300: fprintf(ficparo,"%1d%1d%d",i,j,k);
10301: /* printf(" %.5le",matcov[i][j]); */
10302: }
10303: ll=0;
10304: for(li=1;li <=nlstate; li++){
10305: for(lj=1;lj <=nlstate+ndeath; lj++){
10306: if(lj==li) continue;
10307: for(lk=1;lk<=ncovmodel;lk++){
10308: ll++;
10309: if(ll<=jj){
10310: cb[0]= lk +'a'-1;cb[1]='\0';
10311: if(ll<jj){
10312: if(itimes==1){
10313: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10314: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10315: }else{
10316: printf(" 0.");
10317: fprintf(ficparo," 0.");
10318: }
10319: }else{
10320: if(itimes==1){
10321: printf(" Var(%s%1d%1d)",ca,i,j);
10322: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
10323: }else{
10324: printf(" 0.");
10325: fprintf(ficparo," 0.");
10326: }
10327: }
10328: }
10329: } /* end lk */
10330: } /* end lj */
10331: } /* end li */
10332: printf("\n");
10333: fprintf(ficparo,"\n");
10334: numlinepar++;
10335: } /* end k*/
10336: } /*end j */
10337: } /* end i */
10338: } /* end itimes */
10339:
10340: } /* end of prwizard */
10341: /******************* Gompertz Likelihood ******************************/
10342: double gompertz(double x[])
10343: {
1.302 brouard 10344: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 10345: int i,n=0; /* n is the size of the sample */
10346:
1.220 brouard 10347: for (i=1;i<=imx ; i++) {
1.126 brouard 10348: sump=sump+weight[i];
10349: /* sump=sump+1;*/
10350: num=num+1;
10351: }
1.302 brouard 10352: L=0.0;
10353: /* agegomp=AGEGOMP; */
1.126 brouard 10354: /* for (i=0; i<=imx; i++)
10355: 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]);*/
10356:
1.302 brouard 10357: for (i=1;i<=imx ; i++) {
10358: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
10359: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
10360: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
10361: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
10362: * +
10363: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
10364: */
10365: if (wav[i] > 1 || agedc[i] < AGESUP) {
10366: if (cens[i] == 1){
10367: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
10368: } else if (cens[i] == 0){
1.126 brouard 10369: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 10370: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
10371: } else
10372: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 10373: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 10374: L=L+A*weight[i];
1.126 brouard 10375: /* 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 10376: }
10377: }
1.126 brouard 10378:
1.302 brouard 10379: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 10380:
10381: return -2*L*num/sump;
10382: }
10383:
1.136 brouard 10384: #ifdef GSL
10385: /******************* Gompertz_f Likelihood ******************************/
10386: double gompertz_f(const gsl_vector *v, void *params)
10387: {
1.302 brouard 10388: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 10389: double *x= (double *) v->data;
10390: int i,n=0; /* n is the size of the sample */
10391:
10392: for (i=0;i<=imx-1 ; i++) {
10393: sump=sump+weight[i];
10394: /* sump=sump+1;*/
10395: num=num+1;
10396: }
10397:
10398:
10399: /* for (i=0; i<=imx; i++)
10400: 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]);*/
10401: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
10402: for (i=1;i<=imx ; i++)
10403: {
10404: if (cens[i] == 1 && wav[i]>1)
10405: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
10406:
10407: if (cens[i] == 0 && wav[i]>1)
10408: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
10409: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
10410:
10411: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
10412: if (wav[i] > 1 ) { /* ??? */
10413: LL=LL+A*weight[i];
10414: /* 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]);*/
10415: }
10416: }
10417:
10418: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
10419: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
10420:
10421: return -2*LL*num/sump;
10422: }
10423: #endif
10424:
1.126 brouard 10425: /******************* Printing html file ***********/
1.201 brouard 10426: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 10427: int lastpass, int stepm, int weightopt, char model[],\
10428: int imx, double p[],double **matcov,double agemortsup){
10429: int i,k;
10430:
10431: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
10432: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
10433: for (i=1;i<=2;i++)
10434: 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 10435: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 10436: fprintf(fichtm,"</ul>");
10437:
10438: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
10439:
10440: 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>");
10441:
10442: for (k=agegomp;k<(agemortsup-2);k++)
10443: 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]);
10444:
10445:
10446: fflush(fichtm);
10447: }
10448:
10449: /******************* Gnuplot file **************/
1.201 brouard 10450: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 10451:
10452: char dirfileres[132],optfileres[132];
1.164 brouard 10453:
1.126 brouard 10454: int ng;
10455:
10456:
10457: /*#ifdef windows */
10458: fprintf(ficgp,"cd \"%s\" \n",pathc);
10459: /*#endif */
10460:
10461:
10462: strcpy(dirfileres,optionfilefiname);
10463: strcpy(optfileres,"vpl");
1.199 brouard 10464: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 10465: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 10466: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 10467: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 10468: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
10469:
10470: }
10471:
1.136 brouard 10472: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
10473: {
1.126 brouard 10474:
1.136 brouard 10475: /*-------- data file ----------*/
10476: FILE *fic;
10477: char dummy[]=" ";
1.240 brouard 10478: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 10479: int lstra;
1.136 brouard 10480: int linei, month, year,iout;
1.302 brouard 10481: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 10482: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 10483: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 10484: char *stratrunc;
1.223 brouard 10485:
1.240 brouard 10486: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
10487: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.328 brouard 10488: for(v=1;v<NCOVMAX;v++){
10489: DummyV[v]=0;
10490: FixedV[v]=0;
10491: }
1.126 brouard 10492:
1.240 brouard 10493: for(v=1; v <=ncovcol;v++){
10494: DummyV[v]=0;
10495: FixedV[v]=0;
10496: }
10497: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
10498: DummyV[v]=1;
10499: FixedV[v]=0;
10500: }
10501: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
10502: DummyV[v]=0;
10503: FixedV[v]=1;
10504: }
10505: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
10506: DummyV[v]=1;
10507: FixedV[v]=1;
10508: }
10509: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
10510: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
10511: fprintf(ficlog,"Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
10512: }
1.339 brouard 10513:
10514: ncovcolt=ncovcol+nqv+ntv+nqtv; /* total of covariates in the data, not in the model equation */
10515:
1.136 brouard 10516: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 10517: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10518: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 10519: }
1.126 brouard 10520:
1.302 brouard 10521: /* Is it a BOM UTF-8 Windows file? */
10522: /* First data line */
10523: linei=0;
10524: while(fgets(line, MAXLINE, fic)) {
10525: noffset=0;
10526: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10527: {
10528: noffset=noffset+3;
10529: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
10530: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
10531: fflush(ficlog); return 1;
10532: }
10533: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
10534: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
10535: {
10536: noffset=noffset+2;
1.304 brouard 10537: 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);
10538: 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 10539: fflush(ficlog); return 1;
10540: }
10541: else if( line[0] == 0 && line[1] == 0)
10542: {
10543: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10544: noffset=noffset+4;
1.304 brouard 10545: 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);
10546: 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 10547: fflush(ficlog); return 1;
10548: }
10549: } else{
10550: ;/*printf(" Not a BOM file\n");*/
10551: }
10552: /* If line starts with a # it is a comment */
10553: if (line[noffset] == '#') {
10554: linei=linei+1;
10555: break;
10556: }else{
10557: break;
10558: }
10559: }
10560: fclose(fic);
10561: if((fic=fopen(datafile,"r"))==NULL) {
10562: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10563: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
10564: }
10565: /* Not a Bom file */
10566:
1.136 brouard 10567: i=1;
10568: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
10569: linei=linei+1;
10570: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
10571: if(line[j] == '\t')
10572: line[j] = ' ';
10573: }
10574: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
10575: ;
10576: };
10577: line[j+1]=0; /* Trims blanks at end of line */
10578: if(line[0]=='#'){
10579: fprintf(ficlog,"Comment line\n%s\n",line);
10580: printf("Comment line\n%s\n",line);
10581: continue;
10582: }
10583: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 10584: strcpy(line, linetmp);
1.223 brouard 10585:
10586: /* Loops on waves */
10587: for (j=maxwav;j>=1;j--){
10588: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 10589: cutv(stra, strb, line, ' ');
10590: if(strb[0]=='.') { /* Missing value */
10591: lval=-1;
10592: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
1.341 brouard 10593: cotvar[j][ncovcol+nqv+ntv+iv][i]=-1; /* For performance reasons */
1.238 brouard 10594: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
10595: 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);
10596: 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);
10597: return 1;
10598: }
10599: }else{
10600: errno=0;
10601: /* what_kind_of_number(strb); */
10602: dval=strtod(strb,&endptr);
10603: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
10604: /* if(strb != endptr && *endptr == '\0') */
10605: /* dval=dlval; */
10606: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10607: if( strb[0]=='\0' || (*endptr != '\0')){
10608: 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);
10609: 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);
10610: return 1;
10611: }
10612: cotqvar[j][iv][i]=dval;
1.341 brouard 10613: cotvar[j][ncovcol+nqv+ntv+iv][i]=dval; /* because cotvar starts now at first ntv */
1.238 brouard 10614: }
10615: strcpy(line,stra);
1.223 brouard 10616: }/* end loop ntqv */
1.225 brouard 10617:
1.223 brouard 10618: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 10619: cutv(stra, strb, line, ' ');
10620: if(strb[0]=='.') { /* Missing value */
10621: lval=-1;
10622: }else{
10623: errno=0;
10624: lval=strtol(strb,&endptr,10);
10625: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10626: if( strb[0]=='\0' || (*endptr != '\0')){
10627: 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);
10628: 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);
10629: return 1;
10630: }
10631: }
10632: if(lval <-1 || lval >1){
10633: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10634: 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 10635: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10636: For example, for multinomial values like 1, 2 and 3,\n \
10637: build V1=0 V2=0 for the reference value (1),\n \
10638: V1=1 V2=0 for (2) \n \
1.223 brouard 10639: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10640: output of IMaCh is often meaningless.\n \
1.319 brouard 10641: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 10642: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10643: 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 10644: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10645: For example, for multinomial values like 1, 2 and 3,\n \
10646: build V1=0 V2=0 for the reference value (1),\n \
10647: V1=1 V2=0 for (2) \n \
1.223 brouard 10648: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10649: output of IMaCh is often meaningless.\n \
1.319 brouard 10650: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 10651: return 1;
10652: }
1.341 brouard 10653: cotvar[j][ncovcol+nqv+iv][i]=(double)(lval);
1.238 brouard 10654: strcpy(line,stra);
1.223 brouard 10655: }/* end loop ntv */
1.225 brouard 10656:
1.223 brouard 10657: /* Statuses at wave */
1.137 brouard 10658: cutv(stra, strb, line, ' ');
1.223 brouard 10659: if(strb[0]=='.') { /* Missing value */
1.238 brouard 10660: lval=-1;
1.136 brouard 10661: }else{
1.238 brouard 10662: errno=0;
10663: lval=strtol(strb,&endptr,10);
10664: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
1.347 brouard 10665: if( strb[0]=='\0' || (*endptr != '\0' )){
10666: 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);
10667: 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);
10668: return 1;
10669: }else if( lval==0 || lval > nlstate+ndeath){
1.348 ! brouard 10670: 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);
! 10671: 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 10672: return 1;
10673: }
1.136 brouard 10674: }
1.225 brouard 10675:
1.136 brouard 10676: s[j][i]=lval;
1.225 brouard 10677:
1.223 brouard 10678: /* Date of Interview */
1.136 brouard 10679: strcpy(line,stra);
10680: cutv(stra, strb,line,' ');
1.169 brouard 10681: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10682: }
1.169 brouard 10683: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 10684: month=99;
10685: year=9999;
1.136 brouard 10686: }else{
1.225 brouard 10687: 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);
10688: 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);
10689: return 1;
1.136 brouard 10690: }
10691: anint[j][i]= (double) year;
1.302 brouard 10692: mint[j][i]= (double)month;
10693: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
10694: /* 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]); */
10695: /* 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]); */
10696: /* } */
1.136 brouard 10697: strcpy(line,stra);
1.223 brouard 10698: } /* End loop on waves */
1.225 brouard 10699:
1.223 brouard 10700: /* Date of death */
1.136 brouard 10701: cutv(stra, strb,line,' ');
1.169 brouard 10702: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10703: }
1.169 brouard 10704: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 10705: month=99;
10706: year=9999;
10707: }else{
1.141 brouard 10708: 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 10709: 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);
10710: return 1;
1.136 brouard 10711: }
10712: andc[i]=(double) year;
10713: moisdc[i]=(double) month;
10714: strcpy(line,stra);
10715:
1.223 brouard 10716: /* Date of birth */
1.136 brouard 10717: cutv(stra, strb,line,' ');
1.169 brouard 10718: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10719: }
1.169 brouard 10720: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 10721: month=99;
10722: year=9999;
10723: }else{
1.141 brouard 10724: 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);
10725: 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 10726: return 1;
1.136 brouard 10727: }
10728: if (year==9999) {
1.141 brouard 10729: 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);
10730: 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 10731: return 1;
10732:
1.136 brouard 10733: }
10734: annais[i]=(double)(year);
1.302 brouard 10735: moisnais[i]=(double)(month);
10736: for (j=1;j<=maxwav;j++){
10737: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
10738: 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]);
10739: 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]);
10740: }
10741: }
10742:
1.136 brouard 10743: strcpy(line,stra);
1.225 brouard 10744:
1.223 brouard 10745: /* Sample weight */
1.136 brouard 10746: cutv(stra, strb,line,' ');
10747: errno=0;
10748: dval=strtod(strb,&endptr);
10749: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 10750: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
10751: 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 10752: fflush(ficlog);
10753: return 1;
10754: }
10755: weight[i]=dval;
10756: strcpy(line,stra);
1.225 brouard 10757:
1.223 brouard 10758: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
10759: cutv(stra, strb, line, ' ');
10760: if(strb[0]=='.') { /* Missing value */
1.225 brouard 10761: lval=-1;
1.311 brouard 10762: coqvar[iv][i]=NAN;
10763: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 10764: }else{
1.225 brouard 10765: errno=0;
10766: /* what_kind_of_number(strb); */
10767: dval=strtod(strb,&endptr);
10768: /* if(strb != endptr && *endptr == '\0') */
10769: /* dval=dlval; */
10770: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10771: if( strb[0]=='\0' || (*endptr != '\0')){
10772: 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);
10773: 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);
10774: return 1;
10775: }
10776: coqvar[iv][i]=dval;
1.226 brouard 10777: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 10778: }
10779: strcpy(line,stra);
10780: }/* end loop nqv */
1.136 brouard 10781:
1.223 brouard 10782: /* Covariate values */
1.136 brouard 10783: for (j=ncovcol;j>=1;j--){
10784: cutv(stra, strb,line,' ');
1.223 brouard 10785: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 10786: lval=-1;
1.136 brouard 10787: }else{
1.225 brouard 10788: errno=0;
10789: lval=strtol(strb,&endptr,10);
10790: if( strb[0]=='\0' || (*endptr != '\0')){
10791: 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);
10792: 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);
10793: return 1;
10794: }
1.136 brouard 10795: }
10796: if(lval <-1 || lval >1){
1.225 brouard 10797: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10798: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10799: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10800: For example, for multinomial values like 1, 2 and 3,\n \
10801: build V1=0 V2=0 for the reference value (1),\n \
10802: V1=1 V2=0 for (2) \n \
1.136 brouard 10803: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10804: output of IMaCh is often meaningless.\n \
1.136 brouard 10805: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 10806: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10807: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10808: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10809: For example, for multinomial values like 1, 2 and 3,\n \
10810: build V1=0 V2=0 for the reference value (1),\n \
10811: V1=1 V2=0 for (2) \n \
1.136 brouard 10812: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10813: output of IMaCh is often meaningless.\n \
1.136 brouard 10814: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 10815: return 1;
1.136 brouard 10816: }
10817: covar[j][i]=(double)(lval);
10818: strcpy(line,stra);
10819: }
10820: lstra=strlen(stra);
1.225 brouard 10821:
1.136 brouard 10822: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
10823: stratrunc = &(stra[lstra-9]);
10824: num[i]=atol(stratrunc);
10825: }
10826: else
10827: num[i]=atol(stra);
10828: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
10829: 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;}*/
10830:
10831: i=i+1;
10832: } /* End loop reading data */
1.225 brouard 10833:
1.136 brouard 10834: *imax=i-1; /* Number of individuals */
10835: fclose(fic);
1.225 brouard 10836:
1.136 brouard 10837: return (0);
1.164 brouard 10838: /* endread: */
1.225 brouard 10839: printf("Exiting readdata: ");
10840: fclose(fic);
10841: return (1);
1.223 brouard 10842: }
1.126 brouard 10843:
1.234 brouard 10844: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 10845: char *p1 = *stri, *p2 = *stri;
1.235 brouard 10846: while (*p2 == ' ')
1.234 brouard 10847: p2++;
10848: /* while ((*p1++ = *p2++) !=0) */
10849: /* ; */
10850: /* do */
10851: /* while (*p2 == ' ') */
10852: /* p2++; */
10853: /* while (*p1++ == *p2++); */
10854: *stri=p2;
1.145 brouard 10855: }
10856:
1.330 brouard 10857: int decoderesult( char resultline[], int nres)
1.230 brouard 10858: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
10859: {
1.235 brouard 10860: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 10861: char resultsav[MAXLINE];
1.330 brouard 10862: /* int resultmodel[MAXLINE]; */
1.334 brouard 10863: /* int modelresult[MAXLINE]; */
1.230 brouard 10864: char stra[80], strb[80], strc[80], strd[80],stre[80];
10865:
1.234 brouard 10866: removefirstspace(&resultline);
1.332 brouard 10867: printf("decoderesult:%s\n",resultline);
1.230 brouard 10868:
1.332 brouard 10869: strcpy(resultsav,resultline);
1.342 brouard 10870: /* printf("Decoderesult resultsav=\"%s\" resultline=\"%s\"\n", resultsav, resultline); */
1.230 brouard 10871: if (strlen(resultsav) >1){
1.334 brouard 10872: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' in this resultline */
1.230 brouard 10873: }
1.253 brouard 10874: if(j == 0){ /* Resultline but no = */
10875: TKresult[nres]=0; /* Combination for the nresult and the model */
10876: return (0);
10877: }
1.234 brouard 10878: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.334 brouard 10879: 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, %s.\n",j, cptcovs, model);
10880: 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, %s.\n",j, cptcovs, model);
1.332 brouard 10881: /* return 1;*/
1.234 brouard 10882: }
1.334 brouard 10883: for(k=1; k<=j;k++){ /* Loop on any covariate of the RESULT LINE */
1.234 brouard 10884: if(nbocc(resultsav,'=') >1){
1.318 brouard 10885: 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 10886: /* 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 10887: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.332 brouard 10888: /* If a blank, then strc="V4=" and strd='\0' */
10889: if(strc[0]=='\0'){
10890: printf("Error in resultline, probably a blank after the \"%s\", \"result:%s\", stra=\"%s\" resultsav=\"%s\"\n",strb,resultline, stra, resultsav);
10891: fprintf(ficlog,"Error in resultline, probably a blank after the \"V%s=\", resultline=%s\n",strb,resultline);
10892: return 1;
10893: }
1.234 brouard 10894: }else
10895: cutl(strc,strd,resultsav,'=');
1.318 brouard 10896: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 10897:
1.230 brouard 10898: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 10899: 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 10900: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
10901: /* cptcovsel++; */
10902: if (nbocc(stra,'=') >0)
10903: strcpy(resultsav,stra); /* and analyzes it */
10904: }
1.235 brouard 10905: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10906: /* 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 10907: 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 10908: if(Typevar[k1]==0){ /* Single covariate in model */
10909: /* 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 10910: match=0;
1.318 brouard 10911: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10912: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10913: modelresult[nres][k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10914: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10915: break;
10916: }
10917: }
10918: if(match == 0){
1.338 brouard 10919: 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]);
10920: 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 10921: return 1;
1.234 brouard 10922: }
1.332 brouard 10923: }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*/
10924: /* We feed resultmodel[k1]=k2; */
10925: match=0;
10926: 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 */
10927: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10928: 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 10929: resultmodel[nres][k1]=k2; /* Added here */
1.342 brouard 10930: /* printf("Decoderesult first modelresult[k2=%d]=%d (k1) V%d*AGE\n",k2,k1,Tvar[k1]); */
1.332 brouard 10931: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10932: break;
10933: }
10934: }
10935: if(match == 0){
1.338 brouard 10936: 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]);
10937: 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 10938: return 1;
10939: }
10940: }else if(Typevar[k1]==2){ /* Product No age We want to get the position in the resultline of the product in the model line*/
10941: /* resultmodel[nres][of such a Vn * Vm product k1] is not unique, so can't exist, we feed Tvard[k1][1] and [2] */
10942: match=0;
1.342 brouard 10943: /* 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 10944: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10945: if(Tvardk[k1][1]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10946: /* modelresult[k2]=k1; */
1.342 brouard 10947: /* printf("Decoderesult first Product modelresult[k2=%d]=%d (k1) V%d * \n",k2,k1,Tvarsel[k2]); */
1.332 brouard 10948: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10949: }
10950: }
10951: if(match == 0){
1.338 brouard 10952: printf("Error in result line (Product without age first variable): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][1], resultline, model);
10953: fprintf(ficlog,"Error in result line (Product without age first variable): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][1], resultline, model);
1.332 brouard 10954: return 1;
10955: }
10956: match=0;
10957: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10958: if(Tvardk[k1][2]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10959: /* modelresult[k2]=k1;*/
1.342 brouard 10960: /* printf("Decoderesult second Product modelresult[k2=%d]=%d (k1) * V%d \n ",k2,k1,Tvarsel[k2]); */
1.332 brouard 10961: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10962: break;
10963: }
10964: }
10965: if(match == 0){
1.338 brouard 10966: printf("Error in result line (Product without age second variable): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][2], resultline, model);
10967: fprintf(ficlog,"Error in result line (Product without age second variable): V%d is missing in result : %s according to model=1+age+%s\n",Tvardk[k1][2], resultline, model);
1.332 brouard 10968: return 1;
10969: }
10970: }/* End of testing */
1.333 brouard 10971: }/* End loop cptcovt */
1.235 brouard 10972: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10973: /* Feeds resultmodel[nres][k1]=k2 for single covariate (k1) in the model equation */
1.334 brouard 10974: 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)
10975: * Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 10976: match=0;
1.318 brouard 10977: 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 10978: if(Typevar[k1]==0){ /* Single only */
1.237 brouard 10979: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.330 brouard 10980: 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 10981: 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 10982: ++match;
10983: }
10984: }
10985: }
10986: if(match == 0){
1.338 brouard 10987: printf("Error in result line: variable V%d is missing in model; result: %s, model=1+age+%s\n",Tvarsel[k2], resultline, model);
10988: 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 10989: return 1;
1.234 brouard 10990: }else if(match > 1){
1.338 brouard 10991: printf("Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
10992: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
1.310 brouard 10993: return 1;
1.234 brouard 10994: }
10995: }
1.334 brouard 10996: /* cptcovres=j /\* Number of variables in the resultline is equal to cptcovs and thus useless *\/ */
1.234 brouard 10997: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10998: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 brouard 10999: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
11000: /* 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*/
11001: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 11002: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
11003: /* 1 0 0 0 */
11004: /* 2 1 0 0 */
11005: /* 3 0 1 0 */
1.330 brouard 11006: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 11007: /* 5 0 0 1 */
1.330 brouard 11008: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 11009: /* 7 0 1 1 */
11010: /* 8 1 1 1 */
1.237 brouard 11011: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
11012: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
11013: /* V5*age V5 known which value for nres? */
11014: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.334 brouard 11015: 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.
