Annotation of imach/src/imach.c, revision 1.343
1.343 ! brouard 1: /* $Id: imach.c,v 1.342 2022/09/11 19:54:09 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.343 ! brouard 4: Revision 1.342 2022/09/11 19:54:09 brouard
! 5: Summary: 0.99r38
! 6:
! 7: * imach.c (Module): Adding timevarying products of any kinds,
! 8: should work before shifting cotvar from ncovcol+nqv columns in
! 9: order to have a correspondance between the column of cotvar and
! 10: the id of column.
! 11: (Module): Some cleaning and adding covariates in ILK.txt
! 12:
1.342 brouard 13: Revision 1.341 2022/09/11 07:58:42 brouard
14: Summary: Version 0.99r38
15:
16: After adding change in cotvar.
17:
1.341 brouard 18: Revision 1.340 2022/09/11 07:53:11 brouard
19: Summary: Version imach 0.99r37
20:
21: * imach.c (Module): Adding timevarying products of any kinds,
22: should work before shifting cotvar from ncovcol+nqv columns in
23: order to have a correspondance between the column of cotvar and
24: the id of column.
25:
1.340 brouard 26: Revision 1.339 2022/09/09 17:55:22 brouard
27: Summary: version 0.99r37
28:
29: * imach.c (Module): Many improvements for fixing products of fixed
30: timevarying as well as fixed * fixed, and test with quantitative
31: covariate.
32:
1.339 brouard 33: Revision 1.338 2022/09/04 17:40:33 brouard
34: Summary: 0.99r36
35:
36: * imach.c (Module): Now the easy runs i.e. without result or
37: model=1+age only did not work. The defautl combination should be 1
38: and not 0 because everything hasn't been tranformed yet.
39:
1.338 brouard 40: Revision 1.337 2022/09/02 14:26:02 brouard
41: Summary: version 0.99r35
42:
43: * src/imach.c: Version 0.99r35 because it outputs same results with
44: 1+age+V1+V1*age for females and 1+age for females only
45: (education=1 noweight)
46:
1.337 brouard 47: Revision 1.336 2022/08/31 09:52:36 brouard
48: *** empty log message ***
49:
1.336 brouard 50: Revision 1.335 2022/08/31 08:23:16 brouard
51: Summary: improvements...
52:
1.335 brouard 53: Revision 1.334 2022/08/25 09:08:41 brouard
54: Summary: In progress for quantitative
55:
1.334 brouard 56: Revision 1.333 2022/08/21 09:10:30 brouard
57: * src/imach.c (Module): Version 0.99r33 A lot of changes in
58: reassigning covariates: my first idea was that people will always
59: use the first covariate V1 into the model but in fact they are
60: producing data with many covariates and can use an equation model
61: with some of the covariate; it means that in a model V2+V3 instead
62: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
63: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
64: the equation model is restricted to two variables only (V2, V3)
65: and the combination for V2 should be codtabm(k,1) instead of
66: (codtabm(k,2), and the code should be
67: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
68: made. All of these should be simplified once a day like we did in
69: hpxij() for example by using precov[nres] which is computed in
70: decoderesult for each nres of each resultline. Loop should be done
71: on the equation model globally by distinguishing only product with
72: age (which are changing with age) and no more on type of
73: covariates, single dummies, single covariates.
74:
1.333 brouard 75: Revision 1.332 2022/08/21 09:06:25 brouard
76: Summary: Version 0.99r33
77:
78: * src/imach.c (Module): Version 0.99r33 A lot of changes in
79: reassigning covariates: my first idea was that people will always
80: use the first covariate V1 into the model but in fact they are
81: producing data with many covariates and can use an equation model
82: with some of the covariate; it means that in a model V2+V3 instead
83: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
84: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
85: the equation model is restricted to two variables only (V2, V3)
86: and the combination for V2 should be codtabm(k,1) instead of
87: (codtabm(k,2), and the code should be
88: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
89: made. All of these should be simplified once a day like we did in
90: hpxij() for example by using precov[nres] which is computed in
91: decoderesult for each nres of each resultline. Loop should be done
92: on the equation model globally by distinguishing only product with
93: age (which are changing with age) and no more on type of
94: covariates, single dummies, single covariates.
95:
1.332 brouard 96: Revision 1.331 2022/08/07 05:40:09 brouard
97: *** empty log message ***
98:
1.331 brouard 99: Revision 1.330 2022/08/06 07:18:25 brouard
100: Summary: last 0.99r31
101:
102: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function
103:
1.330 brouard 104: Revision 1.329 2022/08/03 17:29:54 brouard
105: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
106:
1.329 brouard 107: Revision 1.328 2022/07/27 17:40:48 brouard
108: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
109:
1.328 brouard 110: Revision 1.327 2022/07/27 14:47:35 brouard
111: Summary: Still a problem for one-step probabilities in case of quantitative variables
112:
1.327 brouard 113: Revision 1.326 2022/07/26 17:33:55 brouard
114: Summary: some test with nres=1
115:
1.326 brouard 116: Revision 1.325 2022/07/25 14:27:23 brouard
117: Summary: r30
118:
119: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
120: coredumped, revealed by Feiuno, thank you.
121:
1.325 brouard 122: Revision 1.324 2022/07/23 17:44:26 brouard
123: *** empty log message ***
124:
1.324 brouard 125: Revision 1.323 2022/07/22 12:30:08 brouard
126: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
127:
1.323 brouard 128: Revision 1.322 2022/07/22 12:27:48 brouard
129: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
130:
1.322 brouard 131: Revision 1.321 2022/07/22 12:04:24 brouard
132: Summary: r28
133:
134: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
135:
1.321 brouard 136: Revision 1.320 2022/06/02 05:10:11 brouard
137: *** empty log message ***
138:
1.320 brouard 139: Revision 1.319 2022/06/02 04:45:11 brouard
140: * imach.c (Module): Adding the Wald tests from the log to the main
141: htm for better display of the maximum likelihood estimators.
142:
1.319 brouard 143: Revision 1.318 2022/05/24 08:10:59 brouard
144: * imach.c (Module): Some attempts to find a bug of wrong estimates
145: of confidencce intervals with product in the equation modelC
146:
1.318 brouard 147: Revision 1.317 2022/05/15 15:06:23 brouard
148: * imach.c (Module): Some minor improvements
149:
1.317 brouard 150: Revision 1.316 2022/05/11 15:11:31 brouard
151: Summary: r27
152:
1.316 brouard 153: Revision 1.315 2022/05/11 15:06:32 brouard
154: *** empty log message ***
155:
1.315 brouard 156: Revision 1.314 2022/04/13 17:43:09 brouard
157: * imach.c (Module): Adding link to text data files
158:
1.314 brouard 159: Revision 1.313 2022/04/11 15:57:42 brouard
160: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
161:
1.313 brouard 162: Revision 1.312 2022/04/05 21:24:39 brouard
163: *** empty log message ***
164:
1.312 brouard 165: Revision 1.311 2022/04/05 21:03:51 brouard
166: Summary: Fixed quantitative covariates
167:
168: Fixed covariates (dummy or quantitative)
169: with missing values have never been allowed but are ERRORS and
170: program quits. Standard deviations of fixed covariates were
171: wrongly computed. Mean and standard deviations of time varying
172: covariates are still not computed.