11016: * loop on position k1 in the MODEL LINE */
1.331 brouard 11017: /* k counting number of combination of single dummies in the equation model */
11018: /* k4 counting single dummies in the equation model */
11019: /* k4q counting single quantitatives in the equation model */
1.344 brouard 11020: 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 11021: /* 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 11022: /* 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 11023: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
1.332 brouard 11024: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
11025: /* k3 is the position in the nres result line of the k1th variable of the model equation */
11026: /* Tvarsel[k3]: Name of the variable at the k3th position in the result line. */
11027: /* Tvalsel[k3]: Value of the variable at the k3th position in the result line. */
11028: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.334 brouard 11029: /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 11030: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1.330 brouard 11031: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.332 brouard 11032: k3= resultmodel[nres][k1]; /* From position k1 in model get position k3 in result line */
11033: /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
11034: 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 11035: 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 11036: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][Name]=Value; stores the value into the name of the variable. */
1.332 brouard 11037: /* Tinvresult[nres][4]=1 */
1.334 brouard 11038: /* Tresult[nres][k4+1]=Tvalsel[k3];/\* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) *\/ */
11039: Tresult[nres][k3]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
11040: /* Tvresult[nres][k4+1]=(int)Tvarsel[k3];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
11041: Tvresult[nres][k3]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
1.237 brouard 11042: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.334 brouard 11043: precov[nres][k1]=Tvalsel[k3]; /* Value from resultline of the variable at the k1 position in the model */
1.342 brouard 11044: /* 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 11045: k4++;;
1.331 brouard 11046: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Quantitative and single */
1.330 brouard 11047: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.332 brouard 11048: /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.330 brouard 11049: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
1.332 brouard 11050: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 5 =k3q */
11051: k2q=(int)Tvarsel[k3q]; /* Name of variable at k3q th position in the resultline */
11052: /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 11053: /* Tqresult[nres][k4q+1]=Tvalsel[k3q]; /\* Tqresult[nres][1]=25.1 *\/ */
11054: /* Tvresult[nres][k4q+1]=(int)Tvarsel[k3q];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
11055: /* Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /\* Tvqresult[nres][1]=5 *\/ */
11056: Tqresult[nres][k3q]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
11057: Tvresult[nres][k3q]=(int)Tvarsel[k3q];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
11058: Tvqresult[nres][k3q]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
1.237 brouard 11059: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 brouard 11060: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 11061: precov[nres][k1]=Tvalsel[k3q];
1.342 brouard 11062: /* 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 11063: k4q++;;
1.331 brouard 11064: }else if( Dummy[k1]==2 ){ /* For dummy with age product */
11065: /* Tvar[k1]; */ /* Age variable */
1.332 brouard 11066: /* Wrong we want the value of variable name Tvar[k1] */
11067:
11068: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
1.331 brouard 11069: 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)*/
1.334 brouard 11070: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][4]=1 */
1.332 brouard 11071: precov[nres][k1]=Tvalsel[k3];
1.342 brouard 11072: /* 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 11073: }else if( Dummy[k1]==3 ){ /* For quant with age product */
1.332 brouard 11074: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
1.331 brouard 11075: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 11076: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* TinvDoQresult[nres][5]=25.1 */
1.332 brouard 11077: precov[nres][k1]=Tvalsel[k3q];
1.342 brouard 11078: /* 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]); */
1.331 brouard 11079: }else if(Typevar[k1]==2 ){ /* For product quant or dummy (not with age) */
1.332 brouard 11080: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
1.342 brouard 11081: /* 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 11082: }else{
1.332 brouard 11083: printf("Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
11084: fprintf(ficlog,"Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 11085: }
11086: }
1.234 brouard 11087:
1.334 brouard 11088: TKresult[nres]=++k; /* Number of combinations of dummies for the nresult and the model =Tvalsel[k3]*pow(2,k4) + 1*/
1.230 brouard 11089: return (0);
11090: }
1.235 brouard 11091:
1.230 brouard 11092: int decodemodel( char model[], int lastobs)
11093: /**< This routine decodes the model and returns:
1.224 brouard 11094: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
11095: * - nagesqr = 1 if age*age in the model, otherwise 0.
11096: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
11097: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
11098: * - cptcovage number of covariates with age*products =2
11099: * - cptcovs number of simple covariates
1.339 brouard 11100: * ncovcolt=ncovcol+nqv+ntv+nqtv total of covariates in the data, not in the model equation
1.224 brouard 11101: * - 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 11102: * which is a new column after the 9 (ncovcol+nqv+ntv+nqtv) variables.
1.319 brouard 11103: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 11104: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
11105: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
11106: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
11107: */
1.319 brouard 11108: /* 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 11109: {
1.238 brouard 11110: int i, j, k, ks, v;
1.227 brouard 11111: int j1, k1, k2, k3, k4;
1.136 brouard 11112: char modelsav[80];
1.145 brouard 11113: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 11114: char *strpt;
1.136 brouard 11115:
1.145 brouard 11116: /*removespace(model);*/
1.136 brouard 11117: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 11118: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 11119: if (strstr(model,"AGE") !=0){
1.192 brouard 11120: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
11121: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 11122: return 1;
11123: }
1.141 brouard 11124: if (strstr(model,"v") !=0){
1.338 brouard 11125: printf("Error. 'v' must be in upper case 'V' model=1+age+%s ",model);
11126: fprintf(ficlog,"Error. 'v' must be in upper case model=1+age+%s ",model);fflush(ficlog);
1.141 brouard 11127: return 1;
11128: }
1.187 brouard 11129: strcpy(modelsav,model);
11130: if ((strpt=strstr(model,"age*age")) !=0){
1.338 brouard 11131: printf(" strpt=%s, model=1+age+%s\n",strpt, model);
1.187 brouard 11132: if(strpt != model){
1.338 brouard 11133: printf("Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 11134: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 11135: corresponding column of parameters.\n",model);
1.338 brouard 11136: fprintf(ficlog,"Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 11137: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 11138: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 11139: return 1;
1.225 brouard 11140: }
1.187 brouard 11141: nagesqr=1;
11142: if (strstr(model,"+age*age") !=0)
1.234 brouard 11143: substrchaine(modelsav, model, "+age*age");
1.187 brouard 11144: else if (strstr(model,"age*age+") !=0)
1.234 brouard 11145: substrchaine(modelsav, model, "age*age+");
1.187 brouard 11146: else
1.234 brouard 11147: substrchaine(modelsav, model, "age*age");
1.187 brouard 11148: }else
11149: nagesqr=0;
11150: if (strlen(modelsav) >1){
11151: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
11152: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 11153: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 11154: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 11155: * cst, age and age*age
11156: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
11157: /* including age products which are counted in cptcovage.
11158: * but the covariates which are products must be treated
11159: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 11160: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
11161: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 11162:
11163:
1.187 brouard 11164: /* Design
11165: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
11166: * < ncovcol=8 >
11167: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
11168: * k= 1 2 3 4 5 6 7 8
11169: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
1.345 brouard 11170: * covar[k,i], are for fixed covariates, value of kth covariate if not including age for individual i:
1.224 brouard 11171: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
11172: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 11173: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
11174: * Tage[++cptcovage]=k
1.345 brouard 11175: * if products, new covar are created after ncovcol + nqv (quanti fixed) with k1
1.187 brouard 11176: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
11177: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
11178: * 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
11179: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
11180: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
11181: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
1.345 brouard 11182: * < ncovcol=8 8 fixed covariate. Additional starts at 9 (V5*V6) and 10(V7*V8) >
1.187 brouard 11183: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
11184: * k= 1 2 3 4 5 6 7 8 9 10 11 12
1.345 brouard 11185: * Tvard[k]= 2 1 3 3 10 11 8 8 5 6 7 8
11186: * p Tvar[1]@12={2, 1, 3, 3, 9, 10, 8, 8}
1.187 brouard 11187: * p Tprod[1]@2={ 6, 5}
11188: *p Tvard[1][1]@4= {7, 8, 5, 6}
11189: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
11190: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 11191: *How to reorganize? Tvars(orted)
1.187 brouard 11192: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
11193: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
11194: * {2, 1, 4, 8, 5, 6, 3, 7}
11195: * Struct []
11196: */
1.225 brouard 11197:
1.187 brouard 11198: /* This loop fills the array Tvar from the string 'model'.*/
11199: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
11200: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
11201: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
11202: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
11203: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
11204: /* k=1 Tvar[1]=2 (from V2) */
11205: /* k=5 Tvar[5] */
11206: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 11207: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 11208: /* } */
1.198 brouard 11209: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 11210: /*
11211: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 11212: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
11213: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
11214: }
1.187 brouard 11215: cptcovage=0;
1.319 brouard 11216: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
11217: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
11218: 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" */
11219: if (nbocc(modelsav,'+')==0)
11220: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 11221: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
11222: /*scanf("%d",i);*/
1.319 brouard 11223: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
11224: cutl(strc,strd,strb,'*'); /**< k=1 strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
1.234 brouard 11225: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
11226: /* covar is not filled and then is empty */
11227: cptcovprod--;
11228: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 11229: 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 */
1.234 brouard 11230: Typevar[k]=1; /* 1 for age product */
1.319 brouard 11231: cptcovage++; /* Counts the number of covariates which include age as a product */
11232: Tage[cptcovage]=k; /* V2+V1+V4+V3*age Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
1.234 brouard 11233: /*printf("stre=%s ", stre);*/
11234: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
11235: cptcovprod--;
11236: cutl(stre,strb,strc,'V');
11237: Tvar[k]=atoi(stre);
11238: Typevar[k]=1; /* 1 for age product */
11239: cptcovage++;
11240: Tage[cptcovage]=k;
11241: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
11242: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
11243: cptcovn++;
11244: cptcovprodnoage++;k1++;
11245: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
1.339 brouard 11246: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* ncovcolt+k1; For model-covariate k tells which data-covariate to use but
1.234 brouard 11247: because this model-covariate is a construction we invent a new column
11248: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.335 brouard 11249: If already ncovcol=4 and model= V2 + V1 + V1*V4 + age*V3 + V3*V2
1.319 brouard 11250: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
1.339 brouard 11251: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=3 etc */
1.335 brouard 11252: /* Please remark that the new variables are model dependent */
11253: /* If we have 4 variable but the model uses only 3, like in
11254: * model= V1 + age*V1 + V2 + V3 + age*V2 + age*V3 + V1*V2 + V1*V3
11255: * k= 1 2 3 4 5 6 7 8
11256: * Tvar[k]=1 1 2 3 2 3 (5 6) (and not 4 5 because of V4 missing)
11257: * Tage[kk] [1]= 2 [2]=5 [3]=6 kk=1 to cptcovage=3
11258: * Tvar[Tage[kk]][1]=2 [2]=2 [3]=3
11259: */
1.339 brouard 11260: Typevar[k]=2; /* 2 for product */
1.234 brouard 11261: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
11262: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 11263: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 11264: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
1.330 brouard 11265: Tvardk[k][1] =atoi(strc); /* m 1 for V1*/
1.234 brouard 11266: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
1.330 brouard 11267: Tvardk[k][2] =atoi(stre); /* n 4 for V4*/
1.234 brouard 11268: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
11269: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
11270: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 11271: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 11272: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
1.339 brouard 11273: 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 */
11274: for (i=1; i<=lastobs;i++){/* For fixed product */
1.234 brouard 11275: /* Computes the new covariate which is a product of
11276: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
1.339 brouard 11277: covar[ncovcolt+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
11278: }
11279: } /*End of FixedV */
1.234 brouard 11280: } /* End age is not in the model */
11281: } /* End if model includes a product */
1.319 brouard 11282: else { /* not a product */
1.234 brouard 11283: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
11284: /* scanf("%d",i);*/
11285: cutl(strd,strc,strb,'V');
11286: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
11287: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
11288: Tvar[k]=atoi(strd);
11289: Typevar[k]=0; /* 0 for simple covariates */
11290: }
11291: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 11292: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 11293: scanf("%d",i);*/
1.187 brouard 11294: } /* end of loop + on total covariates */
11295: } /* end if strlen(modelsave == 0) age*age might exist */
11296: } /* end if strlen(model == 0) */
1.136 brouard 11297:
11298: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
11299: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 11300:
1.136 brouard 11301: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 11302: printf("cptcovprod=%d ", cptcovprod);
11303: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
11304: scanf("%d ",i);*/
11305:
11306:
1.230 brouard 11307: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
11308: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 11309: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
11310: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
11311: k = 1 2 3 4 5 6 7 8 9
11312: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 11313: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 11314: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
11315: Dummy[k] 1 0 0 0 3 1 1 2 3
11316: Tmodelind[combination of covar]=k;
1.225 brouard 11317: */
11318: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 11319: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 11320: /* 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 11321: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 11322: printf("Model=1+age+%s\n\
1.227 brouard 11323: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
11324: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
11325: 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 11326: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 11327: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
11328: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
11329: 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 11330: for(k=-1;k<=NCOVMAX; k++){ Fixed[k]=0; Dummy[k]=0;}
11331: for(k=1;k<=NCOVMAX; k++){TvarFind[k]=0; TvarVind[k]=0;}
1.343 brouard 11332: for(k=1, ncovf=0, nsd=0, nsq=0, ncovv=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 11333: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 11334: Fixed[k]= 0;
11335: Dummy[k]= 0;
1.225 brouard 11336: ncoveff++;
1.232 brouard 11337: ncovf++;
1.234 brouard 11338: nsd++;
11339: modell[k].maintype= FTYPE;
11340: TvarsD[nsd]=Tvar[k];
11341: TvarsDind[nsd]=k;
1.330 brouard 11342: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 11343: TvarF[ncovf]=Tvar[k];
11344: TvarFind[ncovf]=k;
11345: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
11346: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.339 brouard 11347: /* }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /\* Product of fixed dummy (<=ncovcol) covariates For a fixed product k is higher than ncovcol *\/ */
11348: }else if( Tposprod[k]>0 && Typevar[k]==2 && 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 */
1.234 brouard 11349: Fixed[k]= 0;
11350: Dummy[k]= 0;
11351: ncoveff++;
11352: ncovf++;
11353: modell[k].maintype= FTYPE;
11354: TvarF[ncovf]=Tvar[k];
1.330 brouard 11355: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
1.234 brouard 11356: TvarFind[ncovf]=k;
1.230 brouard 11357: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 11358: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 11359: }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 11360: Fixed[k]= 0;
11361: Dummy[k]= 1;
1.230 brouard 11362: nqfveff++;
1.234 brouard 11363: modell[k].maintype= FTYPE;
11364: modell[k].subtype= FQ;
11365: nsq++;
1.334 brouard 11366: TvarsQ[nsq]=Tvar[k]; /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary see below */
11367: TvarsQind[nsq]=k; /* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.232 brouard 11368: ncovf++;
1.234 brouard 11369: TvarF[ncovf]=Tvar[k];
11370: TvarFind[ncovf]=k;
1.231 brouard 11371: 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 11372: 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 11373: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.339 brouard 11374: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11375: /* model V1+V3+age*V1+age*V3+V1*V3 */
11376: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11377: ncovvt++;
11378: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
11379: TvarVVind[ncovvt]=k; /* TvarVVind[1]=2 (second position in the model equation */
11380:
1.227 brouard 11381: Fixed[k]= 1;
11382: Dummy[k]= 0;
1.225 brouard 11383: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 11384: modell[k].maintype= VTYPE;
11385: modell[k].subtype= VD;
11386: nsd++;
11387: TvarsD[nsd]=Tvar[k];
11388: TvarsDind[nsd]=k;
1.330 brouard 11389: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 11390: ncovv++; /* Only simple time varying variables */
11391: TvarV[ncovv]=Tvar[k];
1.242 brouard 11392: 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 11393: 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 */
11394: 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 11395: 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);
11396: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 11397: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.339 brouard 11398: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11399: /* model V1+V3+age*V1+age*V3+V1*V3 */
11400: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11401: ncovvt++;
11402: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
11403: TvarVVind[ncovvt]=k; /* TvarVV[1]=V3 (first time varying in the model equation */
11404:
1.234 brouard 11405: Fixed[k]= 1;
11406: Dummy[k]= 1;
11407: nqtveff++;
11408: modell[k].maintype= VTYPE;
11409: modell[k].subtype= VQ;
11410: ncovv++; /* Only simple time varying variables */
11411: nsq++;
1.334 brouard 11412: 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) */
11413: 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 11414: TvarV[ncovv]=Tvar[k];
1.242 brouard 11415: 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 11416: 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 */
11417: 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 11418: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
11419: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
1.342 brouard 11420: /* printf("Quasi TmodelQind[%d]=%d,Tvar[TmodelQind[%d]]=V%d, ncovcol=%d, nqv=%d, ntv=%d,Tvar[k]- ncovcol-nqv-ntv=%d\n",nqtveff,k,nqtveff,Tvar[k], ncovcol, nqv, ntv, Tvar[k]- ncovcol-nqv-ntv); */
11421: /* printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv); */
1.227 brouard 11422: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 11423: ncova++;
11424: TvarA[ncova]=Tvar[k];
11425: TvarAind[ncova]=k;
1.231 brouard 11426: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 11427: Fixed[k]= 2;
11428: Dummy[k]= 2;
11429: modell[k].maintype= ATYPE;
11430: modell[k].subtype= APFD;
11431: /* ncoveff++; */
1.227 brouard 11432: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 11433: Fixed[k]= 2;
11434: Dummy[k]= 3;
11435: modell[k].maintype= ATYPE;
11436: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