173:
1.311 brouard 174: Revision 1.310 2022/03/17 08:45:53 brouard
175: Summary: 99r25
176:
177: Improving detection of errors: result lines should be compatible with
178: the model.
179:
1.310 brouard 180: Revision 1.309 2021/05/20 12:39:14 brouard
181: Summary: Version 0.99r24
182:
1.309 brouard 183: Revision 1.308 2021/03/31 13:11:57 brouard
184: Summary: Version 0.99r23
185:
186:
187: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
188:
1.308 brouard 189: Revision 1.307 2021/03/08 18:11:32 brouard
190: Summary: 0.99r22 fixed bug on result:
191:
1.307 brouard 192: Revision 1.306 2021/02/20 15:44:02 brouard
193: Summary: Version 0.99r21
194:
195: * imach.c (Module): Fix bug on quitting after result lines!
196: (Module): Version 0.99r21
197:
1.306 brouard 198: Revision 1.305 2021/02/20 15:28:30 brouard
199: * imach.c (Module): Fix bug on quitting after result lines!
200:
1.305 brouard 201: Revision 1.304 2021/02/12 11:34:20 brouard
202: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
203:
1.304 brouard 204: Revision 1.303 2021/02/11 19:50:15 brouard
205: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
206:
1.303 brouard 207: Revision 1.302 2020/02/22 21:00:05 brouard
208: * (Module): imach.c Update mle=-3 (for computing Life expectancy
209: and life table from the data without any state)
210:
1.302 brouard 211: Revision 1.301 2019/06/04 13:51:20 brouard
212: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
213:
1.301 brouard 214: Revision 1.300 2019/05/22 19:09:45 brouard
215: Summary: version 0.99r19 of May 2019
216:
1.300 brouard 217: Revision 1.299 2019/05/22 18:37:08 brouard
218: Summary: Cleaned 0.99r19
219:
1.299 brouard 220: Revision 1.298 2019/05/22 18:19:56 brouard
221: *** empty log message ***
222:
1.298 brouard 223: Revision 1.297 2019/05/22 17:56:10 brouard
224: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
225:
1.297 brouard 226: Revision 1.296 2019/05/20 13:03:18 brouard
227: Summary: Projection syntax simplified
228:
229:
230: We can now start projections, forward or backward, from the mean date
231: of inteviews up to or down to a number of years of projection:
232: prevforecast=1 yearsfproj=15.3 mobil_average=0
233: or
234: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
235: or
236: prevbackcast=1 yearsbproj=12.3 mobil_average=1
237: or
238: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
239:
1.296 brouard 240: Revision 1.295 2019/05/18 09:52:50 brouard
241: Summary: doxygen tex bug
242:
1.295 brouard 243: Revision 1.294 2019/05/16 14:54:33 brouard
244: Summary: There was some wrong lines added
245:
1.294 brouard 246: Revision 1.293 2019/05/09 15:17:34 brouard
247: *** empty log message ***
248:
1.293 brouard 249: Revision 1.292 2019/05/09 14:17:20 brouard
250: Summary: Some updates
251:
1.292 brouard 252: Revision 1.291 2019/05/09 13:44:18 brouard
253: Summary: Before ncovmax
254:
1.291 brouard 255: Revision 1.290 2019/05/09 13:39:37 brouard
256: Summary: 0.99r18 unlimited number of individuals
257:
258: 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.
259:
1.290 brouard 260: Revision 1.289 2018/12/13 09:16:26 brouard
261: Summary: Bug for young ages (<-30) will be in r17
262:
1.289 brouard 263: Revision 1.288 2018/05/02 20:58:27 brouard
264: Summary: Some bugs fixed
265:
1.288 brouard 266: Revision 1.287 2018/05/01 17:57:25 brouard
267: Summary: Bug fixed by providing frequencies only for non missing covariates
268:
1.287 brouard 269: Revision 1.286 2018/04/27 14:27:04 brouard
270: Summary: some minor bugs
271:
1.286 brouard 272: Revision 1.285 2018/04/21 21:02:16 brouard
273: Summary: Some bugs fixed, valgrind tested
274:
1.285 brouard 275: Revision 1.284 2018/04/20 05:22:13 brouard
276: Summary: Computing mean and stdeviation of fixed quantitative variables
277:
1.284 brouard 278: Revision 1.283 2018/04/19 14:49:16 brouard
279: Summary: Some minor bugs fixed
280:
1.283 brouard 281: Revision 1.282 2018/02/27 22:50:02 brouard
282: *** empty log message ***
283:
1.282 brouard 284: Revision 1.281 2018/02/27 19:25:23 brouard
285: Summary: Adding second argument for quitting
286:
1.281 brouard 287: Revision 1.280 2018/02/21 07:58:13 brouard
288: Summary: 0.99r15
289:
290: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
291:
1.280 brouard 292: Revision 1.279 2017/07/20 13:35:01 brouard
293: Summary: temporary working
294:
1.279 brouard 295: Revision 1.278 2017/07/19 14:09:02 brouard
296: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
297:
1.278 brouard 298: Revision 1.277 2017/07/17 08:53:49 brouard
299: Summary: BOM files can be read now
300:
1.277 brouard 301: Revision 1.276 2017/06/30 15:48:31 brouard
302: Summary: Graphs improvements
303:
1.276 brouard 304: Revision 1.275 2017/06/30 13:39:33 brouard
305: Summary: Saito's color
306:
1.275 brouard 307: Revision 1.274 2017/06/29 09:47:08 brouard
308: Summary: Version 0.99r14
309:
1.274 brouard 310: Revision 1.273 2017/06/27 11:06:02 brouard
311: Summary: More documentation on projections
312:
1.273 brouard 313: Revision 1.272 2017/06/27 10:22:40 brouard
314: Summary: Color of backprojection changed from 6 to 5(yellow)
315:
1.272 brouard 316: Revision 1.271 2017/06/27 10:17:50 brouard
317: Summary: Some bug with rint
318:
1.271 brouard 319: Revision 1.270 2017/05/24 05:45:29 brouard
320: *** empty log message ***
321:
1.270 brouard 322: Revision 1.269 2017/05/23 08:39:25 brouard
323: Summary: Code into subroutine, cleanings
324:
1.269 brouard 325: Revision 1.268 2017/05/18 20:09:32 brouard
326: Summary: backprojection and confidence intervals of backprevalence
327:
1.268 brouard 328: Revision 1.267 2017/05/13 10:25:05 brouard
329: Summary: temporary save for backprojection
330:
1.267 brouard 331: Revision 1.266 2017/05/13 07:26:12 brouard
332: Summary: Version 0.99r13 (improvements and bugs fixed)
333:
1.266 brouard 334: Revision 1.265 2017/04/26 16:22:11 brouard
335: Summary: imach 0.99r13 Some bugs fixed
336:
1.265 brouard 337: Revision 1.264 2017/04/26 06:01:29 brouard
338: Summary: Labels in graphs
339:
1.264 brouard 340: Revision 1.263 2017/04/24 15:23:15 brouard
341: Summary: to save
342:
1.263 brouard 343: Revision 1.262 2017/04/18 16:48:12 brouard
344: *** empty log message ***
345:
1.262 brouard 346: Revision 1.261 2017/04/05 10:14:09 brouard
347: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
348:
1.261 brouard 349: Revision 1.260 2017/04/04 17:46:59 brouard
350: Summary: Gnuplot indexations fixed (humm)
351:
1.260 brouard 352: Revision 1.259 2017/04/04 13:01:16 brouard
353: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
354:
1.259 brouard 355: Revision 1.258 2017/04/03 10:17:47 brouard
356: Summary: Version 0.99r12
357:
358: Some cleanings, conformed with updated documentation.
359:
1.258 brouard 360: Revision 1.257 2017/03/29 16:53:30 brouard
361: Summary: Temp
362:
1.257 brouard 363: Revision 1.256 2017/03/27 05:50:23 brouard
364: Summary: Temporary
365:
1.256 brouard 366: Revision 1.255 2017/03/08 16:02:28 brouard
367: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
368:
1.255 brouard 369: Revision 1.254 2017/03/08 07:13:00 brouard
370: Summary: Fixing data parameter line
371:
1.254 brouard 372: Revision 1.253 2016/12/15 11:59:41 brouard
373: Summary: 0.99 in progress
374:
1.253 brouard 375: Revision 1.252 2016/09/15 21:15:37 brouard
376: *** empty log message ***
377:
1.252 brouard 378: Revision 1.251 2016/09/15 15:01:13 brouard
379: Summary: not working
380:
1.251 brouard 381: Revision 1.250 2016/09/08 16:07:27 brouard
382: Summary: continue
383:
1.250 brouard 384: Revision 1.249 2016/09/07 17:14:18 brouard
385: Summary: Starting values from frequencies
386:
1.249 brouard 387: Revision 1.248 2016/09/07 14:10:18 brouard
388: *** empty log message ***
389:
1.248 brouard 390: Revision 1.247 2016/09/02 11:11:21 brouard
391: *** empty log message ***
392:
1.247 brouard 393: Revision 1.246 2016/09/02 08:49:22 brouard
394: *** empty log message ***
395:
1.246 brouard 396: Revision 1.245 2016/09/02 07:25:01 brouard
397: *** empty log message ***
398:
1.245 brouard 399: Revision 1.244 2016/09/02 07:17:34 brouard
400: *** empty log message ***
401:
1.244 brouard 402: Revision 1.243 2016/09/02 06:45:35 brouard
403: *** empty log message ***
404:
1.243 brouard 405: Revision 1.242 2016/08/30 15:01:20 brouard
406: Summary: Fixing a lots
407:
1.242 brouard 408: Revision 1.241 2016/08/29 17:17:25 brouard
409: Summary: gnuplot problem in Back projection to fix
410:
1.241 brouard 411: Revision 1.240 2016/08/29 07:53:18 brouard
412: Summary: Better
413:
1.240 brouard 414: Revision 1.239 2016/08/26 15:51:03 brouard
415: Summary: Improvement in Powell output in order to copy and paste
416:
417: Author:
418:
1.239 brouard 419: Revision 1.238 2016/08/26 14:23:35 brouard
420: Summary: Starting tests of 0.99
421:
1.238 brouard 422: Revision 1.237 2016/08/26 09:20:19 brouard
423: Summary: to valgrind
424:
1.237 brouard 425: Revision 1.236 2016/08/25 10:50:18 brouard
426: *** empty log message ***
427:
1.236 brouard 428: Revision 1.235 2016/08/25 06:59:23 brouard
429: *** empty log message ***
430:
1.235 brouard 431: Revision 1.234 2016/08/23 16:51:20 brouard
432: *** empty log message ***
433:
1.234 brouard 434: Revision 1.233 2016/08/23 07:40:50 brouard
435: Summary: not working
436:
1.233 brouard 437: Revision 1.232 2016/08/22 14:20:21 brouard
438: Summary: not working
439:
1.232 brouard 440: Revision 1.231 2016/08/22 07:17:15 brouard
441: Summary: not working
442:
1.231 brouard 443: Revision 1.230 2016/08/22 06:55:53 brouard
444: Summary: Not working
445:
1.230 brouard 446: Revision 1.229 2016/07/23 09:45:53 brouard
447: Summary: Completing for func too
448:
1.229 brouard 449: Revision 1.228 2016/07/22 17:45:30 brouard
450: Summary: Fixing some arrays, still debugging
451:
1.227 brouard 452: Revision 1.226 2016/07/12 18:42:34 brouard
453: Summary: temp
454:
1.226 brouard 455: Revision 1.225 2016/07/12 08:40:03 brouard
456: Summary: saving but not running
457:
1.225 brouard 458: Revision 1.224 2016/07/01 13:16:01 brouard
459: Summary: Fixes
460:
1.224 brouard 461: Revision 1.223 2016/02/19 09:23:35 brouard
462: Summary: temporary
463:
1.223 brouard 464: Revision 1.222 2016/02/17 08:14:50 brouard
465: Summary: Probably last 0.98 stable version 0.98r6
466:
1.222 brouard 467: Revision 1.221 2016/02/15 23:35:36 brouard
468: Summary: minor bug
469:
1.220 brouard 470: Revision 1.219 2016/02/15 00:48:12 brouard
471: *** empty log message ***
472:
1.219 brouard 473: Revision 1.218 2016/02/12 11:29:23 brouard
474: Summary: 0.99 Back projections
475:
1.218 brouard 476: Revision 1.217 2015/12/23 17:18:31 brouard
477: Summary: Experimental backcast
478:
1.217 brouard 479: Revision 1.216 2015/12/18 17:32:11 brouard
480: Summary: 0.98r4 Warning and status=-2
481:
482: Version 0.98r4 is now:
483: - displaying an error when status is -1, date of interview unknown and date of death known;
484: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
485: Older changes concerning s=-2, dating from 2005 have been supersed.
486:
1.216 brouard 487: Revision 1.215 2015/12/16 08:52:24 brouard
488: Summary: 0.98r4 working
489:
1.215 brouard 490: Revision 1.214 2015/12/16 06:57:54 brouard
491: Summary: temporary not working
492:
1.214 brouard 493: Revision 1.213 2015/12/11 18:22:17 brouard
494: Summary: 0.98r4
495:
1.213 brouard 496: Revision 1.212 2015/11/21 12:47:24 brouard
497: Summary: minor typo
498:
1.212 brouard 499: Revision 1.211 2015/11/21 12:41:11 brouard
500: Summary: 0.98r3 with some graph of projected cross-sectional
501:
502: Author: Nicolas Brouard
503:
1.211 brouard 504: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 505: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 506: Summary: Adding ftolpl parameter
507: Author: N Brouard
508:
509: We had difficulties to get smoothed confidence intervals. It was due
510: to the period prevalence which wasn't computed accurately. The inner
511: parameter ftolpl is now an outer parameter of the .imach parameter
512: file after estepm. If ftolpl is small 1.e-4 and estepm too,
513: computation are long.