11437: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 11438: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 11439: Fixed[k]= 3;
11440: Dummy[k]= 2;
11441: modell[k].maintype= ATYPE;
11442: modell[k].subtype= APVD; /* Product age * varying dummy */
11443: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 11444: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 11445: Fixed[k]= 3;
11446: Dummy[k]= 3;
11447: modell[k].maintype= ATYPE;
11448: modell[k].subtype= APVQ; /* Product age * varying quantitative */
11449: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 11450: }
1.339 brouard 11451: }else if (Typevar[k] == 2) { /* product 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 */
11452: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11453: /* model V1+V3+age*V1+age*V3+V1*V3 */
11454: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11455: k1=Tposprod[k]; /* Position in the products of product k, Tposprod={0, 0, 0, 0, 1} k1=1 first product but second time varying because of V3 */
11456: ncovvt++;
11457: TvarVV[ncovvt]=Tvard[k1][1]; /* TvarVV[2]=V1 (because TvarVV[1] was V3, first time varying covariates */
11458: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
11459: ncovvt++;
11460: TvarVV[ncovvt]=Tvard[k1][2]; /* TvarVV[3]=V3 */
11461: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
11462:
11463:
11464: if(Tvard[k1][1] <=ncovcol){ /* Vn is dummy fixed, (Tvard[1][1]=V1), (Tvard[1][1]=V3 time varying) */
11465: if(Tvard[k1][2] <=ncovcol){ /* Vm is dummy fixed */
1.240 brouard 11466: Fixed[k]= 1;
11467: Dummy[k]= 0;
11468: modell[k].maintype= FTYPE;
11469: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
11470: ncovf++; /* Fixed variables without age */
11471: TvarF[ncovf]=Tvar[k];
11472: TvarFind[ncovf]=k;
1.339 brouard 11473: }else if(Tvard[k1][2] <=ncovcol+nqv){ /* Vm is quanti fixed */
11474: Fixed[k]= 0; /* Fixed product */
1.240 brouard 11475: Dummy[k]= 1;
11476: modell[k].maintype= FTYPE;
11477: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
11478: ncovf++; /* Varying variables without age */
11479: TvarF[ncovf]=Tvar[k];
11480: TvarFind[ncovf]=k;
1.339 brouard 11481: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is a time varying dummy covariate */
1.240 brouard 11482: Fixed[k]= 1;
11483: Dummy[k]= 0;
11484: modell[k].maintype= VTYPE;
11485: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
11486: ncovv++; /* Varying variables without age */
1.339 brouard 11487: TvarV[ncovv]=Tvar[k]; /* TvarV[1]=Tvar[5]=5 because there is a V4 */
11488: TvarVind[ncovv]=k;/* TvarVind[1]=5 */
11489: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is a time varying quantitative covariate */
1.240 brouard 11490: Fixed[k]= 1;
11491: Dummy[k]= 1;
11492: modell[k].maintype= VTYPE;
11493: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
11494: ncovv++; /* Varying variables without age */
11495: TvarV[ncovv]=Tvar[k];
11496: TvarVind[ncovv]=k;
11497: }
1.339 brouard 11498: }else if(Tvard[k1][1] <=ncovcol+nqv){ /* Vn is fixed quanti */
11499: if(Tvard[k1][2] <=ncovcol){ /* Vm is fixed dummy */
11500: Fixed[k]= 0; /* Fixed product */
1.240 brouard 11501: Dummy[k]= 1;
11502: modell[k].maintype= FTYPE;
11503: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
11504: ncovf++; /* Fixed variables without age */
11505: TvarF[ncovf]=Tvar[k];
11506: TvarFind[ncovf]=k;
1.339 brouard 11507: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is time varying */
1.240 brouard 11508: Fixed[k]= 1;
11509: Dummy[k]= 1;
11510: modell[k].maintype= VTYPE;
11511: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
11512: ncovv++; /* Varying variables without age */
11513: TvarV[ncovv]=Tvar[k];
11514: TvarVind[ncovv]=k;
1.339 brouard 11515: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is time varying quanti */
1.240 brouard 11516: Fixed[k]= 1;
11517: Dummy[k]= 1;
11518: modell[k].maintype= VTYPE;
11519: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
11520: ncovv++; /* Varying variables without age */
11521: TvarV[ncovv]=Tvar[k];
11522: TvarVind[ncovv]=k;
11523: ncovv++; /* Varying variables without age */
11524: TvarV[ncovv]=Tvar[k];
11525: TvarVind[ncovv]=k;
11526: }
1.339 brouard 11527: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){ /* Vn is time varying dummy */
1.240 brouard 11528: if(Tvard[k1][2] <=ncovcol){
11529: Fixed[k]= 1;
11530: Dummy[k]= 1;
11531: modell[k].maintype= VTYPE;
11532: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
11533: ncovv++; /* Varying variables without age */
11534: TvarV[ncovv]=Tvar[k];
11535: TvarVind[ncovv]=k;
11536: }else if(Tvard[k1][2] <=ncovcol+nqv){
11537: Fixed[k]= 1;
11538: Dummy[k]= 1;
11539: modell[k].maintype= VTYPE;
11540: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
11541: ncovv++; /* Varying variables without age */
11542: TvarV[ncovv]=Tvar[k];
11543: TvarVind[ncovv]=k;
11544: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11545: Fixed[k]= 1;
11546: Dummy[k]= 0;
11547: modell[k].maintype= VTYPE;
11548: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
11549: ncovv++; /* Varying variables without age */
11550: TvarV[ncovv]=Tvar[k];
11551: TvarVind[ncovv]=k;
11552: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11553: Fixed[k]= 1;
11554: Dummy[k]= 1;
11555: modell[k].maintype= VTYPE;
11556: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
11557: ncovv++; /* Varying variables without age */
11558: TvarV[ncovv]=Tvar[k];
11559: TvarVind[ncovv]=k;
11560: }
1.339 brouard 11561: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){ /* Vn is time varying quanti */
1.240 brouard 11562: if(Tvard[k1][2] <=ncovcol){
11563: Fixed[k]= 1;
11564: Dummy[k]= 1;
11565: modell[k].maintype= VTYPE;
11566: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
11567: ncovv++; /* Varying variables without age */
11568: TvarV[ncovv]=Tvar[k];
11569: TvarVind[ncovv]=k;
11570: }else if(Tvard[k1][2] <=ncovcol+nqv){
11571: Fixed[k]= 1;
11572: Dummy[k]= 1;
11573: modell[k].maintype= VTYPE;
11574: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
11575: ncovv++; /* Varying variables without age */
11576: TvarV[ncovv]=Tvar[k];
11577: TvarVind[ncovv]=k;
11578: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11579: Fixed[k]= 1;
11580: Dummy[k]= 1;
11581: modell[k].maintype= VTYPE;
11582: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
11583: ncovv++; /* Varying variables without age */
11584: TvarV[ncovv]=Tvar[k];
11585: TvarVind[ncovv]=k;
11586: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11587: Fixed[k]= 1;
11588: Dummy[k]= 1;
11589: modell[k].maintype= VTYPE;
11590: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
11591: ncovv++; /* Varying variables without age */
11592: TvarV[ncovv]=Tvar[k];
11593: TvarVind[ncovv]=k;
11594: }
1.227 brouard 11595: }else{
1.240 brouard 11596: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11597: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11598: } /*end k1*/
1.225 brouard 11599: }else{
1.226 brouard 11600: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
11601: 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 11602: }
1.342 brouard 11603: /* 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]); */
11604: /* printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype); */
1.227 brouard 11605: 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]);
11606: }
11607: /* Searching for doublons in the model */
11608: for(k1=1; k1<= cptcovt;k1++){
11609: for(k2=1; k2 <k1;k2++){
1.285 brouard 11610: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
11611: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 11612: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
11613: if(Tvar[k1]==Tvar[k2]){
1.338 brouard 11614: 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]);
11615: 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 11616: return(1);
11617: }
11618: }else if (Typevar[k1] ==2){
11619: k3=Tposprod[k1];
11620: k4=Tposprod[k2];
11621: 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 11622: 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]]);
11623: 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 11624: return(1);
11625: }
11626: }
1.227 brouard 11627: }
11628: }
1.225 brouard 11629: }
11630: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
11631: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 11632: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
11633: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 11634: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 11635: /*endread:*/
1.225 brouard 11636: printf("Exiting decodemodel: ");
11637: return (1);
1.136 brouard 11638: }
11639:
1.169 brouard 11640: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 11641: {/* Check ages at death */
1.136 brouard 11642: int i, m;
1.218 brouard 11643: int firstone=0;
11644:
1.136 brouard 11645: for (i=1; i<=imx; i++) {
11646: for(m=2; (m<= maxwav); m++) {
11647: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
11648: anint[m][i]=9999;
1.216 brouard 11649: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
11650: s[m][i]=-1;
1.136 brouard 11651: }
11652: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 11653: *nberr = *nberr + 1;
1.218 brouard 11654: if(firstone == 0){
11655: firstone=1;
1.260 brouard 11656: 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 11657: }
1.262 brouard 11658: 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 11659: s[m][i]=-1; /* Droping the death status */
1.136 brouard 11660: }
11661: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 11662: (*nberr)++;
1.259 brouard 11663: 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 11664: 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 11665: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 11666: }
11667: }
11668: }
11669:
11670: for (i=1; i<=imx; i++) {
11671: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
11672: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 11673: 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 11674: if (s[m][i] >= nlstate+1) {
1.169 brouard 11675: if(agedc[i]>0){
11676: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 11677: agev[m][i]=agedc[i];
1.214 brouard 11678: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 11679: }else {
1.136 brouard 11680: if ((int)andc[i]!=9999){
11681: nbwarn++;
11682: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
11683: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
11684: agev[m][i]=-1;
11685: }
11686: }
1.169 brouard 11687: } /* agedc > 0 */
1.214 brouard 11688: } /* end if */
1.136 brouard 11689: else if(s[m][i] !=9){ /* Standard case, age in fractional
11690: years but with the precision of a month */
11691: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
11692: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
11693: agev[m][i]=1;
11694: else if(agev[m][i] < *agemin){
11695: *agemin=agev[m][i];
11696: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
11697: }
11698: else if(agev[m][i] >*agemax){
11699: *agemax=agev[m][i];
1.156 brouard 11700: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 11701: }
11702: /*agev[m][i]=anint[m][i]-annais[i];*/
11703: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 11704: } /* en if 9*/
1.136 brouard 11705: else { /* =9 */
1.214 brouard 11706: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 11707: agev[m][i]=1;
11708: s[m][i]=-1;
11709: }
11710: }
1.214 brouard 11711: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 11712: agev[m][i]=1;
1.214 brouard 11713: else{
11714: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11715: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11716: agev[m][i]=0;
11717: }
11718: } /* End for lastpass */
11719: }
1.136 brouard 11720:
11721: for (i=1; i<=imx; i++) {
11722: for(m=firstpass; (m<=lastpass); m++){
11723: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 11724: (*nberr)++;
1.136 brouard 11725: 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);
11726: 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);
11727: return 1;
11728: }
11729: }
11730: }
11731:
11732: /*for (i=1; i<=imx; i++){
11733: for (m=firstpass; (m<lastpass); m++){
11734: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
11735: }
11736:
11737: }*/
11738:
11739:
1.139 brouard 11740: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
11741: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 11742:
11743: return (0);
1.164 brouard 11744: /* endread:*/
1.136 brouard 11745: printf("Exiting calandcheckages: ");
11746: return (1);
11747: }
11748:
1.172 brouard 11749: #if defined(_MSC_VER)
11750: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11751: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11752: //#include "stdafx.h"
11753: //#include <stdio.h>
11754: //#include <tchar.h>
11755: //#include <windows.h>
11756: //#include <iostream>
11757: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
11758:
11759: LPFN_ISWOW64PROCESS fnIsWow64Process;
11760:
11761: BOOL IsWow64()
11762: {
11763: BOOL bIsWow64 = FALSE;
11764:
11765: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
11766: // (HANDLE, PBOOL);
11767:
11768: //LPFN_ISWOW64PROCESS fnIsWow64Process;
11769:
11770: HMODULE module = GetModuleHandle(_T("kernel32"));
11771: const char funcName[] = "IsWow64Process";
11772: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
11773: GetProcAddress(module, funcName);
11774:
11775: if (NULL != fnIsWow64Process)
11776: {
11777: if (!fnIsWow64Process(GetCurrentProcess(),
11778: &bIsWow64))
11779: //throw std::exception("Unknown error");
11780: printf("Unknown error\n");
11781: }
11782: return bIsWow64 != FALSE;
11783: }
11784: #endif
1.177 brouard 11785:
1.191 brouard 11786: void syscompilerinfo(int logged)
1.292 brouard 11787: {
11788: #include <stdint.h>
11789:
11790: /* #include "syscompilerinfo.h"*/
1.185 brouard 11791: /* command line Intel compiler 32bit windows, XP compatible:*/
11792: /* /GS /W3 /Gy
11793: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
11794: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
11795: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 11796: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
11797: */
11798: /* 64 bits */
1.185 brouard 11799: /*
11800: /GS /W3 /Gy
11801: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
11802: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
11803: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
11804: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
11805: /* Optimization are useless and O3 is slower than O2 */
11806: /*
11807: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
11808: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
11809: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
11810: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
11811: */
1.186 brouard 11812: /* Link is */ /* /OUT:"visual studio
1.185 brouard 11813: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
11814: /PDB:"visual studio
11815: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
11816: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
11817: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
11818: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
11819: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
11820: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
11821: uiAccess='false'"
11822: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
11823: /NOLOGO /TLBID:1
11824: */
1.292 brouard 11825:
11826:
1.177 brouard 11827: #if defined __INTEL_COMPILER
1.178 brouard 11828: #if defined(__GNUC__)
11829: struct utsname sysInfo; /* For Intel on Linux and OS/X */
11830: #endif
1.177 brouard 11831: #elif defined(__GNUC__)
1.179 brouard 11832: #ifndef __APPLE__
1.174 brouard 11833: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 11834: #endif
1.177 brouard 11835: struct utsname sysInfo;
1.178 brouard 11836: int cross = CROSS;
11837: if (cross){
11838: printf("Cross-");
1.191 brouard 11839: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 11840: }
1.174 brouard 11841: #endif
11842:
1.191 brouard 11843: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 11844: #if defined(__clang__)
1.191 brouard 11845: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 11846: #endif
11847: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 11848: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 11849: #endif
11850: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 11851: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 11852: #endif
11853: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 11854: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 11855: #endif
11856: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 11857: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 11858: #endif
11859: #if defined(_MSC_VER)
1.191 brouard 11860: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 11861: #endif
11862: #if defined(__PGI)
1.191 brouard 11863: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 11864: #endif
11865: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 11866: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 11867: #endif
1.191 brouard 11868: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 11869:
1.167 brouard 11870: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
11871: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
11872: // Windows (x64 and x86)
1.191 brouard 11873: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 11874: #elif __unix__ // all unices, not all compilers
11875: // Unix
1.191 brouard 11876: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 11877: #elif __linux__
11878: // linux
1.191 brouard 11879: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 11880: #elif __APPLE__
1.174 brouard 11881: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 11882: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 11883: #endif
11884:
11885: /* __MINGW32__ */
11886: /* __CYGWIN__ */
11887: /* __MINGW64__ */
11888: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
11889: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
11890: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
11891: /* _WIN64 // Defined for applications for Win64. */
11892: /* _M_X64 // Defined for compilations that target x64 processors. */
11893: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 11894:
1.167 brouard 11895: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 11896: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 11897: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 11898: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 11899: #else
1.191 brouard 11900: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 11901: #endif
11902:
1.169 brouard 11903: #if defined(__GNUC__)
11904: # if defined(__GNUC_PATCHLEVEL__)
11905: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11906: + __GNUC_MINOR__ * 100 \
11907: + __GNUC_PATCHLEVEL__)
11908: # else
11909: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11910: + __GNUC_MINOR__ * 100)
11911: # endif
1.174 brouard 11912: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 11913: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 11914:
11915: if (uname(&sysInfo) != -1) {
11916: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 11917: 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 11918: }
11919: else
11920: perror("uname() error");
1.179 brouard 11921: //#ifndef __INTEL_COMPILER
11922: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 11923: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 11924: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 11925: #endif
1.169 brouard 11926: #endif
1.172 brouard 11927:
1.286 brouard 11928: // void main ()
1.172 brouard 11929: // {
1.169 brouard 11930: #if defined(_MSC_VER)
1.174 brouard 11931: if (IsWow64()){
1.191 brouard 11932: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
11933: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 11934: }
11935: else{
1.191 brouard 11936: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
11937: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 11938: }
1.172 brouard 11939: // printf("\nPress Enter to continue...");
11940: // getchar();
11941: // }
11942:
1.169 brouard 11943: #endif
11944:
1.167 brouard 11945:
1.219 brouard 11946: }
1.136 brouard 11947:
1.219 brouard 11948: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 11949: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.332 brouard 11950: /* Computes the prevalence limit for each combination of the dummy covariates */
1.235 brouard 11951: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 11952: /* double ftolpl = 1.e-10; */
1.180 brouard 11953: double age, agebase, agelim;
1.203 brouard 11954: double tot;
1.180 brouard 11955:
1.202 brouard 11956: strcpy(filerespl,"PL_");
11957: strcat(filerespl,fileresu);
11958: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 11959: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
11960: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 11961: }
1.288 brouard 11962: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
11963: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 11964: pstamp(ficrespl);
1.288 brouard 11965: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 11966: fprintf(ficrespl,"#Age ");
11967: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
11968: fprintf(ficrespl,"\n");
1.180 brouard 11969:
1.219 brouard 11970: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 11971:
1.219 brouard 11972: agebase=ageminpar;
11973: agelim=agemaxpar;
1.180 brouard 11974:
1.227 brouard 11975: /* i1=pow(2,ncoveff); */
1.234 brouard 11976: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 11977: if (cptcovn < 1){i1=1;}
1.180 brouard 11978:
1.337 brouard 11979: /* for(k=1; k<=i1;k++){ /\* For each combination k of dummy covariates in the model *\/ */
1.238 brouard 11980: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 11981: k=TKresult[nres];
1.338 brouard 11982: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 11983: /* if(i1 != 1 && TKresult[nres]!= k) /\* We found the combination k corresponding to the resultline value of dummies *\/ */
11984: /* continue; */
1.235 brouard 11985:
1.238 brouard 11986: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11987: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
11988: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
11989: /* k=k+1; */
11990: /* to clean */
1.332 brouard 11991: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11992: fprintf(ficrespl,"#******");
11993: printf("#******");
11994: fprintf(ficlog,"#******");
1.337 brouard 11995: 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 11996: /* fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Here problem for varying dummy*\/ */
1.337 brouard 11997: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11998: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11999: fprintf(ficrespl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12000: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12001: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12002: }
12003: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
12004: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12005: /* fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12006: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12007: /* } */
1.238 brouard 12008: fprintf(ficrespl,"******\n");
12009: printf("******\n");
12010: fprintf(ficlog,"******\n");
12011: if(invalidvarcomb[k]){
12012: printf("\nCombination (%d) ignored because no case \n",k);
12013: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
12014: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
12015: continue;
12016: }
1.219 brouard 12017:
1.238 brouard 12018: fprintf(ficrespl,"#Age ");
1.337 brouard 12019: /* for(j=1;j<=cptcoveff;j++) { */
12020: /* fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12021: /* } */
12022: for(j=1;j<=cptcovs;j++) { /* New the quanti variable is added */
12023: fprintf(ficrespl,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12024: }
12025: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
12026: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 12027:
1.238 brouard 12028: for (age=agebase; age<=agelim; age++){
12029: /* for (age=agebase; age<=agebase; age++){ */
1.337 brouard 12030: /**< Computes the prevalence limit in each live state at age x and for covariate combination (k and) nres */
12031: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres); /* Nicely done */
1.238 brouard 12032: fprintf(ficrespl,"%.0f ",age );
1.337 brouard 12033: /* for(j=1;j<=cptcoveff;j++) */
12034: /* fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12035: for(j=1;j<=cptcovs;j++)
12036: fprintf(ficrespl,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12037: tot=0.;
12038: for(i=1; i<=nlstate;i++){
12039: tot += prlim[i][i];
12040: fprintf(ficrespl," %.5f", prlim[i][i]);
12041: }
12042: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
12043: } /* Age */
12044: /* was end of cptcod */
1.337 brouard 12045: } /* nres */
12046: /* } /\* for each combination *\/ */
1.219 brouard 12047: return 0;
1.180 brouard 12048: }
12049:
1.218 brouard 12050: 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 12051: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 12052:
12053: /* Computes the back prevalence limit for any combination of covariate values
12054: * at any age between ageminpar and agemaxpar
12055: */
1.235 brouard 12056: int i, j, k, i1, nres=0 ;
1.217 brouard 12057: /* double ftolpl = 1.e-10; */
12058: double age, agebase, agelim;
12059: double tot;
1.218 brouard 12060: /* double ***mobaverage; */
12061: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 12062:
12063: strcpy(fileresplb,"PLB_");
12064: strcat(fileresplb,fileresu);
12065: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 12066: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
12067: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 12068: }
1.288 brouard 12069: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
12070: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 12071: pstamp(ficresplb);
1.288 brouard 12072: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 12073: fprintf(ficresplb,"#Age ");
12074: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
12075: fprintf(ficresplb,"\n");
12076:
1.218 brouard 12077:
12078: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
12079:
12080: agebase=ageminpar;
12081: agelim=agemaxpar;
12082:
12083:
1.227 brouard 12084: i1=pow(2,cptcoveff);
1.218 brouard 12085: if (cptcovn < 1){i1=1;}
1.227 brouard 12086:
1.238 brouard 12087: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.338 brouard 12088: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
12089: k=TKresult[nres];
12090: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
12091: /* if(i1 != 1 && TKresult[nres]!= k) */
12092: /* continue; */
12093: /* /\*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*\/ */
1.238 brouard 12094: fprintf(ficresplb,"#******");
12095: printf("#******");
12096: fprintf(ficlog,"#******");
1.338 brouard 12097: 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) */
12098: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12099: fprintf(ficresplb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12100: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12101: }
1.338 brouard 12102: /* for(j=1;j<=cptcoveff ;j++) {/\* all covariates *\/ */
12103: /* fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12104: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12105: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12106: /* } */
12107: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
12108: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12109: /* fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12110: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12111: /* } */
1.238 brouard 12112: fprintf(ficresplb,"******\n");
12113: printf("******\n");
12114: fprintf(ficlog,"******\n");
12115: if(invalidvarcomb[k]){
12116: printf("\nCombination (%d) ignored because no cases \n",k);
12117: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
12118: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
12119: continue;
12120: }
1.218 brouard 12121:
1.238 brouard 12122: fprintf(ficresplb,"#Age ");
1.338 brouard 12123: for(j=1;j<=cptcovs;j++) {
12124: fprintf(ficresplb,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12125: }
12126: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
12127: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 12128:
12129:
1.238 brouard 12130: for (age=agebase; age<=agelim; age++){
12131: /* for (age=agebase; age<=agebase; age++){ */
12132: if(mobilavproj > 0){
12133: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
12134: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 12135: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 12136: }else if (mobilavproj == 0){
12137: 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);
12138: 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);
12139: exit(1);
12140: }else{
12141: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 12142: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 12143: /* printf("TOTOT\n"); */
12144: /* exit(1); */
1.238 brouard 12145: }
12146: fprintf(ficresplb,"%.0f ",age );
1.338 brouard 12147: for(j=1;j<=cptcovs;j++)
12148: fprintf(ficresplb,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12149: tot=0.;
12150: for(i=1; i<=nlstate;i++){
12151: tot += bprlim[i][i];
12152: fprintf(ficresplb," %.5f", bprlim[i][i]);
12153: }
12154: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
12155: } /* Age */
12156: /* was end of cptcod */
1.255 brouard 12157: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.338 brouard 12158: /* } /\* end of any combination *\/ */
1.238 brouard 12159: } /* end of nres */
1.218 brouard 12160: /* hBijx(p, bage, fage); */
12161: /* fclose(ficrespijb); */
12162:
12163: return 0;
1.217 brouard 12164: }
1.218 brouard 12165:
1.180 brouard 12166: int hPijx(double *p, int bage, int fage){
12167: /*------------- h Pij x at various ages ------------*/
1.336 brouard 12168: /* to be optimized with precov */
1.180 brouard 12169: int stepsize;
12170: int agelim;
12171: int hstepm;
12172: int nhstepm;
1.235 brouard 12173: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 12174:
12175: double agedeb;
12176: double ***p3mat;
12177:
1.337 brouard 12178: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
12179: if((ficrespij=fopen(filerespij,"w"))==NULL) {
12180: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
12181: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
12182: }
12183: printf("Computing pij: result on file '%s' \n", filerespij);
12184: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
12185:
12186: stepsize=(int) (stepm+YEARM-1)/YEARM;
12187: /*if (stepm<=24) stepsize=2;*/
12188:
12189: agelim=AGESUP;
12190: hstepm=stepsize*YEARM; /* Every year of age */
12191: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
12192:
12193: /* hstepm=1; aff par mois*/
12194: pstamp(ficrespij);
12195: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
12196: i1= pow(2,cptcoveff);
12197: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
12198: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
12199: /* k=k+1; */
12200: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
12201: k=TKresult[nres];
1.338 brouard 12202: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 12203: /* for(k=1; k<=i1;k++){ */
12204: /* if(i1 != 1 && TKresult[nres]!= k) */
12205: /* continue; */
12206: fprintf(ficrespij,"\n#****** ");
12207: for(j=1;j<=cptcovs;j++){
12208: fprintf(ficrespij," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12209: /* fprintf(ficrespij,"@wV%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12210: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
12211: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12212: /* fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12213: }
12214: fprintf(ficrespij,"******\n");
12215:
12216: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
12217: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
12218: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
12219:
12220: /* nhstepm=nhstepm*YEARM; aff par mois*/
12221:
12222: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12223: oldm=oldms;savm=savms;
12224: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
12225: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
12226: for(i=1; i<=nlstate;i++)
12227: for(j=1; j<=nlstate+ndeath;j++)
12228: fprintf(ficrespij," %1d-%1d",i,j);
12229: fprintf(ficrespij,"\n");
12230: for (h=0; h<=nhstepm; h++){
12231: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
12232: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.183 brouard 12233: for(i=1; i<=nlstate;i++)
12234: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 12235: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.183 brouard 12236: fprintf(ficrespij,"\n");
12237: }
1.337 brouard 12238: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12239: fprintf(ficrespij,"\n");
1.180 brouard 12240: }
1.337 brouard 12241: }
12242: /*}*/
12243: return 0;
1.180 brouard 12244: }
1.218 brouard 12245:
12246: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 12247: /*------------- h Bij x at various ages ------------*/
1.336 brouard 12248: /* To be optimized with precov */
1.217 brouard 12249: int stepsize;
1.218 brouard 12250: /* int agelim; */
12251: int ageminl;
1.217 brouard 12252: int hstepm;
12253: int nhstepm;
1.238 brouard 12254: int h, i, i1, j, k, nres;
1.218 brouard 12255:
1.217 brouard 12256: double agedeb;
12257: double ***p3mat;
1.218 brouard 12258:
12259: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
12260: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
12261: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
12262: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
12263: }
12264: printf("Computing pij back: result on file '%s' \n", filerespijb);
12265: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
12266:
12267: stepsize=(int) (stepm+YEARM-1)/YEARM;
12268: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 12269:
1.218 brouard 12270: /* agelim=AGESUP; */
1.289 brouard 12271: ageminl=AGEINF; /* was 30 */
1.218 brouard 12272: hstepm=stepsize*YEARM; /* Every year of age */
12273: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
12274:
12275: /* hstepm=1; aff par mois*/
12276: pstamp(ficrespijb);
1.255 brouard 12277: 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 12278: i1= pow(2,cptcoveff);
1.218 brouard 12279: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
12280: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
12281: /* k=k+1; */
1.238 brouard 12282: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 12283: k=TKresult[nres];
1.338 brouard 12284: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 12285: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
12286: /* if(i1 != 1 && TKresult[nres]!= k) */
12287: /* continue; */
12288: fprintf(ficrespijb,"\n#****** ");
12289: for(j=1;j<=cptcovs;j++){
1.338 brouard 12290: fprintf(ficrespijb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.337 brouard 12291: /* for(j=1;j<=cptcoveff;j++) */
12292: /* fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12293: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
12294: /* fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12295: }
12296: fprintf(ficrespijb,"******\n");
12297: if(invalidvarcomb[k]){ /* Is it necessary here? */
12298: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
12299: continue;
12300: }
12301:
12302: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
12303: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
12304: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
12305: 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 */
12306: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
12307:
12308: /* nhstepm=nhstepm*YEARM; aff par mois*/
12309:
12310: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
12311: /* and memory limitations if stepm is small */
12312:
12313: /* oldm=oldms;savm=savms; */
12314: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
12315: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
12316: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
12317: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
12318: for(i=1; i<=nlstate;i++)
12319: for(j=1; j<=nlstate+ndeath;j++)
12320: fprintf(ficrespijb," %1d-%1d",i,j);
12321: fprintf(ficrespijb,"\n");
12322: for (h=0; h<=nhstepm; h++){
12323: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
12324: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
12325: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
1.217 brouard 12326: for(i=1; i<=nlstate;i++)
12327: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 12328: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.217 brouard 12329: fprintf(ficrespijb,"\n");
1.337 brouard 12330: }
12331: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12332: fprintf(ficrespijb,"\n");
12333: } /* end age deb */
12334: /* } /\* end combination *\/ */
1.238 brouard 12335: } /* end nres */
1.218 brouard 12336: return 0;
12337: } /* hBijx */
1.217 brouard 12338:
1.180 brouard 12339:
1.136 brouard 12340: /***********************************************/
12341: /**************** Main Program *****************/
12342: /***********************************************/
12343:
12344: int main(int argc, char *argv[])
12345: {
12346: #ifdef GSL
12347: const gsl_multimin_fminimizer_type *T;
12348: size_t iteri = 0, it;
12349: int rval = GSL_CONTINUE;
12350: int status = GSL_SUCCESS;
12351: double ssval;
12352: #endif
12353: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 12354: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
12355: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 12356: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 12357: int jj, ll, li, lj, lk;
1.136 brouard 12358: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 12359: int num_filled;
1.136 brouard 12360: int itimes;
12361: int NDIM=2;
12362: int vpopbased=0;
1.235 brouard 12363: int nres=0;
1.258 brouard 12364: int endishere=0;
1.277 brouard 12365: int noffset=0;
1.274 brouard 12366: int ncurrv=0; /* Temporary variable */
12367:
1.164 brouard 12368: char ca[32], cb[32];
1.136 brouard 12369: /* FILE *fichtm; *//* Html File */
12370: /* FILE *ficgp;*/ /*Gnuplot File */
12371: struct stat info;
1.191 brouard 12372: double agedeb=0.;
1.194 brouard 12373:
12374: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 12375: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 12376:
1.165 brouard 12377: double fret;
1.191 brouard 12378: double dum=0.; /* Dummy variable */
1.136 brouard 12379: double ***p3mat;
1.218 brouard 12380: /* double ***mobaverage; */
1.319 brouard 12381: double wald;
1.164 brouard 12382:
12383: char line[MAXLINE];
1.197 brouard 12384: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
12385:
1.234 brouard 12386: char modeltemp[MAXLINE];
1.332 brouard 12387: char resultline[MAXLINE], resultlineori[MAXLINE];
1.230 brouard 12388:
1.136 brouard 12389: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 12390: char *tok, *val; /* pathtot */
1.334 brouard 12391: /* int firstobs=1, lastobs=10; /\* nobs = lastobs-firstobs declared globally ;*\/ */
1.195 brouard 12392: int c, h , cpt, c2;
1.191 brouard 12393: int jl=0;
12394: int i1, j1, jk, stepsize=0;
1.194 brouard 12395: int count=0;
12396:
1.164 brouard 12397: int *tab;
1.136 brouard 12398: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 12399: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
12400: /* double anprojf, mprojf, jprojf; */
12401: /* double jintmean,mintmean,aintmean; */
12402: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
12403: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
12404: double yrfproj= 10.0; /* Number of years of forward projections */
12405: double yrbproj= 10.0; /* Number of years of backward projections */
12406: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 12407: int mobilav=0,popforecast=0;
1.191 brouard 12408: int hstepm=0, nhstepm=0;
1.136 brouard 12409: int agemortsup;
12410: float sumlpop=0.;
12411: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
12412: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
12413:
1.191 brouard 12414: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 12415: double ftolpl=FTOL;
12416: double **prlim;
1.217 brouard 12417: double **bprlim;
1.317 brouard 12418: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
12419: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 12420: double ***paramstart; /* Matrix of starting parameter values */
12421: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 12422: double **matcov; /* Matrix of covariance */
1.203 brouard 12423: double **hess; /* Hessian matrix */
1.136 brouard 12424: double ***delti3; /* Scale */
12425: double *delti; /* Scale */
12426: double ***eij, ***vareij;
12427: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 12428:
1.136 brouard 12429: double *epj, vepp;
1.164 brouard 12430:
1.273 brouard 12431: double dateprev1, dateprev2;
1.296 brouard 12432: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
12433: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
12434:
1.217 brouard 12435:
1.136 brouard 12436: double **ximort;
1.145 brouard 12437: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 12438: int *dcwave;
12439:
1.164 brouard 12440: char z[1]="c";
1.136 brouard 12441:
12442: /*char *strt;*/
12443: char strtend[80];
1.126 brouard 12444:
1.164 brouard 12445:
1.126 brouard 12446: /* setlocale (LC_ALL, ""); */
12447: /* bindtextdomain (PACKAGE, LOCALEDIR); */
12448: /* textdomain (PACKAGE); */
12449: /* setlocale (LC_CTYPE, ""); */
12450: /* setlocale (LC_MESSAGES, ""); */
12451:
12452: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 12453: rstart_time = time(NULL);
12454: /* (void) gettimeofday(&start_time,&tzp);*/
12455: start_time = *localtime(&rstart_time);
1.126 brouard 12456: curr_time=start_time;
1.157 brouard 12457: /*tml = *localtime(&start_time.tm_sec);*/
12458: /* strcpy(strstart,asctime(&tml)); */
12459: strcpy(strstart,asctime(&start_time));
1.126 brouard 12460:
12461: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 12462: /* tp.tm_sec = tp.tm_sec +86400; */
12463: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 12464: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
12465: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
12466: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 12467: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 12468: /* strt=asctime(&tmg); */
12469: /* printf("Time(after) =%s",strstart); */
12470: /* (void) time (&time_value);
12471: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
12472: * tm = *localtime(&time_value);
12473: * strstart=asctime(&tm);
12474: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
12475: */
12476:
12477: nberr=0; /* Number of errors and warnings */
12478: nbwarn=0;
1.184 brouard 12479: #ifdef WIN32
12480: _getcwd(pathcd, size);
12481: #else
1.126 brouard 12482: getcwd(pathcd, size);
1.184 brouard 12483: #endif
1.191 brouard 12484: syscompilerinfo(0);
1.196 brouard 12485: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 12486: if(argc <=1){
12487: printf("\nEnter the parameter file name: ");
1.205 brouard 12488: if(!fgets(pathr,FILENAMELENGTH,stdin)){
12489: printf("ERROR Empty parameter file name\n");
12490: goto end;
12491: }
1.126 brouard 12492: i=strlen(pathr);
12493: if(pathr[i-1]=='\n')
12494: pathr[i-1]='\0';
1.156 brouard 12495: i=strlen(pathr);
1.205 brouard 12496: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 12497: pathr[i-1]='\0';
1.205 brouard 12498: }
12499: i=strlen(pathr);
12500: if( i==0 ){
12501: printf("ERROR Empty parameter file name\n");
12502: goto end;
12503: }
12504: for (tok = pathr; tok != NULL; ){
1.126 brouard 12505: printf("Pathr |%s|\n",pathr);
12506: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
12507: printf("val= |%s| pathr=%s\n",val,pathr);
12508: strcpy (pathtot, val);
12509: if(pathr[0] == '\0') break; /* Dirty */
12510: }
12511: }
1.281 brouard 12512: else if (argc<=2){
12513: strcpy(pathtot,argv[1]);
12514: }
1.126 brouard 12515: else{
12516: strcpy(pathtot,argv[1]);
1.281 brouard 12517: strcpy(z,argv[2]);
12518: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 12519: }
12520: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
12521: /*cygwin_split_path(pathtot,path,optionfile);
12522: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
12523: /* cutv(path,optionfile,pathtot,'\\');*/
12524:
12525: /* Split argv[0], imach program to get pathimach */
12526: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
12527: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12528: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12529: /* strcpy(pathimach,argv[0]); */
12530: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
12531: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
12532: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 12533: #ifdef WIN32
12534: _chdir(path); /* Can be a relative path */
12535: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
12536: #else
1.126 brouard 12537: chdir(path); /* Can be a relative path */
1.184 brouard 12538: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
12539: #endif
12540: printf("Current directory %s!\n",pathcd);
1.126 brouard 12541: strcpy(command,"mkdir ");
12542: strcat(command,optionfilefiname);
12543: if((outcmd=system(command)) != 0){
1.169 brouard 12544: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 12545: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
12546: /* fclose(ficlog); */
12547: /* exit(1); */
12548: }
12549: /* if((imk=mkdir(optionfilefiname))<0){ */
12550: /* perror("mkdir"); */
12551: /* } */
12552:
12553: /*-------- arguments in the command line --------*/
12554:
1.186 brouard 12555: /* Main Log file */
1.126 brouard 12556: strcat(filelog, optionfilefiname);
12557: strcat(filelog,".log"); /* */
12558: if((ficlog=fopen(filelog,"w"))==NULL) {
12559: printf("Problem with logfile %s\n",filelog);
12560: goto end;
12561: }
12562: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 12563: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 12564: fprintf(ficlog,"\nEnter the parameter file name: \n");
12565: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