514:
1.209 brouard 515: Revision 1.208 2015/11/17 14:31:57 brouard
516: Summary: temporary
517:
1.208 brouard 518: Revision 1.207 2015/10/27 17:36:57 brouard
519: *** empty log message ***
520:
1.207 brouard 521: Revision 1.206 2015/10/24 07:14:11 brouard
522: *** empty log message ***
523:
1.206 brouard 524: Revision 1.205 2015/10/23 15:50:53 brouard
525: Summary: 0.98r3 some clarification for graphs on likelihood contributions
526:
1.205 brouard 527: Revision 1.204 2015/10/01 16:20:26 brouard
528: Summary: Some new graphs of contribution to likelihood
529:
1.204 brouard 530: Revision 1.203 2015/09/30 17:45:14 brouard
531: Summary: looking at better estimation of the hessian
532:
533: Also a better criteria for convergence to the period prevalence And
534: therefore adding the number of years needed to converge. (The
535: prevalence in any alive state shold sum to one
536:
1.203 brouard 537: Revision 1.202 2015/09/22 19:45:16 brouard
538: Summary: Adding some overall graph on contribution to likelihood. Might change
539:
1.202 brouard 540: Revision 1.201 2015/09/15 17:34:58 brouard
541: Summary: 0.98r0
542:
543: - Some new graphs like suvival functions
544: - Some bugs fixed like model=1+age+V2.
545:
1.201 brouard 546: Revision 1.200 2015/09/09 16:53:55 brouard
547: Summary: Big bug thanks to Flavia
548:
549: Even model=1+age+V2. did not work anymore
550:
1.200 brouard 551: Revision 1.199 2015/09/07 14:09:23 brouard
552: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
553:
1.199 brouard 554: Revision 1.198 2015/09/03 07:14:39 brouard
555: Summary: 0.98q5 Flavia
556:
1.198 brouard 557: Revision 1.197 2015/09/01 18:24:39 brouard
558: *** empty log message ***
559:
1.197 brouard 560: Revision 1.196 2015/08/18 23:17:52 brouard
561: Summary: 0.98q5
562:
1.196 brouard 563: Revision 1.195 2015/08/18 16:28:39 brouard
564: Summary: Adding a hack for testing purpose
565:
566: After reading the title, ftol and model lines, if the comment line has
567: a q, starting with #q, the answer at the end of the run is quit. It
568: permits to run test files in batch with ctest. The former workaround was
569: $ echo q | imach foo.imach
570:
1.195 brouard 571: Revision 1.194 2015/08/18 13:32:00 brouard
572: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
573:
1.194 brouard 574: Revision 1.193 2015/08/04 07:17:42 brouard
575: Summary: 0.98q4
576:
1.193 brouard 577: Revision 1.192 2015/07/16 16:49:02 brouard
578: Summary: Fixing some outputs
579:
1.192 brouard 580: Revision 1.191 2015/07/14 10:00:33 brouard
581: Summary: Some fixes
582:
1.191 brouard 583: Revision 1.190 2015/05/05 08:51:13 brouard
584: Summary: Adding digits in output parameters (7 digits instead of 6)
585:
586: Fix 1+age+.
587:
1.190 brouard 588: Revision 1.189 2015/04/30 14:45:16 brouard
589: Summary: 0.98q2
590:
1.189 brouard 591: Revision 1.188 2015/04/30 08:27:53 brouard
592: *** empty log message ***
593:
1.188 brouard 594: Revision 1.187 2015/04/29 09:11:15 brouard
595: *** empty log message ***
596:
1.187 brouard 597: Revision 1.186 2015/04/23 12:01:52 brouard
598: Summary: V1*age is working now, version 0.98q1
599:
600: Some codes had been disabled in order to simplify and Vn*age was
601: working in the optimization phase, ie, giving correct MLE parameters,
602: but, as usual, outputs were not correct and program core dumped.
603:
1.186 brouard 604: Revision 1.185 2015/03/11 13:26:42 brouard
605: Summary: Inclusion of compile and links command line for Intel Compiler
606:
1.185 brouard 607: Revision 1.184 2015/03/11 11:52:39 brouard
608: Summary: Back from Windows 8. Intel Compiler
609:
1.184 brouard 610: Revision 1.183 2015/03/10 20:34:32 brouard
611: Summary: 0.98q0, trying with directest, mnbrak fixed
612:
613: We use directest instead of original Powell test; probably no
614: incidence on the results, but better justifications;
615: We fixed Numerical Recipes mnbrak routine which was wrong and gave
616: wrong results.
617:
1.183 brouard 618: Revision 1.182 2015/02/12 08:19:57 brouard
619: Summary: Trying to keep directest which seems simpler and more general
620: Author: Nicolas Brouard
621:
1.182 brouard 622: Revision 1.181 2015/02/11 23:22:24 brouard
623: Summary: Comments on Powell added
624:
625: Author:
626:
1.181 brouard 627: Revision 1.180 2015/02/11 17:33:45 brouard
628: Summary: Finishing move from main to function (hpijx and prevalence_limit)
629:
1.180 brouard 630: Revision 1.179 2015/01/04 09:57:06 brouard
631: Summary: back to OS/X
632:
1.179 brouard 633: Revision 1.178 2015/01/04 09:35:48 brouard
634: *** empty log message ***
635:
1.178 brouard 636: Revision 1.177 2015/01/03 18:40:56 brouard
637: Summary: Still testing ilc32 on OSX
638:
1.177 brouard 639: Revision 1.176 2015/01/03 16:45:04 brouard
640: *** empty log message ***
641:
1.176 brouard 642: Revision 1.175 2015/01/03 16:33:42 brouard
643: *** empty log message ***
644:
1.175 brouard 645: Revision 1.174 2015/01/03 16:15:49 brouard
646: Summary: Still in cross-compilation
647:
1.174 brouard 648: Revision 1.173 2015/01/03 12:06:26 brouard
649: Summary: trying to detect cross-compilation
650:
1.173 brouard 651: Revision 1.172 2014/12/27 12:07:47 brouard
652: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
653:
1.172 brouard 654: Revision 1.171 2014/12/23 13:26:59 brouard
655: Summary: Back from Visual C
656:
657: Still problem with utsname.h on Windows
658:
1.171 brouard 659: Revision 1.170 2014/12/23 11:17:12 brouard
660: Summary: Cleaning some \%% back to %%
661:
662: The escape was mandatory for a specific compiler (which one?), but too many warnings.
663:
1.170 brouard 664: Revision 1.169 2014/12/22 23:08:31 brouard
665: Summary: 0.98p
666:
667: Outputs some informations on compiler used, OS etc. Testing on different platforms.
668:
1.169 brouard 669: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 670: Summary: update
1.169 brouard 671:
1.168 brouard 672: Revision 1.167 2014/12/22 13:50:56 brouard
673: Summary: Testing uname and compiler version and if compiled 32 or 64
674:
675: Testing on Linux 64
676:
1.167 brouard 677: Revision 1.166 2014/12/22 11:40:47 brouard
678: *** empty log message ***
679:
1.166 brouard 680: Revision 1.165 2014/12/16 11:20:36 brouard
681: Summary: After compiling on Visual C
682:
683: * imach.c (Module): Merging 1.61 to 1.162
684:
1.165 brouard 685: Revision 1.164 2014/12/16 10:52:11 brouard
686: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
687:
688: * imach.c (Module): Merging 1.61 to 1.162
689:
1.164 brouard 690: Revision 1.163 2014/12/16 10:30:11 brouard
691: * imach.c (Module): Merging 1.61 to 1.162
692:
1.163 brouard 693: Revision 1.162 2014/09/25 11:43:39 brouard
694: Summary: temporary backup 0.99!