12566: path=%s \n\
12567: optionfile=%s\n\
12568: optionfilext=%s\n\
1.156 brouard 12569: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 12570:
1.197 brouard 12571: syscompilerinfo(1);
1.167 brouard 12572:
1.126 brouard 12573: printf("Local time (at start):%s",strstart);
12574: fprintf(ficlog,"Local time (at start): %s",strstart);
12575: fflush(ficlog);
12576: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 12577: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 12578:
12579: /* */
12580: strcpy(fileres,"r");
12581: strcat(fileres, optionfilefiname);
1.201 brouard 12582: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 12583: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 12584: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 12585:
1.186 brouard 12586: /* Main ---------arguments file --------*/
1.126 brouard 12587:
12588: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 12589: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
12590: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 12591: fflush(ficlog);
1.149 brouard 12592: /* goto end; */
12593: exit(70);
1.126 brouard 12594: }
12595:
12596: strcpy(filereso,"o");
1.201 brouard 12597: strcat(filereso,fileresu);
1.126 brouard 12598: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
12599: printf("Problem with Output resultfile: %s\n", filereso);
12600: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
12601: fflush(ficlog);
12602: goto end;
12603: }
1.278 brouard 12604: /*-------- Rewriting parameter file ----------*/
12605: strcpy(rfileres,"r"); /* "Rparameterfile */
12606: strcat(rfileres,optionfilefiname); /* Parameter file first name */
12607: strcat(rfileres,"."); /* */
12608: strcat(rfileres,optionfilext); /* Other files have txt extension */
12609: if((ficres =fopen(rfileres,"w"))==NULL) {
12610: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
12611: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
12612: fflush(ficlog);
12613: goto end;
12614: }
12615: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 12616:
1.278 brouard 12617:
1.126 brouard 12618: /* Reads comments: lines beginning with '#' */
12619: numlinepar=0;
1.277 brouard 12620: /* Is it a BOM UTF-8 Windows file? */
12621: /* First parameter line */
1.197 brouard 12622: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 12623: noffset=0;
12624: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
12625: {
12626: noffset=noffset+3;
12627: printf("# File is an UTF8 Bom.\n"); // 0xBF
12628: }
1.302 brouard 12629: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
12630: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 12631: {
12632: noffset=noffset+2;
12633: printf("# File is an UTF16BE BOM file\n");
12634: }
12635: else if( line[0] == 0 && line[1] == 0)
12636: {
12637: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
12638: noffset=noffset+4;
12639: printf("# File is an UTF16BE BOM file\n");
12640: }
12641: } else{
12642: ;/*printf(" Not a BOM file\n");*/
12643: }
12644:
1.197 brouard 12645: /* If line starts with a # it is a comment */
1.277 brouard 12646: if (line[noffset] == '#') {
1.197 brouard 12647: numlinepar++;
12648: fputs(line,stdout);
12649: fputs(line,ficparo);
1.278 brouard 12650: fputs(line,ficres);
1.197 brouard 12651: fputs(line,ficlog);
12652: continue;
12653: }else
12654: break;
12655: }
12656: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
12657: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
12658: if (num_filled != 5) {
12659: printf("Should be 5 parameters\n");
1.283 brouard 12660: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 12661: }
1.126 brouard 12662: numlinepar++;
1.197 brouard 12663: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 12664: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12665: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12666: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 12667: }
12668: /* Second parameter line */
12669: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 12670: /* while(fscanf(ficpar,"%[^\n]", line)) { */
12671: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 12672: if (line[0] == '#') {
12673: numlinepar++;
1.283 brouard 12674: printf("%s",line);
12675: fprintf(ficres,"%s",line);
12676: fprintf(ficparo,"%s",line);
12677: fprintf(ficlog,"%s",line);
1.197 brouard 12678: continue;
12679: }else
12680: break;
12681: }
1.223 brouard 12682: 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", \
12683: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
12684: if (num_filled != 11) {
12685: 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 12686: printf("but line=%s\n",line);
1.283 brouard 12687: 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");
12688: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 12689: }
1.286 brouard 12690: if( lastpass > maxwav){
12691: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12692: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12693: fflush(ficlog);
12694: goto end;
12695: }
12696: 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 12697: 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 12698: 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 12699: 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 12700: }
1.203 brouard 12701: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 12702: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 12703: /* Third parameter line */
12704: while(fgets(line, MAXLINE, ficpar)) {
12705: /* If line starts with a # it is a comment */
12706: if (line[0] == '#') {
12707: numlinepar++;
1.283 brouard 12708: printf("%s",line);
12709: fprintf(ficres,"%s",line);
12710: fprintf(ficparo,"%s",line);
12711: fprintf(ficlog,"%s",line);
1.197 brouard 12712: continue;
12713: }else
12714: break;
12715: }
1.201 brouard 12716: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 12717: if (num_filled != 1){
1.302 brouard 12718: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
12719: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 12720: model[0]='\0';
12721: goto end;
12722: }
12723: else{
12724: if (model[0]=='+'){
12725: for(i=1; i<=strlen(model);i++)
12726: modeltemp[i-1]=model[i];
1.201 brouard 12727: strcpy(model,modeltemp);
1.197 brouard 12728: }
12729: }
1.338 brouard 12730: /* printf(" model=1+age%s modeltemp= %s, model=1+age+%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 12731: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 12732: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
12733: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
12734: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 12735: }
12736: /* 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); */
12737: /* numlinepar=numlinepar+3; /\* In general *\/ */
12738: /* 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 12739: /* 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); */
12740: /* 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 12741: fflush(ficlog);
1.190 brouard 12742: /* if(model[0]=='#'|| model[0]== '\0'){ */
12743: if(model[0]=='#'){
1.279 brouard 12744: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
12745: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
12746: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 12747: if(mle != -1){
1.279 brouard 12748: 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 12749: exit(1);
12750: }
12751: }
1.126 brouard 12752: while((c=getc(ficpar))=='#' && c!= EOF){
12753: ungetc(c,ficpar);
12754: fgets(line, MAXLINE, ficpar);
12755: numlinepar++;
1.195 brouard 12756: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
12757: z[0]=line[1];
1.342 brouard 12758: }else if(line[1]=='d'){ /* For debugging individual values of covariates in ficresilk */
1.343 brouard 12759: debugILK=1;printf("DebugILK\n");
1.195 brouard 12760: }
12761: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 12762: fputs(line, stdout);
12763: //puts(line);
1.126 brouard 12764: fputs(line,ficparo);
12765: fputs(line,ficlog);
12766: }
12767: ungetc(c,ficpar);
12768:
12769:
1.290 brouard 12770: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
12771: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
12772: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
1.341 brouard 12773: /* if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /\**< Time varying covariate (dummy and quantitative)*\/ */
12774: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs); /**< Might be better */
1.136 brouard 12775: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
12776: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
12777: v1+v2*age+v2*v3 makes cptcovn = 3
12778: */
12779: if (strlen(model)>1)
1.187 brouard 12780: 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 12781: else
1.187 brouard 12782: ncovmodel=2; /* Constant and age */
1.133 brouard 12783: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
12784: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 12785: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
12786: 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);
12787: 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);
12788: fflush(stdout);
12789: fclose (ficlog);
12790: goto end;
12791: }
1.126 brouard 12792: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12793: delti=delti3[1][1];
12794: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
12795: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 12796: /* We could also provide initial parameters values giving by simple logistic regression
12797: * only one way, that is without matrix product. We will have nlstate maximizations */
12798: /* for(i=1;i<nlstate;i++){ */
12799: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12800: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12801: /* } */
1.126 brouard 12802: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 12803: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
12804: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12805: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12806: fclose (ficparo);
12807: fclose (ficlog);
12808: goto end;
12809: exit(0);
1.220 brouard 12810: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 12811: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 12812: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
12813: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12814: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12815: matcov=matrix(1,npar,1,npar);
1.203 brouard 12816: hess=matrix(1,npar,1,npar);
1.220 brouard 12817: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 12818: /* Read guessed parameters */
1.126 brouard 12819: /* Reads comments: lines beginning with '#' */
12820: while((c=getc(ficpar))=='#' && c!= EOF){
12821: ungetc(c,ficpar);
12822: fgets(line, MAXLINE, ficpar);
12823: numlinepar++;
1.141 brouard 12824: fputs(line,stdout);
1.126 brouard 12825: fputs(line,ficparo);
12826: fputs(line,ficlog);
12827: }
12828: ungetc(c,ficpar);
12829:
12830: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 12831: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 12832: for(i=1; i <=nlstate; i++){
1.234 brouard 12833: j=0;
1.126 brouard 12834: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 12835: if(jj==i) continue;
12836: j++;
1.292 brouard 12837: while((c=getc(ficpar))=='#' && c!= EOF){
12838: ungetc(c,ficpar);
12839: fgets(line, MAXLINE, ficpar);
12840: numlinepar++;
12841: fputs(line,stdout);
12842: fputs(line,ficparo);
12843: fputs(line,ficlog);
12844: }
12845: ungetc(c,ficpar);
1.234 brouard 12846: fscanf(ficpar,"%1d%1d",&i1,&j1);
12847: if ((i1 != i) || (j1 != jj)){
12848: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 12849: It might be a problem of design; if ncovcol and the model are correct\n \
12850: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 12851: exit(1);
12852: }
12853: fprintf(ficparo,"%1d%1d",i1,j1);
12854: if(mle==1)
12855: printf("%1d%1d",i,jj);
12856: fprintf(ficlog,"%1d%1d",i,jj);
12857: for(k=1; k<=ncovmodel;k++){
12858: fscanf(ficpar," %lf",¶m[i][j][k]);
12859: if(mle==1){
12860: printf(" %lf",param[i][j][k]);
12861: fprintf(ficlog," %lf",param[i][j][k]);
12862: }
12863: else
12864: fprintf(ficlog," %lf",param[i][j][k]);
12865: fprintf(ficparo," %lf",param[i][j][k]);
12866: }
12867: fscanf(ficpar,"\n");
12868: numlinepar++;
12869: if(mle==1)
12870: printf("\n");
12871: fprintf(ficlog,"\n");
12872: fprintf(ficparo,"\n");
1.126 brouard 12873: }
12874: }
12875: fflush(ficlog);
1.234 brouard 12876:
1.251 brouard 12877: /* Reads parameters values */
1.126 brouard 12878: p=param[1][1];
1.251 brouard 12879: pstart=paramstart[1][1];
1.126 brouard 12880:
12881: /* Reads comments: lines beginning with '#' */
12882: while((c=getc(ficpar))=='#' && c!= EOF){
12883: ungetc(c,ficpar);
12884: fgets(line, MAXLINE, ficpar);
12885: numlinepar++;
1.141 brouard 12886: fputs(line,stdout);
1.126 brouard 12887: fputs(line,ficparo);
12888: fputs(line,ficlog);
12889: }
12890: ungetc(c,ficpar);
12891:
12892: for(i=1; i <=nlstate; i++){
12893: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 12894: fscanf(ficpar,"%1d%1d",&i1,&j1);
12895: if ( (i1-i) * (j1-j) != 0){
12896: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
12897: exit(1);
12898: }
12899: printf("%1d%1d",i,j);
12900: fprintf(ficparo,"%1d%1d",i1,j1);
12901: fprintf(ficlog,"%1d%1d",i1,j1);
12902: for(k=1; k<=ncovmodel;k++){
12903: fscanf(ficpar,"%le",&delti3[i][j][k]);
12904: printf(" %le",delti3[i][j][k]);
12905: fprintf(ficparo," %le",delti3[i][j][k]);
12906: fprintf(ficlog," %le",delti3[i][j][k]);
12907: }
12908: fscanf(ficpar,"\n");
12909: numlinepar++;
12910: printf("\n");
12911: fprintf(ficparo,"\n");
12912: fprintf(ficlog,"\n");
1.126 brouard 12913: }
12914: }
12915: fflush(ficlog);
1.234 brouard 12916:
1.145 brouard 12917: /* Reads covariance matrix */
1.126 brouard 12918: delti=delti3[1][1];
1.220 brouard 12919:
12920:
1.126 brouard 12921: /* 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 12922:
1.126 brouard 12923: /* Reads comments: lines beginning with '#' */
12924: while((c=getc(ficpar))=='#' && c!= EOF){
12925: ungetc(c,ficpar);
12926: fgets(line, MAXLINE, ficpar);
12927: numlinepar++;
1.141 brouard 12928: fputs(line,stdout);
1.126 brouard 12929: fputs(line,ficparo);
12930: fputs(line,ficlog);
12931: }
12932: ungetc(c,ficpar);
1.220 brouard 12933:
1.126 brouard 12934: matcov=matrix(1,npar,1,npar);
1.203 brouard 12935: hess=matrix(1,npar,1,npar);
1.131 brouard 12936: for(i=1; i <=npar; i++)
12937: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 12938:
1.194 brouard 12939: /* Scans npar lines */
1.126 brouard 12940: for(i=1; i <=npar; i++){
1.226 brouard 12941: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 12942: if(count != 3){
1.226 brouard 12943: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12944: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12945: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12946: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12947: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12948: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12949: exit(1);
1.220 brouard 12950: }else{
1.226 brouard 12951: if(mle==1)
12952: printf("%1d%1d%d",i1,j1,jk);
12953: }
12954: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
12955: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 12956: for(j=1; j <=i; j++){
1.226 brouard 12957: fscanf(ficpar," %le",&matcov[i][j]);
12958: if(mle==1){
12959: printf(" %.5le",matcov[i][j]);
12960: }
12961: fprintf(ficlog," %.5le",matcov[i][j]);
12962: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 12963: }
12964: fscanf(ficpar,"\n");
12965: numlinepar++;
12966: if(mle==1)
1.220 brouard 12967: printf("\n");
1.126 brouard 12968: fprintf(ficlog,"\n");
12969: fprintf(ficparo,"\n");
12970: }
1.194 brouard 12971: /* End of read covariance matrix npar lines */
1.126 brouard 12972: for(i=1; i <=npar; i++)
12973: for(j=i+1;j<=npar;j++)
1.226 brouard 12974: matcov[i][j]=matcov[j][i];
1.126 brouard 12975:
12976: if(mle==1)
12977: printf("\n");
12978: fprintf(ficlog,"\n");
12979:
12980: fflush(ficlog);
12981:
12982: } /* End of mle != -3 */
1.218 brouard 12983:
1.186 brouard 12984: /* Main data
12985: */
1.290 brouard 12986: nobs=lastobs-firstobs+1; /* was = lastobs;*/
12987: /* num=lvector(1,n); */
12988: /* moisnais=vector(1,n); */
12989: /* annais=vector(1,n); */
12990: /* moisdc=vector(1,n); */
12991: /* andc=vector(1,n); */
12992: /* weight=vector(1,n); */
12993: /* agedc=vector(1,n); */
12994: /* cod=ivector(1,n); */
12995: /* for(i=1;i<=n;i++){ */
12996: num=lvector(firstobs,lastobs);
12997: moisnais=vector(firstobs,lastobs);
12998: annais=vector(firstobs,lastobs);
12999: moisdc=vector(firstobs,lastobs);
13000: andc=vector(firstobs,lastobs);
13001: weight=vector(firstobs,lastobs);
13002: agedc=vector(firstobs,lastobs);
13003: cod=ivector(firstobs,lastobs);
13004: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 13005: num[i]=0;
13006: moisnais[i]=0;
13007: annais[i]=0;
13008: moisdc[i]=0;
13009: andc[i]=0;
13010: agedc[i]=0;
13011: cod[i]=0;
13012: weight[i]=1.0; /* Equal weights, 1 by default */
13013: }
1.290 brouard 13014: mint=matrix(1,maxwav,firstobs,lastobs);
13015: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 13016: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.336 brouard 13017: /* printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel)); */
1.126 brouard 13018: tab=ivector(1,NCOVMAX);
1.144 brouard 13019: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 13020: 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 13021:
1.136 brouard 13022: /* Reads data from file datafile */
13023: if (readdata(datafile, firstobs, lastobs, &imx)==1)
13024: goto end;
13025:
13026: /* Calculation of the number of parameters from char model */
1.234 brouard 13027: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 13028: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
13029: k=3 V4 Tvar[k=3]= 4 (from V4)
13030: k=2 V1 Tvar[k=2]= 1 (from V1)
13031: k=1 Tvar[1]=2 (from V2)
1.234 brouard 13032: */
13033:
13034: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
13035: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 brouard 13036: TnsdVar=ivector(1,NCOVMAX); /* */
1.335 brouard 13037: /* for(i=1; i<=NCOVMAX;i++) TnsdVar[i]=3; */
1.234 brouard 13038: TvarsD=ivector(1,NCOVMAX); /* */
13039: TvarsQind=ivector(1,NCOVMAX); /* */
13040: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 13041: TvarF=ivector(1,NCOVMAX); /* */
13042: TvarFind=ivector(1,NCOVMAX); /* */
13043: TvarV=ivector(1,NCOVMAX); /* */
13044: TvarVind=ivector(1,NCOVMAX); /* */
13045: TvarA=ivector(1,NCOVMAX); /* */
13046: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 13047: TvarFD=ivector(1,NCOVMAX); /* */
13048: TvarFDind=ivector(1,NCOVMAX); /* */
13049: TvarFQ=ivector(1,NCOVMAX); /* */
13050: TvarFQind=ivector(1,NCOVMAX); /* */
13051: TvarVD=ivector(1,NCOVMAX); /* */
13052: TvarVDind=ivector(1,NCOVMAX); /* */
13053: TvarVQ=ivector(1,NCOVMAX); /* */
13054: TvarVQind=ivector(1,NCOVMAX); /* */
1.339 brouard 13055: TvarVV=ivector(1,NCOVMAX); /* */
13056: TvarVVind=ivector(1,NCOVMAX); /* */
1.231 brouard 13057:
1.230 brouard 13058: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 13059: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 13060: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
13061: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
13062: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 13063: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
13064: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
13065: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
13066: */
13067: /* For model-covariate k tells which data-covariate to use but
13068: because this model-covariate is a construction we invent a new column
13069: ncovcol + k1
13070: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
13071: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 13072: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
13073: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 13074: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
13075: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 13076: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 13077: */
1.145 brouard 13078: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
13079: 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 13080: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
13081: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.330 brouard 13082: Tvardk=imatrix(1,NCOVMAX,1,2);
1.145 brouard 13083: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 13084: 4 covariates (3 plus signs)
13085: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 13086: */
13087: for(i=1;i<NCOVMAX;i++)
13088: Tage[i]=0;
1.230 brouard 13089: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 13090: * individual dummy, fixed or varying:
13091: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
13092: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 13093: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
13094: * V1 df, V2 qf, V3 & V4 dv, V5 qv
13095: * Tmodelind[1]@9={9,0,3,2,}*/
13096: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
13097: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 13098: * individual quantitative, fixed or varying:
13099: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
13100: * 3, 1, 0, 0, 0, 0, 0, 0},
13101: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 13102: /* Main decodemodel */
13103:
1.187 brouard 13104:
1.223 brouard 13105: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 13106: goto end;
13107:
1.137 brouard 13108: if((double)(lastobs-imx)/(double)imx > 1.10){
13109: nbwarn++;
13110: 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);
13111: 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);
13112: }
1.136 brouard 13113: /* if(mle==1){*/
1.137 brouard 13114: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
13115: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 13116: }
13117:
13118: /*-calculation of age at interview from date of interview and age at death -*/
13119: agev=matrix(1,maxwav,1,imx);
13120:
13121: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
13122: goto end;
13123:
1.126 brouard 13124:
1.136 brouard 13125: agegomp=(int)agemin;
1.290 brouard 13126: free_vector(moisnais,firstobs,lastobs);
13127: free_vector(annais,firstobs,lastobs);
1.126 brouard 13128: /* free_matrix(mint,1,maxwav,1,n);
13129: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 13130: /* free_vector(moisdc,1,n); */
13131: /* free_vector(andc,1,n); */
1.145 brouard 13132: /* */
13133:
1.126 brouard 13134: wav=ivector(1,imx);
1.214 brouard 13135: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
13136: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
13137: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
13138: 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.*/
13139: bh=imatrix(1,lastpass-firstpass+2,1,imx);
13140: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 13141:
13142: /* Concatenates waves */
1.214 brouard 13143: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
13144: Death is a valid wave (if date is known).