695:
1.162 brouard 696: Revision 1.1 2014/09/16 11:06:58 brouard
697: Summary: With some code (wrong) for nlopt
698:
699: Author:
700:
701: Revision 1.161 2014/09/15 20:41:41 brouard
702: Summary: Problem with macro SQR on Intel compiler
703:
1.161 brouard 704: Revision 1.160 2014/09/02 09:24:05 brouard
705: *** empty log message ***
706:
1.160 brouard 707: Revision 1.159 2014/09/01 10:34:10 brouard
708: Summary: WIN32
709: Author: Brouard
710:
1.159 brouard 711: Revision 1.158 2014/08/27 17:11:51 brouard
712: *** empty log message ***
713:
1.158 brouard 714: Revision 1.157 2014/08/27 16:26:55 brouard
715: Summary: Preparing windows Visual studio version
716: Author: Brouard
717:
718: In order to compile on Visual studio, time.h is now correct and time_t
719: and tm struct should be used. difftime should be used but sometimes I
720: just make the differences in raw time format (time(&now).
721: Trying to suppress #ifdef LINUX
722: Add xdg-open for __linux in order to open default browser.
723:
1.157 brouard 724: Revision 1.156 2014/08/25 20:10:10 brouard
725: *** empty log message ***
726:
1.156 brouard 727: Revision 1.155 2014/08/25 18:32:34 brouard
728: Summary: New compile, minor changes
729: Author: Brouard
730:
1.155 brouard 731: Revision 1.154 2014/06/20 17:32:08 brouard
732: Summary: Outputs now all graphs of convergence to period prevalence
733:
1.154 brouard 734: Revision 1.153 2014/06/20 16:45:46 brouard
735: Summary: If 3 live state, convergence to period prevalence on same graph
736: Author: Brouard
737:
1.153 brouard 738: Revision 1.152 2014/06/18 17:54:09 brouard
739: Summary: open browser, use gnuplot on same dir than imach if not found in the path
740:
1.152 brouard 741: Revision 1.151 2014/06/18 16:43:30 brouard
742: *** empty log message ***
743:
1.151 brouard 744: Revision 1.150 2014/06/18 16:42:35 brouard
745: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
746: Author: brouard
747:
1.150 brouard 748: Revision 1.149 2014/06/18 15:51:14 brouard
749: Summary: Some fixes in parameter files errors
750: Author: Nicolas Brouard
751:
1.149 brouard 752: Revision 1.148 2014/06/17 17:38:48 brouard
753: Summary: Nothing new
754: Author: Brouard
755:
756: Just a new packaging for OS/X version 0.98nS
757:
1.148 brouard 758: Revision 1.147 2014/06/16 10:33:11 brouard
759: *** empty log message ***
760:
1.147 brouard 761: Revision 1.146 2014/06/16 10:20:28 brouard
762: Summary: Merge
763: Author: Brouard
764:
765: Merge, before building revised version.
766:
1.146 brouard 767: Revision 1.145 2014/06/10 21:23:15 brouard
768: Summary: Debugging with valgrind
769: Author: Nicolas Brouard
770:
771: Lot of changes in order to output the results with some covariates
772: After the Edimburgh REVES conference 2014, it seems mandatory to
773: improve the code.
774: No more memory valgrind error but a lot has to be done in order to
775: continue the work of splitting the code into subroutines.
776: Also, decodemodel has been improved. Tricode is still not
777: optimal. nbcode should be improved. Documentation has been added in
778: the source code.
779:
1.144 brouard 780: Revision 1.143 2014/01/26 09:45:38 brouard
781: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
782:
783: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
784: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
785:
1.143 brouard 786: Revision 1.142 2014/01/26 03:57:36 brouard
787: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
788:
789: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
790:
1.142 brouard 791: Revision 1.141 2014/01/26 02:42:01 brouard
792: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
793:
1.141 brouard 794: Revision 1.140 2011/09/02 10:37:54 brouard
795: Summary: times.h is ok with mingw32 now.
796:
1.140 brouard 797: Revision 1.139 2010/06/14 07:50:17 brouard
798: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
799: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
800:
1.139 brouard 801: Revision 1.138 2010/04/30 18:19:40 brouard
802: *** empty log message ***
803:
1.138 brouard 804: Revision 1.137 2010/04/29 18:11:38 brouard
805: (Module): Checking covariates for more complex models
806: than V1+V2. A lot of change to be done. Unstable.
807:
1.137 brouard 808: Revision 1.136 2010/04/26 20:30:53 brouard
809: (Module): merging some libgsl code. Fixing computation
810: of likelione (using inter/intrapolation if mle = 0) in order to
811: get same likelihood as if mle=1.
812: Some cleaning of code and comments added.
813:
1.136 brouard 814: Revision 1.135 2009/10/29 15:33:14 brouard
815: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
816:
1.135 brouard 817: Revision 1.134 2009/10/29 13:18:53 brouard
818: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
819:
1.134 brouard 820: Revision 1.133 2009/07/06 10:21:25 brouard
821: just nforces
822:
1.133 brouard 823: Revision 1.132 2009/07/06 08:22:05 brouard
824: Many tings
825:
1.132 brouard 826: Revision 1.131 2009/06/20 16:22:47 brouard
827: Some dimensions resccaled
828:
1.131 brouard 829: Revision 1.130 2009/05/26 06:44:34 brouard
830: (Module): Max Covariate is now set to 20 instead of 8. A
831: lot of cleaning with variables initialized to 0. Trying to make
832: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
833:
1.130 brouard 834: Revision 1.129 2007/08/31 13:49:27 lievre
835: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
836:
1.129 lievre 837: Revision 1.128 2006/06/30 13:02:05 brouard
838: (Module): Clarifications on computing e.j
839:
1.128 brouard 840: Revision 1.127 2006/04/28 18:11:50 brouard
841: (Module): Yes the sum of survivors was wrong since
842: imach-114 because nhstepm was no more computed in the age
843: loop. Now we define nhstepma in the age loop.
844: (Module): In order to speed up (in case of numerous covariates) we
845: compute health expectancies (without variances) in a first step
846: and then all the health expectancies with variances or standard
847: deviation (needs data from the Hessian matrices) which slows the
848: computation.
849: In the future we should be able to stop the program is only health
850: expectancies and graph are needed without standard deviations.
851:
1.127 brouard 852: Revision 1.126 2006/04/28 17:23:28 brouard
853: (Module): Yes the sum of survivors was wrong since
854: imach-114 because nhstepm was no more computed in the age
855: loop. Now we define nhstepma in the age loop.
856: Version 0.98h
857:
1.126 brouard 858: Revision 1.125 2006/04/04 15:20:31 lievre
859: Errors in calculation of health expectancies. Age was not initialized.
860: Forecasting file added.
861:
862: Revision 1.124 2006/03/22 17:13:53 lievre
863: Parameters are printed with %lf instead of %f (more numbers after the comma).
864: The log-likelihood is printed in the log file
865:
866: Revision 1.123 2006/03/20 10:52:43 brouard
867: * imach.c (Module): <title> changed, corresponds to .htm file
868: name. <head> headers where missing.
869:
870: * imach.c (Module): Weights can have a decimal point as for
871: English (a comma might work with a correct LC_NUMERIC environment,
872: otherwise the weight is truncated).
873: Modification of warning when the covariates values are not 0 or
874: 1.
875: Version 0.98g
876:
877: Revision 1.122 2006/03/20 09:45:41 brouard
878: (Module): Weights can have a decimal point as for
879: English (a comma might work with a correct LC_NUMERIC environment,
880: otherwise the weight is truncated).
881: Modification of warning when the covariates values are not 0 or
882: 1.
883: Version 0.98g
884:
885: Revision 1.121 2006/03/16 17:45:01 lievre
886: * imach.c (Module): Comments concerning covariates added
887:
888: * imach.c (Module): refinements in the computation of lli if
889: status=-2 in order to have more reliable computation if stepm is
890: not 1 month. Version 0.98f
891:
892: Revision 1.120 2006/03/16 15:10:38 lievre
893: (Module): refinements in the computation of lli if
894: status=-2 in order to have more reliable computation if stepm is
895: not 1 month. Version 0.98f
896:
897: Revision 1.119 2006/03/15 17:42:26 brouard
898: (Module): Bug if status = -2, the loglikelihood was
899: computed as likelihood omitting the logarithm. Version O.98e
900:
901: Revision 1.118 2006/03/14 18:20:07 brouard
902: (Module): varevsij Comments added explaining the second
903: table of variances if popbased=1 .
904: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
905: (Module): Function pstamp added
906: (Module): Version 0.98d
907:
908: Revision 1.117 2006/03/14 17:16:22 brouard
909: (Module): varevsij Comments added explaining the second
910: table of variances if popbased=1 .
911: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
912: (Module): Function pstamp added
913: (Module): Version 0.98d
914:
915: Revision 1.116 2006/03/06 10:29:27 brouard
916: (Module): Variance-covariance wrong links and
917: varian-covariance of ej. is needed (Saito).
918:
919: Revision 1.115 2006/02/27 12:17:45 brouard
920: (Module): One freematrix added in mlikeli! 0.98c
921:
922: Revision 1.114 2006/02/26 12:57:58 brouard
923: (Module): Some improvements in processing parameter
924: filename with strsep.
925:
926: Revision 1.113 2006/02/24 14:20:24 brouard
927: (Module): Memory leaks checks with valgrind and:
928: datafile was not closed, some imatrix were not freed and on matrix
929: allocation too.
930:
931: Revision 1.112 2006/01/30 09:55:26 brouard
932: (Module): Back to gnuplot.exe instead of wgnuplot.exe
933:
934: Revision 1.111 2006/01/25 20:38:18 brouard
935: (Module): Lots of cleaning and bugs added (Gompertz)
936: (Module): Comments can be added in data file. Missing date values
937: can be a simple dot '.'.
938:
939: Revision 1.110 2006/01/25 00:51:50 brouard
940: (Module): Lots of cleaning and bugs added (Gompertz)
941:
942: Revision 1.109 2006/01/24 19:37:15 brouard
943: (Module): Comments (lines starting with a #) are allowed in data.
944:
945: Revision 1.108 2006/01/19 18:05:42 lievre
946: Gnuplot problem appeared...
947: To be fixed
948:
949: Revision 1.107 2006/01/19 16:20:37 brouard
950: Test existence of gnuplot in imach path
951:
952: Revision 1.106 2006/01/19 13:24:36 brouard
953: Some cleaning and links added in html output
954:
955: Revision 1.105 2006/01/05 20:23:19 lievre
956: *** empty log message ***
957:
958: Revision 1.104 2005/09/30 16:11:43 lievre
959: (Module): sump fixed, loop imx fixed, and simplifications.
960: (Module): If the status is missing at the last wave but we know
961: that the person is alive, then we can code his/her status as -2
962: (instead of missing=-1 in earlier versions) and his/her
963: contributions to the likelihood is 1 - Prob of dying from last
964: health status (= 1-p13= p11+p12 in the easiest case of somebody in
965: the healthy state at last known wave). Version is 0.98
966:
967: Revision 1.103 2005/09/30 15:54:49 lievre
968: (Module): sump fixed, loop imx fixed, and simplifications.
969:
970: Revision 1.102 2004/09/15 17:31:30 brouard
971: Add the possibility to read data file including tab characters.
972:
973: Revision 1.101 2004/09/15 10:38:38 brouard
974: Fix on curr_time
975:
976: Revision 1.100 2004/07/12 18:29:06 brouard
977: Add version for Mac OS X. Just define UNIX in Makefile
978:
979: Revision 1.99 2004/06/05 08:57:40 brouard
980: *** empty log message ***
981:
982: Revision 1.98 2004/05/16 15:05:56 brouard
983: New version 0.97 . First attempt to estimate force of mortality
984: directly from the data i.e. without the need of knowing the health
985: state at each age, but using a Gompertz model: log u =a + b*age .
986: This is the basic analysis of mortality and should be done before any
987: other analysis, in order to test if the mortality estimated from the
988: cross-longitudinal survey is different from the mortality estimated
989: from other sources like vital statistic data.
990:
991: The same imach parameter file can be used but the option for mle should be -3.
992:
1.324 brouard 993: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 994: former routines in order to include the new code within the former code.
995:
996: The output is very simple: only an estimate of the intercept and of
997: the slope with 95% confident intervals.
998:
999: Current limitations:
1000: A) Even if you enter covariates, i.e. with the
1001: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
1002: B) There is no computation of Life Expectancy nor Life Table.
1003:
1004: Revision 1.97 2004/02/20 13:25:42 lievre
1005: Version 0.96d. Population forecasting command line is (temporarily)
1006: suppressed.
1007:
1008: Revision 1.96 2003/07/15 15:38:55 brouard
1009: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
1010: rewritten within the same printf. Workaround: many printfs.
1011:
1012: Revision 1.95 2003/07/08 07:54:34 brouard
1013: * imach.c (Repository):
1014: (Repository): Using imachwizard code to output a more meaningful covariance
1015: matrix (cov(a12,c31) instead of numbers.
1016:
1017: Revision 1.94 2003/06/27 13:00:02 brouard
1018: Just cleaning
1019:
1020: Revision 1.93 2003/06/25 16:33:55 brouard
1021: (Module): On windows (cygwin) function asctime_r doesn't
1022: exist so I changed back to asctime which exists.
1023: (Module): Version 0.96b
1024:
1025: Revision 1.92 2003/06/25 16:30:45 brouard
1026: (Module): On windows (cygwin) function asctime_r doesn't
1027: exist so I changed back to asctime which exists.
1028:
1029: Revision 1.91 2003/06/25 15:30:29 brouard
1030: * imach.c (Repository): Duplicated warning errors corrected.
1031: (Repository): Elapsed time after each iteration is now output. It
1032: helps to forecast when convergence will be reached. Elapsed time
1033: is stamped in powell. We created a new html file for the graphs
1034: concerning matrix of covariance. It has extension -cov.htm.
1035:
1036: Revision 1.90 2003/06/24 12:34:15 brouard
1037: (Module): Some bugs corrected for windows. Also, when
1038: mle=-1 a template is output in file "or"mypar.txt with the design
1039: of the covariance matrix to be input.
1040:
1041: Revision 1.89 2003/06/24 12:30:52 brouard
1042: (Module): Some bugs corrected for windows. Also, when
1043: mle=-1 a template is output in file "or"mypar.txt with the design
1044: of the covariance matrix to be input.
1045:
1046: Revision 1.88 2003/06/23 17:54:56 brouard
1047: * 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.