13145: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
13146: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
13147: and mw[mi+1][i]. dh depends on stepm.
13148: */
13149:
1.126 brouard 13150: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 13151: /* Concatenates waves */
1.145 brouard 13152:
1.290 brouard 13153: free_vector(moisdc,firstobs,lastobs);
13154: free_vector(andc,firstobs,lastobs);
1.215 brouard 13155:
1.126 brouard 13156: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
13157: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
13158: ncodemax[1]=1;
1.145 brouard 13159: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 13160: cptcoveff=0;
1.220 brouard 13161: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
1.335 brouard 13162: 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 13163: }
13164:
13165: ncovcombmax=pow(2,cptcoveff);
1.338 brouard 13166: invalidvarcomb=ivector(0, ncovcombmax);
13167: for(i=0;i<ncovcombmax;i++)
1.227 brouard 13168: invalidvarcomb[i]=0;
13169:
1.211 brouard 13170: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 13171: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 13172: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 13173:
1.200 brouard 13174: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 13175: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 13176: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 13177: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
13178: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
13179: * (currently 0 or 1) in the data.
13180: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
13181: * corresponding modality (h,j).
13182: */
13183:
1.145 brouard 13184: h=0;
13185: /*if (cptcovn > 0) */
1.126 brouard 13186: m=pow(2,cptcoveff);
13187:
1.144 brouard 13188: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 13189: * For k=4 covariates, h goes from 1 to m=2**k
13190: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
13191: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 13192: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
13193: *______________________________ *______________________
13194: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
13195: * 2 2 1 1 1 * 1 0 0 0 1
13196: * 3 i=2 1 2 1 1 * 2 0 0 1 0
13197: * 4 2 2 1 1 * 3 0 0 1 1
13198: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
13199: * 6 2 1 2 1 * 5 0 1 0 1
13200: * 7 i=4 1 2 2 1 * 6 0 1 1 0
13201: * 8 2 2 2 1 * 7 0 1 1 1
13202: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
13203: * 10 2 1 1 2 * 9 1 0 0 1
13204: * 11 i=6 1 2 1 2 * 10 1 0 1 0
13205: * 12 2 2 1 2 * 11 1 0 1 1
13206: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
13207: * 14 2 1 2 2 * 13 1 1 0 1
13208: * 15 i=8 1 2 2 2 * 14 1 1 1 0
13209: * 16 2 2 2 2 * 15 1 1 1 1
13210: */
1.212 brouard 13211: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 13212: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
13213: * and the value of each covariate?
13214: * V1=1, V2=1, V3=2, V4=1 ?
13215: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
13216: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
13217: * In order to get the real value in the data, we use nbcode
13218: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
13219: * We are keeping this crazy system in order to be able (in the future?)
13220: * to have more than 2 values (0 or 1) for a covariate.
13221: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
13222: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
13223: * bbbbbbbb
13224: * 76543210
13225: * h-1 00000101 (6-1=5)
1.219 brouard 13226: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 13227: * &
13228: * 1 00000001 (1)
1.219 brouard 13229: * 00000000 = 1 & ((h-1) >> (k-1))
13230: * +1= 00000001 =1
1.211 brouard 13231: *
13232: * h=14, k=3 => h'=h-1=13, k'=k-1=2
13233: * h' 1101 =2^3+2^2+0x2^1+2^0
13234: * >>k' 11
13235: * & 00000001
13236: * = 00000001
13237: * +1 = 00000010=2 = codtabm(14,3)
13238: * Reverse h=6 and m=16?
13239: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
13240: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
13241: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
13242: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
13243: * V3=decodtabm(14,3,2**4)=2
13244: * h'=13 1101 =2^3+2^2+0x2^1+2^0
13245: *(h-1) >> (j-1) 0011 =13 >> 2
13246: * &1 000000001
13247: * = 000000001
13248: * +1= 000000010 =2
13249: * 2211
13250: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
13251: * V3=2
1.220 brouard 13252: * codtabm and decodtabm are identical
1.211 brouard 13253: */
13254:
1.145 brouard 13255:
13256: free_ivector(Ndum,-1,NCOVMAX);
13257:
13258:
1.126 brouard 13259:
1.186 brouard 13260: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 13261: strcpy(optionfilegnuplot,optionfilefiname);
13262: if(mle==-3)
1.201 brouard 13263: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 13264: strcat(optionfilegnuplot,".gp");
13265:
13266: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
13267: printf("Problem with file %s",optionfilegnuplot);
13268: }
13269: else{
1.204 brouard 13270: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 13271: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 13272: //fprintf(ficgp,"set missing 'NaNq'\n");
13273: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 13274: }
13275: /* fclose(ficgp);*/
1.186 brouard 13276:
13277:
13278: /* Initialisation of --------- index.htm --------*/
1.126 brouard 13279:
13280: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
13281: if(mle==-3)
1.201 brouard 13282: strcat(optionfilehtm,"-MORT_");
1.126 brouard 13283: strcat(optionfilehtm,".htm");
13284: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 13285: printf("Problem with %s \n",optionfilehtm);
13286: exit(0);
1.126 brouard 13287: }
13288:
13289: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
13290: strcat(optionfilehtmcov,"-cov.htm");
13291: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
13292: printf("Problem with %s \n",optionfilehtmcov), exit(0);
13293: }
13294: else{
13295: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
13296: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 13297: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 13298: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
13299: }
13300:
1.335 brouard 13301: fprintf(fichtm,"<html><head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n\
13302: <title>IMaCh %s</title></head>\n\
13303: <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n\
13304: <font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>\
13305: -EUROREVES-Institut de longévité-2013-2022-Japan Society for the Promotion of Sciences 日本学術振興会 \
13306: (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - \
13307: <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \n", optionfilehtm);
13308:
13309: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 13310: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 13311: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.337 brouard 13312: 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 13313: \n\
13314: <hr size=\"2\" color=\"#EC5E5E\">\
13315: <ul><li><h4>Parameter files</h4>\n\
13316: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
13317: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
13318: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
13319: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
13320: - Date and time at start: %s</ul>\n",\
1.335 brouard 13321: version,fullversion,optionfilehtm,optionfilehtm,title,datafile,datafile,firstpass,lastpass,stepm, weightopt, model, \
1.126 brouard 13322: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
13323: fileres,fileres,\
13324: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
13325: fflush(fichtm);
13326:
13327: strcpy(pathr,path);
13328: strcat(pathr,optionfilefiname);
1.184 brouard 13329: #ifdef WIN32
13330: _chdir(optionfilefiname); /* Move to directory named optionfile */
13331: #else
1.126 brouard 13332: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 13333: #endif
13334:
1.126 brouard 13335:
1.220 brouard 13336: /* Calculates basic frequencies. Computes observed prevalence at single age
13337: and for any valid combination of covariates
1.126 brouard 13338: and prints on file fileres'p'. */
1.251 brouard 13339: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 13340: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 13341:
13342: fprintf(fichtm,"\n");
1.286 brouard 13343: 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 13344: ftol, stepm);
13345: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
13346: ncurrv=1;
13347: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
13348: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
13349: ncurrv=i;
13350: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 13351: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 13352: ncurrv=i;
13353: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 13354: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 13355: ncurrv=i;
13356: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
13357: 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", \
13358: nlstate, ndeath, maxwav, mle, weightopt);
13359:
13360: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
13361: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
13362:
13363:
1.317 brouard 13364: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 13365: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
13366: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 13367: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 13368: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 13369: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13370: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13371: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13372: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 13373:
1.126 brouard 13374: /* For Powell, parameters are in a vector p[] starting at p[1]
13375: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
13376: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
13377:
13378: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 13379: /* For mortality only */
1.126 brouard 13380: if (mle==-3){
1.136 brouard 13381: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 13382: for(i=1;i<=NDIM;i++)
13383: for(j=1;j<=NDIM;j++)
13384: ximort[i][j]=0.;
1.186 brouard 13385: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 13386: cens=ivector(firstobs,lastobs);
13387: ageexmed=vector(firstobs,lastobs);
13388: agecens=vector(firstobs,lastobs);
13389: dcwave=ivector(firstobs,lastobs);
1.223 brouard 13390:
1.126 brouard 13391: for (i=1; i<=imx; i++){
13392: dcwave[i]=-1;
13393: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 13394: if (s[m][i]>nlstate) {
13395: dcwave[i]=m;
13396: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
13397: break;
13398: }
1.126 brouard 13399: }
1.226 brouard 13400:
1.126 brouard 13401: for (i=1; i<=imx; i++) {
13402: if (wav[i]>0){
1.226 brouard 13403: ageexmed[i]=agev[mw[1][i]][i];
13404: j=wav[i];
13405: agecens[i]=1.;
13406:
13407: if (ageexmed[i]> 1 && wav[i] > 0){
13408: agecens[i]=agev[mw[j][i]][i];
13409: cens[i]= 1;
13410: }else if (ageexmed[i]< 1)
13411: cens[i]= -1;
13412: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
13413: cens[i]=0 ;
1.126 brouard 13414: }
13415: else cens[i]=-1;
13416: }
13417:
13418: for (i=1;i<=NDIM;i++) {
13419: for (j=1;j<=NDIM;j++)
1.226 brouard 13420: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 13421: }
13422:
1.302 brouard 13423: p[1]=0.0268; p[NDIM]=0.083;
13424: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 13425:
13426:
1.136 brouard 13427: #ifdef GSL
13428: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 13429: #else
1.126 brouard 13430: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 13431: #endif
1.201 brouard 13432: strcpy(filerespow,"POW-MORT_");
13433: strcat(filerespow,fileresu);
1.126 brouard 13434: if((ficrespow=fopen(filerespow,"w"))==NULL) {
13435: printf("Problem with resultfile: %s\n", filerespow);
13436: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
13437: }
1.136 brouard 13438: #ifdef GSL
13439: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 13440: #else
1.126 brouard 13441: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 13442: #endif
1.126 brouard 13443: /* for (i=1;i<=nlstate;i++)
13444: for(j=1;j<=nlstate+ndeath;j++)
13445: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
13446: */
13447: fprintf(ficrespow,"\n");
1.136 brouard 13448: #ifdef GSL
13449: /* gsl starts here */
13450: T = gsl_multimin_fminimizer_nmsimplex;
13451: gsl_multimin_fminimizer *sfm = NULL;
13452: gsl_vector *ss, *x;
13453: gsl_multimin_function minex_func;
13454:
13455: /* Initial vertex size vector */
13456: ss = gsl_vector_alloc (NDIM);
13457:
13458: if (ss == NULL){
13459: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
13460: }
13461: /* Set all step sizes to 1 */
13462: gsl_vector_set_all (ss, 0.001);
13463:
13464: /* Starting point */
1.126 brouard 13465:
1.136 brouard 13466: x = gsl_vector_alloc (NDIM);
13467:
13468: if (x == NULL){
13469: gsl_vector_free(ss);
13470: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
13471: }
13472:
13473: /* Initialize method and iterate */
13474: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 13475: /* gsl_vector_set(x, 0, 0.0268); */
13476: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 13477: gsl_vector_set(x, 0, p[1]);
13478: gsl_vector_set(x, 1, p[2]);
13479:
13480: minex_func.f = &gompertz_f;
13481: minex_func.n = NDIM;
13482: minex_func.params = (void *)&p; /* ??? */
13483:
13484: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
13485: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
13486:
13487: printf("Iterations beginning .....\n\n");
13488: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
13489:
13490: iteri=0;
13491: while (rval == GSL_CONTINUE){
13492: iteri++;
13493: status = gsl_multimin_fminimizer_iterate(sfm);
13494:
13495: if (status) printf("error: %s\n", gsl_strerror (status));
13496: fflush(0);
13497:
13498: if (status)
13499: break;
13500:
13501: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
13502: ssval = gsl_multimin_fminimizer_size (sfm);
13503:
13504: if (rval == GSL_SUCCESS)
13505: printf ("converged to a local maximum at\n");
13506:
13507: printf("%5d ", iteri);
13508: for (it = 0; it < NDIM; it++){
13509: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
13510: }
13511: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
13512: }
13513:
13514: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
13515:
13516: gsl_vector_free(x); /* initial values */
13517: gsl_vector_free(ss); /* inital step size */
13518: for (it=0; it<NDIM; it++){
13519: p[it+1]=gsl_vector_get(sfm->x,it);
13520: fprintf(ficrespow," %.12lf", p[it]);
13521: }
13522: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
13523: #endif
13524: #ifdef POWELL
13525: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
13526: #endif
1.126 brouard 13527: fclose(ficrespow);
13528:
1.203 brouard 13529: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 13530:
13531: for(i=1; i <=NDIM; i++)
13532: for(j=i+1;j<=NDIM;j++)
1.220 brouard 13533: matcov[i][j]=matcov[j][i];
1.126 brouard 13534:
13535: printf("\nCovariance matrix\n ");
1.203 brouard 13536: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 13537: for(i=1; i <=NDIM; i++) {
13538: for(j=1;j<=NDIM;j++){
1.220 brouard 13539: printf("%f ",matcov[i][j]);
13540: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 13541: }
1.203 brouard 13542: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 13543: }
13544:
13545: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 13546: for (i=1;i<=NDIM;i++) {
1.126 brouard 13547: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 13548: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
13549: }
1.302 brouard 13550: lsurv=vector(agegomp,AGESUP);
13551: lpop=vector(agegomp,AGESUP);
13552: tpop=vector(agegomp,AGESUP);
1.126 brouard 13553: lsurv[agegomp]=100000;
13554:
13555: for (k=agegomp;k<=AGESUP;k++) {
13556: agemortsup=k;
13557: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
13558: }
13559:
13560: for (k=agegomp;k<agemortsup;k++)
13561: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
13562:
13563: for (k=agegomp;k<agemortsup;k++){
13564: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
13565: sumlpop=sumlpop+lpop[k];
13566: }
13567:
13568: tpop[agegomp]=sumlpop;
13569: for (k=agegomp;k<(agemortsup-3);k++){
13570: /* tpop[k+1]=2;*/
13571: tpop[k+1]=tpop[k]-lpop[k];
13572: }
13573:
13574:
13575: printf("\nAge lx qx dx Lx Tx e(x)\n");
13576: for (k=agegomp;k<(agemortsup-2);k++)
13577: 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]);
13578:
13579:
13580: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 13581: ageminpar=50;
13582: agemaxpar=100;
1.194 brouard 13583: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
13584: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13585: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13586: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
13587: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13588: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13589: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13590: }else{
13591: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
13592: 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 13593: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 13594: }
1.201 brouard 13595: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 13596: stepm, weightopt,\
13597: model,imx,p,matcov,agemortsup);
13598:
1.302 brouard 13599: free_vector(lsurv,agegomp,AGESUP);
13600: free_vector(lpop,agegomp,AGESUP);
13601: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 13602: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 13603: free_ivector(dcwave,firstobs,lastobs);
13604: free_vector(agecens,firstobs,lastobs);
13605: free_vector(ageexmed,firstobs,lastobs);
13606: free_ivector(cens,firstobs,lastobs);
1.220 brouard 13607: #ifdef GSL
1.136 brouard 13608: #endif
1.186 brouard 13609: } /* Endof if mle==-3 mortality only */
1.205 brouard 13610: /* Standard */
13611: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
13612: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13613: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 13614: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 13615: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
13616: for (k=1; k<=npar;k++)
13617: printf(" %d %8.5f",k,p[k]);
13618: printf("\n");
1.205 brouard 13619: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
13620: /* mlikeli uses func not funcone */
1.247 brouard 13621: /* for(i=1;i<nlstate;i++){ */
13622: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
13623: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
13624: /* } */
1.205 brouard 13625: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
13626: }
13627: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
13628: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13629: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
13630: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13631: }
13632: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 13633: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13634: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
1.335 brouard 13635: /* exit(0); */
1.126 brouard 13636: for (k=1; k<=npar;k++)
13637: printf(" %d %8.5f",k,p[k]);
13638: printf("\n");
13639:
13640: /*--------- results files --------------*/
1.283 brouard 13641: /* 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 13642:
13643:
13644: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13645: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 13646: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13647:
13648: printf("#model= 1 + age ");
13649: fprintf(ficres,"#model= 1 + age ");
13650: fprintf(ficlog,"#model= 1 + age ");
13651: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
13652: </ul>", model);
13653:
13654: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
13655: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
13656: if(nagesqr==1){
13657: printf(" + age*age ");
13658: fprintf(ficres," + age*age ");
13659: fprintf(ficlog," + age*age ");
13660: fprintf(fichtm, "<th>+ age*age</th>");
13661: }
13662: for(j=1;j <=ncovmodel-2;j++){
13663: if(Typevar[j]==0) {
13664: printf(" + V%d ",Tvar[j]);
13665: fprintf(ficres," + V%d ",Tvar[j]);
13666: fprintf(ficlog," + V%d ",Tvar[j]);
13667: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13668: }else if(Typevar[j]==1) {
13669: printf(" + V%d*age ",Tvar[j]);
13670: fprintf(ficres," + V%d*age ",Tvar[j]);
13671: fprintf(ficlog," + V%d*age ",Tvar[j]);
13672: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13673: }else if(Typevar[j]==2) {
13674: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13675: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13676: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13677: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13678: }
13679: }
13680: printf("\n");
13681: fprintf(ficres,"\n");
13682: fprintf(ficlog,"\n");
13683: fprintf(fichtm, "</tr>");
13684: fprintf(fichtm, "\n");
13685:
13686:
1.126 brouard 13687: for(i=1,jk=1; i <=nlstate; i++){
13688: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 13689: if (k != i) {
1.319 brouard 13690: fprintf(fichtm, "<tr>");
1.225 brouard 13691: printf("%d%d ",i,k);
13692: fprintf(ficlog,"%d%d ",i,k);
13693: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 13694: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13695: for(j=1; j <=ncovmodel; j++){
13696: printf("%12.7f ",p[jk]);
13697: fprintf(ficlog,"%12.7f ",p[jk]);
13698: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 13699: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 13700: jk++;
13701: }
13702: printf("\n");
13703: fprintf(ficlog,"\n");
13704: fprintf(ficres,"\n");
1.319 brouard 13705: fprintf(fichtm, "</tr>\n");
1.225 brouard 13706: }
1.126 brouard 13707: }
13708: }
1.319 brouard 13709: /* fprintf(fichtm,"</tr>\n"); */
13710: fprintf(fichtm,"</table>\n");
13711: fprintf(fichtm, "\n");
13712:
1.203 brouard 13713: if(mle != 0){