1048:
1049: Revision 1.87 2003/06/18 12:26:01 brouard
1050: Version 0.96
1051:
1052: Revision 1.86 2003/06/17 20:04:08 brouard
1053: (Module): Change position of html and gnuplot routines and added
1054: routine fileappend.
1055:
1056: Revision 1.85 2003/06/17 13:12:43 brouard
1057: * imach.c (Repository): Check when date of death was earlier that
1058: current date of interview. It may happen when the death was just
1059: prior to the death. In this case, dh was negative and likelihood
1060: was wrong (infinity). We still send an "Error" but patch by
1061: assuming that the date of death was just one stepm after the
1062: interview.
1063: (Repository): Because some people have very long ID (first column)
1064: we changed int to long in num[] and we added a new lvector for
1065: memory allocation. But we also truncated to 8 characters (left
1066: truncation)
1067: (Repository): No more line truncation errors.
1068:
1069: Revision 1.84 2003/06/13 21:44:43 brouard
1070: * imach.c (Repository): Replace "freqsummary" at a correct
1071: place. It differs from routine "prevalence" which may be called
1072: many times. Probs is memory consuming and must be used with
1073: parcimony.
1074: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
1075:
1076: Revision 1.83 2003/06/10 13:39:11 lievre
1077: *** empty log message ***
1078:
1079: Revision 1.82 2003/06/05 15:57:20 brouard
1080: Add log in imach.c and fullversion number is now printed.
1081:
1082: */
1083: /*
1084: Interpolated Markov Chain
1085:
1086: Short summary of the programme:
1087:
1.227 brouard 1088: This program computes Healthy Life Expectancies or State-specific
1089: (if states aren't health statuses) Expectancies from
1090: cross-longitudinal data. Cross-longitudinal data consist in:
1091:
1092: -1- a first survey ("cross") where individuals from different ages
1093: are interviewed on their health status or degree of disability (in
1094: the case of a health survey which is our main interest)
1095:
1096: -2- at least a second wave of interviews ("longitudinal") which
1097: measure each change (if any) in individual health status. Health
1098: expectancies are computed from the time spent in each health state
1099: according to a model. More health states you consider, more time is
1100: necessary to reach the Maximum Likelihood of the parameters involved
1101: in the model. The simplest model is the multinomial logistic model
1102: where pij is the probability to be observed in state j at the second
1103: wave conditional to be observed in state i at the first
1104: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1105: etc , where 'age' is age and 'sex' is a covariate. If you want to
1106: have a more complex model than "constant and age", you should modify
1107: the program where the markup *Covariates have to be included here
1108: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1109: convergence.
1110:
1111: The advantage of this computer programme, compared to a simple
1112: multinomial logistic model, is clear when the delay between waves is not
1113: identical for each individual. Also, if a individual missed an
1114: intermediate interview, the information is lost, but taken into
1115: account using an interpolation or extrapolation.
1116:
1117: hPijx is the probability to be observed in state i at age x+h
1118: conditional to the observed state i at age x. The delay 'h' can be
1119: split into an exact number (nh*stepm) of unobserved intermediate
1120: states. This elementary transition (by month, quarter,
1121: semester or year) is modelled as a multinomial logistic. The hPx
1122: matrix is simply the matrix product of nh*stepm elementary matrices
1123: and the contribution of each individual to the likelihood is simply
1124: hPijx.
1125:
1126: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1127: of the life expectancies. It also computes the period (stable) prevalence.
1128:
1129: Back prevalence and projections:
1.227 brouard 1130:
1131: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1132: double agemaxpar, double ftolpl, int *ncvyearp, double
1133: dateprev1,double dateprev2, int firstpass, int lastpass, int
1134: mobilavproj)
1135:
1136: Computes the back prevalence limit for any combination of
1137: covariate values k at any age between ageminpar and agemaxpar and
1138: returns it in **bprlim. In the loops,
1139:
1140: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1141: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1142:
1143: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1144: Computes for any combination of covariates k and any age between bage and fage
1145: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1146: oldm=oldms;savm=savms;
1.227 brouard 1147:
1.267 brouard 1148: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1149: Computes the transition matrix starting at age 'age' over
1150: 'nhstepm*hstepm*stepm' months (i.e. until
1151: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1152: nhstepm*hstepm matrices.
1153:
1154: Returns p3mat[i][j][h] after calling
1155: p3mat[i][j][h]=matprod2(newm,
1156: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1157: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1158: oldm);
1.226 brouard 1159:
1160: Important routines
1161:
1162: - func (or funcone), computes logit (pij) distinguishing
1163: o fixed variables (single or product dummies or quantitative);
1164: o varying variables by:
1165: (1) wave (single, product dummies, quantitative),
1166: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1167: % fixed dummy (treated) or quantitative (not done because time-consuming);
1168: % varying dummy (not done) or quantitative (not done);
1169: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1170: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1171: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1172: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1173: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1174:
1.226 brouard 1175:
1176:
1.324 brouard 1177: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1178: Institut national d'études démographiques, Paris.
1.126 brouard 1179: This software have been partly granted by Euro-REVES, a concerted action
1180: from the European Union.
1181: It is copyrighted identically to a GNU software product, ie programme and
1182: software can be distributed freely for non commercial use. Latest version
1183: can be accessed at http://euroreves.ined.fr/imach .
1184:
1185: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1186: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1187:
1188: **********************************************************************/
1189: /*
1190: main
1191: read parameterfile
1192: read datafile
1193: concatwav
1194: freqsummary
1195: if (mle >= 1)
1196: mlikeli
1197: print results files
1198: if mle==1
1199: computes hessian
1200: read end of parameter file: agemin, agemax, bage, fage, estepm
1201: begin-prev-date,...
1202: open gnuplot file
1203: open html file
1.145 brouard 1204: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1205: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1206: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1207: freexexit2 possible for memory heap.