13714: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 13715: ftolhess=ftol; /* Usually correct */
1.203 brouard 13716: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
13717: 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");
13718: 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 13719: 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 13720: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
13721: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
13722: if(nagesqr==1){
13723: printf(" + age*age ");
13724: fprintf(ficres," + age*age ");
13725: fprintf(ficlog," + age*age ");
13726: fprintf(fichtm, "<th>+ age*age</th>");
13727: }
13728: for(j=1;j <=ncovmodel-2;j++){
13729: if(Typevar[j]==0) {
13730: printf(" + V%d ",Tvar[j]);
13731: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13732: }else if(Typevar[j]==1) {
13733: printf(" + V%d*age ",Tvar[j]);
13734: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13735: }else if(Typevar[j]==2) {
13736: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13737: }
13738: }
13739: fprintf(fichtm, "</tr>\n");
13740:
1.203 brouard 13741: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 13742: for(k=1; k <=(nlstate+ndeath); k++){
13743: if (k != i) {
1.319 brouard 13744: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 13745: printf("%d%d ",i,k);
13746: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 13747: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13748: for(j=1; j <=ncovmodel; j++){
1.319 brouard 13749: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 13750: 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]));
13751: 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 13752: if(fabs(wald) > 1.96){
1.321 brouard 13753: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 13754: }else{
13755: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
13756: }
1.324 brouard 13757: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 13758: 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 13759: jk++;
13760: }
13761: printf("\n");
13762: fprintf(ficlog,"\n");
1.319 brouard 13763: fprintf(fichtm, "</tr>\n");
1.225 brouard 13764: }
13765: }
1.193 brouard 13766: }
1.203 brouard 13767: } /* end of hesscov and Wald tests */
1.319 brouard 13768: fprintf(fichtm,"</table>\n");
1.225 brouard 13769:
1.203 brouard 13770: /* */
1.126 brouard 13771: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
13772: printf("# Scales (for hessian or gradient estimation)\n");
13773: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
13774: for(i=1,jk=1; i <=nlstate; i++){
13775: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 13776: if (j!=i) {
13777: fprintf(ficres,"%1d%1d",i,j);
13778: printf("%1d%1d",i,j);
13779: fprintf(ficlog,"%1d%1d",i,j);
13780: for(k=1; k<=ncovmodel;k++){
13781: printf(" %.5e",delti[jk]);
13782: fprintf(ficlog," %.5e",delti[jk]);
13783: fprintf(ficres," %.5e",delti[jk]);
13784: jk++;
13785: }
13786: printf("\n");
13787: fprintf(ficlog,"\n");
13788: fprintf(ficres,"\n");
13789: }
1.126 brouard 13790: }
13791: }
13792:
13793: 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");
1.203 brouard 13794: if(mle >= 1) /* To big for the screen */
1.126 brouard 13795: 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");
13796: 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");
13797: /* # 121 Var(a12)\n\ */
13798: /* # 122 Cov(b12,a12) Var(b12)\n\ */
13799: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
13800: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
13801: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
13802: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
13803: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
13804: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
13805:
13806:
13807: /* Just to have a covariance matrix which will be more understandable
13808: even is we still don't want to manage dictionary of variables
13809: */
13810: for(itimes=1;itimes<=2;itimes++){
13811: jj=0;
13812: for(i=1; i <=nlstate; i++){
1.225 brouard 13813: for(j=1; j <=nlstate+ndeath; j++){
13814: if(j==i) continue;
13815: for(k=1; k<=ncovmodel;k++){
13816: jj++;
13817: ca[0]= k+'a'-1;ca[1]='\0';
13818: if(itimes==1){
13819: if(mle>=1)
13820: printf("#%1d%1d%d",i,j,k);
13821: fprintf(ficlog,"#%1d%1d%d",i,j,k);
13822: fprintf(ficres,"#%1d%1d%d",i,j,k);
13823: }else{
13824: if(mle>=1)
13825: printf("%1d%1d%d",i,j,k);
13826: fprintf(ficlog,"%1d%1d%d",i,j,k);
13827: fprintf(ficres,"%1d%1d%d",i,j,k);
13828: }
13829: ll=0;
13830: for(li=1;li <=nlstate; li++){
13831: for(lj=1;lj <=nlstate+ndeath; lj++){
13832: if(lj==li) continue;
13833: for(lk=1;lk<=ncovmodel;lk++){
13834: ll++;
13835: if(ll<=jj){
13836: cb[0]= lk +'a'-1;cb[1]='\0';
13837: if(ll<jj){
13838: if(itimes==1){
13839: if(mle>=1)
13840: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13841: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13842: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13843: }else{
13844: if(mle>=1)
13845: printf(" %.5e",matcov[jj][ll]);
13846: fprintf(ficlog," %.5e",matcov[jj][ll]);
13847: fprintf(ficres," %.5e",matcov[jj][ll]);
13848: }
13849: }else{
13850: if(itimes==1){
13851: if(mle>=1)
13852: printf(" Var(%s%1d%1d)",ca,i,j);
13853: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
13854: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
13855: }else{
13856: if(mle>=1)
13857: printf(" %.7e",matcov[jj][ll]);
13858: fprintf(ficlog," %.7e",matcov[jj][ll]);
13859: fprintf(ficres," %.7e",matcov[jj][ll]);
13860: }
13861: }
13862: }
13863: } /* end lk */
13864: } /* end lj */
13865: } /* end li */
13866: if(mle>=1)
13867: printf("\n");
13868: fprintf(ficlog,"\n");
13869: fprintf(ficres,"\n");
13870: numlinepar++;
13871: } /* end k*/
13872: } /*end j */
1.126 brouard 13873: } /* end i */
13874: } /* end itimes */
13875:
13876: fflush(ficlog);
13877: fflush(ficres);
1.225 brouard 13878: while(fgets(line, MAXLINE, ficpar)) {
13879: /* If line starts with a # it is a comment */
13880: if (line[0] == '#') {
13881: numlinepar++;
13882: fputs(line,stdout);
13883: fputs(line,ficparo);
13884: fputs(line,ficlog);
1.299 brouard 13885: fputs(line,ficres);
1.225 brouard 13886: continue;
13887: }else
13888: break;
13889: }
13890:
1.209 brouard 13891: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
13892: /* ungetc(c,ficpar); */
13893: /* fgets(line, MAXLINE, ficpar); */
13894: /* fputs(line,stdout); */
13895: /* fputs(line,ficparo); */
13896: /* } */
13897: /* ungetc(c,ficpar); */
1.126 brouard 13898:
13899: estepm=0;
1.209 brouard 13900: 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 13901:
13902: if (num_filled != 6) {
13903: 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);
13904: 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);
13905: goto end;
13906: }
13907: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
13908: }
13909: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
13910: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
13911:
1.209 brouard 13912: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 13913: if (estepm==0 || estepm < stepm) estepm=stepm;
13914: if (fage <= 2) {
13915: bage = ageminpar;
13916: fage = agemaxpar;
13917: }
13918:
13919: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 13920: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
13921: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 13922:
1.186 brouard 13923: /* Other stuffs, more or less useful */
1.254 brouard 13924: while(fgets(line, MAXLINE, ficpar)) {
13925: /* If line starts with a # it is a comment */
13926: if (line[0] == '#') {
13927: numlinepar++;
13928: fputs(line,stdout);
13929: fputs(line,ficparo);
13930: fputs(line,ficlog);
1.299 brouard 13931: fputs(line,ficres);
1.254 brouard 13932: continue;
13933: }else
13934: break;
13935: }
13936:
13937: 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){
13938:
13939: if (num_filled != 7) {
13940: 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);
13941: 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);
13942: goto end;
13943: }
13944: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13945: 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);
13946: 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);
13947: 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 13948: }
1.254 brouard 13949:
13950: while(fgets(line, MAXLINE, ficpar)) {
13951: /* If line starts with a # it is a comment */
13952: if (line[0] == '#') {
13953: numlinepar++;
13954: fputs(line,stdout);
13955: fputs(line,ficparo);
13956: fputs(line,ficlog);
1.299 brouard 13957: fputs(line,ficres);
1.254 brouard 13958: continue;
13959: }else
13960: break;
1.126 brouard 13961: }
13962:
13963:
13964: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
13965: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
13966:
1.254 brouard 13967: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
13968: if (num_filled != 1) {
13969: 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);
13970: 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);
13971: goto end;
13972: }
13973: printf("pop_based=%d\n",popbased);
13974: fprintf(ficlog,"pop_based=%d\n",popbased);
13975: fprintf(ficparo,"pop_based=%d\n",popbased);
13976: fprintf(ficres,"pop_based=%d\n",popbased);
13977: }
13978:
1.258 brouard 13979: /* Results */
1.332 brouard 13980: /* Value of covariate in each resultine will be compututed (if product) and sorted according to model rank */
13981: /* It is precov[] because we need the varying age in order to compute the real cov[] of the model equation */
13982: precov=matrix(1,MAXRESULTLINESPONE,1,NCOVMAX+1);
1.307 brouard 13983: endishere=0;
1.258 brouard 13984: nresult=0;
1.308 brouard 13985: parameterline=0;
1.258 brouard 13986: do{
13987: if(!fgets(line, MAXLINE, ficpar)){
13988: endishere=1;
1.308 brouard 13989: parameterline=15;
1.258 brouard 13990: }else if (line[0] == '#') {
13991: /* If line starts with a # it is a comment */
1.254 brouard 13992: numlinepar++;
13993: fputs(line,stdout);
13994: fputs(line,ficparo);
13995: fputs(line,ficlog);
1.299 brouard 13996: fputs(line,ficres);
1.254 brouard 13997: continue;
1.258 brouard 13998: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
13999: parameterline=11;
1.296 brouard 14000: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 14001: parameterline=12;
1.307 brouard 14002: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 14003: parameterline=13;
1.307 brouard 14004: }
1.258 brouard 14005: else{
14006: parameterline=14;
1.254 brouard 14007: }
1.308 brouard 14008: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 14009: case 11:
1.296 brouard 14010: 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)){
14011: 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 14012: 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);
14013: 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);
14014: 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);
14015: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 14016: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
14017: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 14018: prvforecast = 1;
14019: }
14020: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 14021: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
14022: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
14023: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 14024: prvforecast = 2;
14025: }
14026: else {
14027: 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);
14028: 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);
14029: goto end;
1.258 brouard 14030: }
1.254 brouard 14031: break;
1.258 brouard 14032: case 12:
1.296 brouard 14033: 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)){
14034: 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);
14035: 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);
14036: 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);
14037: 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);
14038: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 14039: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
14040: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 14041: prvbackcast = 1;
14042: }
14043: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 14044: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
14045: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
14046: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 14047: prvbackcast = 2;
14048: }
14049: else {
14050: 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);
14051: 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);
14052: goto end;
1.258 brouard 14053: }
1.230 brouard 14054: break;
1.258 brouard 14055: case 13:
1.332 brouard 14056: num_filled=sscanf(line,"result:%[^\n]\n",resultlineori);
1.307 brouard 14057: nresult++; /* Sum of resultlines */
1.342 brouard 14058: /* printf("Result %d: result:%s\n",nresult, resultlineori); */
1.332 brouard 14059: /* removefirstspace(&resultlineori); */
14060:
14061: if(strstr(resultlineori,"v") !=0){
14062: printf("Error. 'v' must be in upper case 'V' result: %s ",resultlineori);
14063: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultlineori);fflush(ficlog);
14064: return 1;
14065: }
14066: trimbb(resultline, resultlineori); /* Suppressing double blank in the resultline */
1.342 brouard 14067: /* printf("Decoderesult resultline=\"%s\" resultlineori=\"%s\"\n", resultline, resultlineori); */
1.318 brouard 14068: if(nresult > MAXRESULTLINESPONE-1){
14069: 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);
14070: 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 14071: goto end;
14072: }
1.332 brouard 14073:
1.310 brouard 14074: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 14075: fprintf(ficparo,"result: %s\n",resultline);
14076: fprintf(ficres,"result: %s\n",resultline);
14077: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 14078: } else
14079: goto end;
1.307 brouard 14080: break;
14081: case 14:
14082: printf("Error: Unknown command '%s'\n",line);
14083: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 14084: if(line[0] == ' ' || line[0] == '\n'){
14085: printf("It should not be an empty line '%s'\n",line);
14086: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
14087: }
1.307 brouard 14088: if(ncovmodel >=2 && nresult==0 ){
14089: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
14090: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 14091: }
1.307 brouard 14092: /* goto end; */
14093: break;
1.308 brouard 14094: case 15:
14095: printf("End of resultlines.\n");
14096: fprintf(ficlog,"End of resultlines.\n");
14097: break;
14098: default: /* parameterline =0 */
1.307 brouard 14099: nresult=1;
14100: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 14101: } /* End switch parameterline */
14102: }while(endishere==0); /* End do */
1.126 brouard 14103:
1.230 brouard 14104: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 14105: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 14106:
14107: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 14108: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 14109: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 14110: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
14111: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 14112: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 14113: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
14114: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 14115: }else{
1.270 brouard 14116: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 14117: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
14118: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
14119: if(prvforecast==1){
14120: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
14121: jprojd=jproj1;
14122: mprojd=mproj1;
14123: anprojd=anproj1;
14124: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
14125: jprojf=jproj2;
14126: mprojf=mproj2;
14127: anprojf=anproj2;
14128: } else if(prvforecast == 2){
14129: dateprojd=dateintmean;
14130: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
14131: dateprojf=dateintmean+yrfproj;
14132: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
14133: }
14134: if(prvbackcast==1){
14135: datebackd=(jback1+12*mback1+365*anback1)/365;
14136: jbackd=jback1;
14137: mbackd=mback1;
14138: anbackd=anback1;
14139: datebackf=(jback2+12*mback2+365*anback2)/365;
14140: jbackf=jback2;
14141: mbackf=mback2;
14142: anbackf=anback2;
14143: } else if(prvbackcast == 2){
14144: datebackd=dateintmean;
14145: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
14146: datebackf=dateintmean-yrbproj;
14147: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
14148: }
14149:
14150: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 14151: }
14152: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 14153: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
14154: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 14155:
1.225 brouard 14156: /*------------ free_vector -------------*/
14157: /* chdir(path); */
1.220 brouard 14158:
1.215 brouard 14159: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
14160: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
14161: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
14162: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 14163: free_lvector(num,firstobs,lastobs);
14164: free_vector(agedc,firstobs,lastobs);
1.126 brouard 14165: /*free_matrix(covar,0,NCOVMAX,1,n);*/
14166: /*free_matrix(covar,1,NCOVMAX,1,n);*/
14167: fclose(ficparo);
14168: fclose(ficres);
1.220 brouard 14169:
14170:
1.186 brouard 14171: /* Other results (useful)*/
1.220 brouard 14172:
14173:
1.126 brouard 14174: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 14175: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
14176: prlim=matrix(1,nlstate,1,nlstate);
1.332 brouard 14177: /* Computes the prevalence limit for each combination k of the dummy covariates by calling prevalim(k) */
1.209 brouard 14178: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 14179: fclose(ficrespl);
14180:
14181: /*------------- h Pij x at various ages ------------*/
1.180 brouard 14182: /*#include "hpijx.h"*/
1.332 brouard 14183: /** 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?*/
14184: /* calls hpxij with combination k */
1.180 brouard 14185: hPijx(p, bage, fage);
1.145 brouard 14186: fclose(ficrespij);
1.227 brouard 14187:
1.220 brouard 14188: /* ncovcombmax= pow(2,cptcoveff); */
1.332 brouard 14189: /*-------------- Variance of one-step probabilities for a combination ij or for nres ?---*/
1.145 brouard 14190: k=1;
1.126 brouard 14191: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 14192:
1.269 brouard 14193: /* Prevalence for each covariate combination in probs[age][status][cov] */
14194: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
14195: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 14196: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 14197: for(k=1;k<=ncovcombmax;k++)
14198: probs[i][j][k]=0.;
1.269 brouard 14199: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
14200: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 14201: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 14202: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
14203: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 14204: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 14205: for(k=1;k<=ncovcombmax;k++)
14206: mobaverages[i][j][k]=0.;
1.219 brouard 14207: mobaverage=mobaverages;
14208: if (mobilav!=0) {
1.235 brouard 14209: printf("Movingaveraging observed prevalence\n");
1.258 brouard 14210: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 14211: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
14212: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
14213: printf(" Error in movingaverage mobilav=%d\n",mobilav);
14214: }
1.269 brouard 14215: } else if (mobilavproj !=0) {
1.235 brouard 14216: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 14217: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 14218: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
14219: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
14220: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
14221: }
1.269 brouard 14222: }else{
14223: printf("Internal error moving average\n");
14224: fflush(stdout);
14225: exit(1);
1.219 brouard 14226: }
14227: }/* end if moving average */
1.227 brouard 14228:
1.126 brouard 14229: /*---------- Forecasting ------------------*/
1.296 brouard 14230: if(prevfcast==1){
14231: /* /\* if(stepm ==1){*\/ */
14232: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
14233: /*This done previously after freqsummary.*/
14234: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
14235: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
14236:
14237: /* } else if (prvforecast==2){ */
14238: /* /\* if(stepm ==1){*\/ */
14239: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
14240: /* } */
14241: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
14242: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 14243: }
1.269 brouard 14244:
1.296 brouard 14245: /* Prevbcasting */
14246: if(prevbcast==1){
1.219 brouard 14247: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14248: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14249: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14250:
14251: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
14252:
14253: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 14254:
1.219 brouard 14255: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
14256: fclose(ficresplb);
14257:
1.222 brouard 14258: hBijx(p, bage, fage, mobaverage);
14259: fclose(ficrespijb);
1.219 brouard 14260:
1.296 brouard 14261: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
14262: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
14263: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
14264: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
14265: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
14266: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
14267:
14268:
1.269 brouard 14269: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14270:
14271:
1.269 brouard 14272: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 14273: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
14274: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
14275: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 14276: } /* end Prevbcasting */
1.268 brouard 14277:
1.186 brouard 14278:
14279: /* ------ Other prevalence ratios------------ */
1.126 brouard 14280:
1.215 brouard 14281: free_ivector(wav,1,imx);
14282: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
14283: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
14284: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 14285:
14286:
1.127 brouard 14287: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 14288:
1.201 brouard 14289: strcpy(filerese,"E_");
14290: strcat(filerese,fileresu);
1.126 brouard 14291: if((ficreseij=fopen(filerese,"w"))==NULL) {
14292: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
14293: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
14294: }
1.208 brouard 14295: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
14296: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 14297:
14298: pstamp(ficreseij);
1.219 brouard 14299:
1.235 brouard 14300: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
14301: if (cptcovn < 1){i1=1;}
14302:
14303: for(nres=1; nres <= nresult; nres++) /* For each resultline */
14304: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 14305: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 14306: continue;
1.219 brouard 14307: fprintf(ficreseij,"\n#****** ");
1.235 brouard 14308: printf("\n#****** ");
1.225 brouard 14309: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 14310: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
14311: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 14312: }
14313: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.337 brouard 14314: printf(" V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]); /* TvarsQ[j] gives the name of the jth quantitative (fixed or time v) */
14315: fprintf(ficreseij,"V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]);
1.219 brouard 14316: }
14317: fprintf(ficreseij,"******\n");
1.235 brouard 14318: printf("******\n");
1.219 brouard 14319:
14320: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14321: oldm=oldms;savm=savms;
1.330 brouard 14322: /* 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 14323: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 14324:
1.219 brouard 14325: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 14326: }
14327: fclose(ficreseij);
1.208 brouard 14328: printf("done evsij\n");fflush(stdout);
14329: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 14330:
1.218 brouard 14331:
1.227 brouard 14332: /*---------- State-specific expectancies and variances ------------*/
1.336 brouard 14333: /* Should be moved in a function */
1.201 brouard 14334: strcpy(filerest,"T_");
14335: strcat(filerest,fileresu);
1.127 brouard 14336: if((ficrest=fopen(filerest,"w"))==NULL) {
14337: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
14338: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
14339: }
1.208 brouard 14340: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
14341: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 14342: strcpy(fileresstde,"STDE_");
14343: strcat(fileresstde,fileresu);
1.126 brouard 14344: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 14345: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
14346: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 14347: }
1.227 brouard 14348: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
14349: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 14350:
1.201 brouard 14351: strcpy(filerescve,"CVE_");
14352: strcat(filerescve,fileresu);
1.126 brouard 14353: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 14354: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
14355: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 14356: }
1.227 brouard 14357: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
14358: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 14359:
1.201 brouard 14360: strcpy(fileresv,"V_");
14361: strcat(fileresv,fileresu);
1.126 brouard 14362: if((ficresvij=fopen(fileresv,"w"))==NULL) {
14363: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
14364: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
14365: }
1.227 brouard 14366: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
14367: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 14368:
1.235 brouard 14369: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
14370: if (cptcovn < 1){i1=1;}
14371:
1.334 brouard 14372: for(nres=1; nres <= nresult; nres++) /* For each resultline, find the combination and output results according to the values of dummies and then quanti. */
14373: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying. For each nres and each value at position k
14374: * we know Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline
14375: * Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline
14376: * and Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
14377: /* */
14378: 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 14379: continue;
1.321 brouard 14380: printf("\n# model %s \n#****** Result for:", model);
14381: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
14382: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.334 brouard 14383: /* It might not be a good idea to mix dummies and quantitative */
14384: /* 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 *\/ */
14385: 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 */
14386: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* Output by variables in the resultline *\/ */
14387: /* Tvaraff[j] is the name of the dummy variable in position j in the equation model:
14388: * Tvaraff[1]@9={4, 3, 0, 0, 0, 0, 0, 0, 0}, in model=V5+V4+V3+V4*V3+V5*age
14389: * (V5 is quanti) V4 and V3 are dummies
14390: * TnsdVar[4] is the position 1 and TnsdVar[3]=2 in codtabm(k,l)(V4 V3)=V4 V3
14391: * l=1 l=2
14392: * k=1 1 1 0 0
14393: * k=2 2 1 1 0
14394: * k=3 [1] [2] 0 1
14395: * k=4 2 2 1 1
14396: * If nres=1 result: V3=1 V4=0 then k=3 and outputs
14397: * If nres=2 result: V4=1 V3=0 then k=2 and outputs
14398: * nres=1 =>k=3 j=1 V4= nbcode[4][codtabm(3,1)=1)=0; j=2 V3= nbcode[3][codtabm(3,2)=2]=1
14399: * nres=2 =>k=2 j=1 V4= nbcode[4][codtabm(2,1)=2)=1; j=2 V3= nbcode[3][codtabm(2,2)=1]=0
14400: */
14401: /* Tvresult[nres][j] Name of the variable at position j in this resultline */
14402: /* Tresult[nres][j] Value of this variable at position j could be a float if quantitative */
14403: /* We give up with the combinations!! */
1.342 brouard 14404: /* if(debugILK) */
14405: /* 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 14406:
14407: 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 14408: /* 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] */
14409: 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 */
14410: 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 */
14411: 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 14412: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
14413: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
14414: }else{
14415: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
14416: }
14417: /* fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14418: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14419: }else if(Dummy[modelresult[nres][j]]==1){ /* Quanti variable */
14420: /* For each selected (single) quantitative value */
1.337 brouard 14421: printf(" V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
14422: fprintf(ficlog," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
14423: fprintf(ficrest," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
1.334 brouard 14424: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
14425: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
14426: }else{
14427: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
14428: }
14429: }else{
14430: 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 */
14431: 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 */
14432: exit(1);
14433: }
1.335 brouard 14434: } /* End loop for each variable in the resultline */
1.334 brouard 14435: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
14436: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /\* Wrong j is not in the equation model *\/ */
14437: /* fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14438: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14439: /* } */
1.208 brouard 14440: fprintf(ficrest,"******\n");
1.227 brouard 14441: fprintf(ficlog,"******\n");
14442: printf("******\n");
1.208 brouard 14443:
14444: fprintf(ficresstdeij,"\n#****** ");
14445: fprintf(ficrescveij,"\n#****** ");
1.337 brouard 14446: /* It could have been: for(j=1;j<=cptcoveff;j++) {printf("V=%d=%lg",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);} */
14447: /* But it won't be sorted and depends on how the resultline is ordered */
1.225 brouard 14448: for(j=1;j<=cptcoveff;j++) {
1.334 brouard 14449: fprintf(ficresstdeij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14450: /* fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14451: /* fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14452: }
14453: 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 14454: fprintf(ficresstdeij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
14455: fprintf(ficrescveij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
1.235 brouard 14456: }
1.208 brouard 14457: fprintf(ficresstdeij,"******\n");
14458: fprintf(ficrescveij,"******\n");
14459:
14460: fprintf(ficresvij,"\n#****** ");
1.238 brouard 14461: /* pstamp(ficresvij); */
1.225 brouard 14462: for(j=1;j<=cptcoveff;j++)
1.335 brouard 14463: fprintf(ficresvij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14464: /* fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[TnsdVar[Tvaraff[j]]])]); */
1.235 brouard 14465: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 14466: /* fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]); /\* To solve *\/ */
1.337 brouard 14467: fprintf(ficresvij," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /* Solved */
1.235 brouard 14468: }
1.208 brouard 14469: fprintf(ficresvij,"******\n");
14470:
14471: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14472: oldm=oldms;savm=savms;
1.235 brouard 14473: printf(" cvevsij ");
14474: fprintf(ficlog, " cvevsij ");
14475: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 14476: printf(" end cvevsij \n ");
14477: fprintf(ficlog, " end cvevsij \n ");
14478:
14479: /*
14480: */
14481: /* goto endfree; */
14482:
14483: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14484: pstamp(ficrest);
14485:
1.269 brouard 14486: epj=vector(1,nlstate+1);
1.208 brouard 14487: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 14488: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
14489: cptcod= 0; /* To be deleted */
14490: printf("varevsij vpopbased=%d \n",vpopbased);
14491: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 14492: 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 */
1.227 brouard 14493: fprintf(ficrest,"# Total life expectancy with std error and decomposition into time to be expected in each health state\n# (weighted average of eij where weights are ");
14494: if(vpopbased==1)
14495: 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);
14496: else
1.288 brouard 14497: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.335 brouard 14498: fprintf(ficrest,"# Age popbased mobilav e.. (std) "); /* Adding covariate values? */
1.227 brouard 14499: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
14500: fprintf(ficrest,"\n");
14501: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 14502: printf("Computing age specific forward period (stable) prevalences in each health state \n");
14503: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 14504: for(age=bage; age <=fage ;age++){
1.235 brouard 14505: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 14506: if (vpopbased==1) {
14507: if(mobilav ==0){
14508: for(i=1; i<=nlstate;i++)
14509: prlim[i][i]=probs[(int)age][i][k];
14510: }else{ /* mobilav */
14511: for(i=1; i<=nlstate;i++)
14512: prlim[i][i]=mobaverage[(int)age][i][k];
14513: }
14514: }
1.219 brouard 14515:
1.227 brouard 14516: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
14517: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
14518: /* printf(" age %4.0f ",age); */
14519: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
14520: for(i=1, epj[j]=0.;i <=nlstate;i++) {
14521: epj[j] += prlim[i][i]*eij[i][j][(int)age];
14522: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
14523: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
14524: }
14525: epj[nlstate+1] +=epj[j];
14526: }
14527: /* printf(" age %4.0f \n",age); */
1.219 brouard 14528:
1.227 brouard 14529: for(i=1, vepp=0.;i <=nlstate;i++)
14530: for(j=1;j <=nlstate;j++)
14531: vepp += vareij[i][j][(int)age];
14532: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
14533: for(j=1;j <=nlstate;j++){
14534: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
14535: }
14536: fprintf(ficrest,"\n");
14537: }
1.208 brouard 14538: } /* End vpopbased */
1.269 brouard 14539: free_vector(epj,1,nlstate+1);
1.208 brouard 14540: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
14541: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 14542: printf("done selection\n");fflush(stdout);
14543: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 14544:
1.335 brouard 14545: } /* End k selection or end covariate selection for nres */
1.227 brouard 14546:
14547: printf("done State-specific expectancies\n");fflush(stdout);
14548: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
14549:
1.335 brouard 14550: /* variance-covariance of forward period prevalence */
1.269 brouard 14551: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14552:
1.227 brouard 14553:
1.290 brouard 14554: free_vector(weight,firstobs,lastobs);
1.330 brouard 14555: free_imatrix(Tvardk,1,NCOVMAX,1,2);
1.227 brouard 14556: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 14557: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
14558: free_matrix(anint,1,maxwav,firstobs,lastobs);
14559: free_matrix(mint,1,maxwav,firstobs,lastobs);
14560: free_ivector(cod,firstobs,lastobs);
1.227 brouard 14561: free_ivector(tab,1,NCOVMAX);
14562: fclose(ficresstdeij);
14563: fclose(ficrescveij);
14564: fclose(ficresvij);
14565: fclose(ficrest);
14566: fclose(ficpar);
14567:
14568:
1.126 brouard 14569: /*---------- End : free ----------------*/
1.219 brouard 14570: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 14571: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
14572: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 14573: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
14574: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 14575: } /* mle==-3 arrives here for freeing */
1.227 brouard 14576: /* endfree:*/
14577: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
14578: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
14579: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.341 brouard 14580: /* if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs); */
14581: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs);
1.290 brouard 14582: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
14583: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
14584: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 14585: free_matrix(matcov,1,npar,1,npar);
14586: free_matrix(hess,1,npar,1,npar);
14587: /*free_vector(delti,1,npar);*/
14588: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14589: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 14590: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 14591: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14592:
14593: free_ivector(ncodemax,1,NCOVMAX);
14594: free_ivector(ncodemaxwundef,1,NCOVMAX);
14595: free_ivector(Dummy,-1,NCOVMAX);
14596: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 14597: free_ivector(DummyV,1,NCOVMAX);
14598: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 14599: free_ivector(Typevar,-1,NCOVMAX);
14600: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 14601: free_ivector(TvarsQ,1,NCOVMAX);
14602: free_ivector(TvarsQind,1,NCOVMAX);
14603: free_ivector(TvarsD,1,NCOVMAX);
1.330 brouard 14604: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 14605: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 14606: free_ivector(TvarFD,1,NCOVMAX);
14607: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 14608: free_ivector(TvarF,1,NCOVMAX);
14609: free_ivector(TvarFind,1,NCOVMAX);
14610: free_ivector(TvarV,1,NCOVMAX);
14611: free_ivector(TvarVind,1,NCOVMAX);
14612: free_ivector(TvarA,1,NCOVMAX);
14613: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 14614: free_ivector(TvarFQ,1,NCOVMAX);
14615: free_ivector(TvarFQind,1,NCOVMAX);
14616: free_ivector(TvarVD,1,NCOVMAX);
14617: free_ivector(TvarVDind,1,NCOVMAX);
14618: free_ivector(TvarVQ,1,NCOVMAX);
14619: free_ivector(TvarVQind,1,NCOVMAX);
1.339 brouard 14620: free_ivector(TvarVV,1,NCOVMAX);
14621: free_ivector(TvarVVind,1,NCOVMAX);
14622:
1.230 brouard 14623: free_ivector(Tvarsel,1,NCOVMAX);
14624: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 14625: free_ivector(Tposprod,1,NCOVMAX);
14626: free_ivector(Tprod,1,NCOVMAX);
14627: free_ivector(Tvaraff,1,NCOVMAX);
1.338 brouard 14628: free_ivector(invalidvarcomb,0,ncovcombmax);
1.227 brouard 14629: free_ivector(Tage,1,NCOVMAX);
14630: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 14631: free_ivector(TmodelInvind,1,NCOVMAX);
14632: free_ivector(TmodelInvQind,1,NCOVMAX);
1.332 brouard 14633:
14634: free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /* Could be elsewhere ?*/
14635:
1.227 brouard 14636: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
14637: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 14638: fflush(fichtm);
14639: fflush(ficgp);
14640:
1.227 brouard 14641:
1.126 brouard 14642: if((nberr >0) || (nbwarn>0)){
1.216 brouard 14643: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
14644: 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 14645: }else{
14646: printf("End of Imach\n");
14647: fprintf(ficlog,"End of Imach\n");
14648: }
14649: printf("See log file on %s\n",filelog);
14650: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 14651: /*(void) gettimeofday(&end_time,&tzp);*/
14652: rend_time = time(NULL);
14653: end_time = *localtime(&rend_time);
14654: /* tml = *localtime(&end_time.tm_sec); */
14655: strcpy(strtend,asctime(&end_time));
1.126 brouard 14656: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
14657: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 14658: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 14659:
1.157 brouard 14660: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
14661: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
14662: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 14663: /* printf("Total time was %d uSec.\n", total_usecs);*/
14664: /* if(fileappend(fichtm,optionfilehtm)){ */
14665: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14666: fclose(fichtm);
14667: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14668: fclose(fichtmcov);
14669: fclose(ficgp);
14670: fclose(ficlog);
14671: /*------ End -----------*/
1.227 brouard 14672:
1.281 brouard 14673:
14674: /* Executes gnuplot */
1.227 brouard 14675:
14676: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 14677: #ifdef WIN32
1.227 brouard 14678: if (_chdir(pathcd) != 0)
14679: printf("Can't move to directory %s!\n",path);
14680: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 14681: #else
1.227 brouard 14682: if(chdir(pathcd) != 0)
14683: printf("Can't move to directory %s!\n", path);
14684: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 14685: #endif
1.126 brouard 14686: printf("Current directory %s!\n",pathcd);
14687: /*strcat(plotcmd,CHARSEPARATOR);*/
14688: sprintf(plotcmd,"gnuplot");
1.157 brouard 14689: #ifdef _WIN32
1.126 brouard 14690: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
14691: #endif
14692: if(!stat(plotcmd,&info)){
1.158 brouard 14693: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14694: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 14695: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 14696: }else
14697: strcpy(pplotcmd,plotcmd);
1.157 brouard 14698: #ifdef __unix
1.126 brouard 14699: strcpy(plotcmd,GNUPLOTPROGRAM);
14700: if(!stat(plotcmd,&info)){
1.158 brouard 14701: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14702: }else
14703: strcpy(pplotcmd,plotcmd);
14704: #endif
14705: }else
14706: strcpy(pplotcmd,plotcmd);
14707:
14708: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 14709: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 14710: strcpy(pplotcmd,plotcmd);
1.227 brouard 14711:
1.126 brouard 14712: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 14713: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 14714: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 14715: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 14716: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 14717: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 14718: strcpy(plotcmd,pplotcmd);
14719: }
1.126 brouard 14720: }
1.158 brouard 14721: printf(" Successful, please wait...");
1.126 brouard 14722: while (z[0] != 'q') {
14723: /* chdir(path); */
1.154 brouard 14724: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 14725: scanf("%s",z);
14726: /* if (z[0] == 'c') system("./imach"); */
14727: if (z[0] == 'e') {
1.158 brouard 14728: #ifdef __APPLE__
1.152 brouard 14729: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 14730: #elif __linux
14731: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 14732: #else
1.152 brouard 14733: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 14734: #endif
14735: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
14736: system(pplotcmd);
1.126 brouard 14737: }
14738: else if (z[0] == 'g') system(plotcmd);
14739: else if (z[0] == 'q') exit(0);
14740: }
1.227 brouard 14741: end:
1.126 brouard 14742: while (z[0] != 'q') {
1.195 brouard 14743: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 14744: scanf("%s",z);
14745: }
1.283 brouard 14746: printf("End\n");
1.282 brouard 14747: exit(0);
1.126 brouard 14748: }
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