1208:
1209: h Pij x | pij_nom ficrestpij
1210: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1211: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1212: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1213:
1214: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1215: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1216: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1217: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1218: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1219:
1.126 brouard 1220: forecasting if prevfcast==1 prevforecast call prevalence()
1221: health expectancies
1222: Variance-covariance of DFLE
1223: prevalence()
1224: movingaverage()
1225: varevsij()
1226: if popbased==1 varevsij(,popbased)
1227: total life expectancies
1228: Variance of period (stable) prevalence
1229: end
1230: */
1231:
1.187 brouard 1232: /* #define DEBUG */
1233: /* #define DEBUGBRENT */
1.203 brouard 1234: /* #define DEBUGLINMIN */
1235: /* #define DEBUGHESS */
1236: #define DEBUGHESSIJ
1.224 brouard 1237: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1238: #define POWELL /* Instead of NLOPT */
1.224 brouard 1239: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1240: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1241: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1242: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1243:
1244: #include <math.h>
1245: #include <stdio.h>
1246: #include <stdlib.h>
1247: #include <string.h>
1.226 brouard 1248: #include <ctype.h>
1.159 brouard 1249:
1250: #ifdef _WIN32
1251: #include <io.h>
1.172 brouard 1252: #include <windows.h>
1253: #include <tchar.h>
1.159 brouard 1254: #else
1.126 brouard 1255: #include <unistd.h>
1.159 brouard 1256: #endif
1.126 brouard 1257:
1258: #include <limits.h>
1259: #include <sys/types.h>
1.171 brouard 1260:
1261: #if defined(__GNUC__)
1262: #include <sys/utsname.h> /* Doesn't work on Windows */
1263: #endif
1264:
1.126 brouard 1265: #include <sys/stat.h>
1266: #include <errno.h>
1.159 brouard 1267: /* extern int errno; */
1.126 brouard 1268:
1.157 brouard 1269: /* #ifdef LINUX */
1270: /* #include <time.h> */
1271: /* #include "timeval.h" */
1272: /* #else */
1273: /* #include <sys/time.h> */
1274: /* #endif */
1275:
1.126 brouard 1276: #include <time.h>
1277:
1.136 brouard 1278: #ifdef GSL
1279: #include <gsl/gsl_errno.h>
1280: #include <gsl/gsl_multimin.h>
1281: #endif
1282:
1.167 brouard 1283:
1.162 brouard 1284: #ifdef NLOPT
1285: #include <nlopt.h>
1286: typedef struct {
1287: double (* function)(double [] );
1288: } myfunc_data ;
1289: #endif
1290:
1.126 brouard 1291: /* #include <libintl.h> */
1292: /* #define _(String) gettext (String) */
1293:
1.251 brouard 1294: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1295:
1296: #define GNUPLOTPROGRAM "gnuplot"
1.343 ! brouard 1297: #define GNUPLOTVERSION 5.1
! 1298: double gnuplotversion=GNUPLOTVERSION;
1.126 brouard 1299: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1300: #define FILENAMELENGTH 256
1.126 brouard 1301:
1302: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1303: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1304:
1.144 brouard 1305: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1306: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1307:
1308: #define NINTERVMAX 8
1.144 brouard 1309: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1310: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1311: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1312: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1313: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1314: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1315: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1316: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1317: /* #define AGESUP 130 */
1.288 brouard 1318: /* #define AGESUP 150 */
1319: #define AGESUP 200
1.268 brouard 1320: #define AGEINF 0
1.218 brouard 1321: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1322: #define AGEBASE 40
1.194 brouard 1323: #define AGEOVERFLOW 1.e20
1.164 brouard 1324: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1325: #ifdef _WIN32
1326: #define DIRSEPARATOR '\\'
1327: #define CHARSEPARATOR "\\"
1328: #define ODIRSEPARATOR '/'
1329: #else
1.126 brouard 1330: #define DIRSEPARATOR '/'
1331: #define CHARSEPARATOR "/"
1332: #define ODIRSEPARATOR '\\'
1333: #endif
1334:
1.343 ! brouard 1335: /* $Id: imach.c,v 1.342 2022/09/11 19:54:09 brouard Exp $ */
1.126 brouard 1336: /* $State: Exp $ */
1.196 brouard 1337: #include "version.h"
1338: char version[]=__IMACH_VERSION__;
1.337 brouard 1339: 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";
1.343 ! brouard 1340: char fullversion[]="$Revision: 1.342 $ $Date: 2022/09/11 19:54:09 $";
1.126 brouard 1341: char strstart[80];
1342: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1343: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.342 brouard 1344: int debugILK=0; /* debugILK is set by a #d in a comment line */
1.187 brouard 1345: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 brouard 1346: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
1347: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
1.335 brouard 1348: 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 1349: 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 1350: int cptcovs=0; /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
1351: int cptcovsnq=0; /**< cptcovsnq number of SIMPLE covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1352: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1353: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.335 brouard 1354: 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 1355: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1356: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.339 brouard 1357: int ncovvt=0; /* Total number of effective (wave) varying covariates (dummy or quantitative or products [without age]) in the model */
1.232 brouard 1358: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1359: int nsd=0; /**< Total number of single dummy variables (output) */
1360: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1361: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1362: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1363: int ntveff=0; /**< ntveff number of effective time varying variables */
1364: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1365: int cptcov=0; /* Working variable */
1.334 brouard 1366: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs+1 declared globally ;*/
1.290 brouard 1367: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1368: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1369: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1370: int nlstate=2; /* Number of live states */
1371: int ndeath=1; /* Number of dead states */
1.130 brouard 1372: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.339 brouard 1373: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable*/
1374: int ncovcolt=0; /* ncovcolt=ncovcol+nqv+ntv+nqtv; total of covariates in the data, not in the model equation*/
1.126 brouard 1375: int popbased=0;
1376:
1377: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1378: int maxwav=0; /* Maxim number of waves */
1379: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1380: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1381: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1382: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1383: int mle=1, weightopt=0;
1.126 brouard 1384: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1385: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1386: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1387: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1388: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1389: int selected(int kvar); /* Is covariate kvar selected for printing results */
1390:
1.130 brouard 1391: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1392: double **matprod2(); /* test */
1.126 brouard 1393: double **oldm, **newm, **savm; /* Working pointers to matrices */
1394: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1395: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1396:
1.136 brouard 1397: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1398: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1399: FILE *ficlog, *ficrespow;
1.130 brouard 1400: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1401: double fretone; /* Only one call to likelihood */
1.130 brouard 1402: long ipmx=0; /* Number of contributions */
1.126 brouard 1403: double sw; /* Sum of weights */
1404: char filerespow[FILENAMELENGTH];
1405: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1406: FILE *ficresilk;
1407: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1408: FILE *ficresprobmorprev;
1409: FILE *fichtm, *fichtmcov; /* Html File */
1410: FILE *ficreseij;
1411: char filerese[FILENAMELENGTH];
1412: FILE *ficresstdeij;
1413: char fileresstde[FILENAMELENGTH];
1414: FILE *ficrescveij;
1415: char filerescve[FILENAMELENGTH];
1416: FILE *ficresvij;
1417: char fileresv[FILENAMELENGTH];
1.269 brouard 1418:
1.126 brouard 1419: char title[MAXLINE];
1.234 brouard 1420: char model[MAXLINE]; /**< The model line */
1.217 brouard 1421: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1422: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1423: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1424: char command[FILENAMELENGTH];
1425: int outcmd=0;
1426:
1.217 brouard 1427: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1428: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1429: char filelog[FILENAMELENGTH]; /* Log file */
1430: char filerest[FILENAMELENGTH];
1431: char fileregp[FILENAMELENGTH];
1432: char popfile[FILENAMELENGTH];
1433:
1434: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1435:
1.157 brouard 1436: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1437: /* struct timezone tzp; */
1438: /* extern int gettimeofday(); */
1439: struct tm tml, *gmtime(), *localtime();
1440:
1441: extern time_t time();
1442:
1443: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1444: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1445: struct tm tm;
1446:
1.126 brouard 1447: char strcurr[80], strfor[80];
1448:
1449: char *endptr;
1450: long lval;
1451: double dval;
1452:
1453: #define NR_END 1
1454: #define FREE_ARG char*
1455: #define FTOL 1.0e-10
1456:
1457: #define NRANSI
1.240 brouard 1458: #define ITMAX 200
1459: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1460:
1461: #define TOL 2.0e-4
1462:
1463: #define CGOLD 0.3819660
1464: #define ZEPS 1.0e-10
1465: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1466:
1467: #define GOLD 1.618034
1468: #define GLIMIT 100.0
1469: #define TINY 1.0e-20
1470:
1471: static double maxarg1,maxarg2;
1472: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1473: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1474:
1475: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1476: #define rint(a) floor(a+0.5)
1.166 brouard 1477: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1478: #define mytinydouble 1.0e-16
1.166 brouard 1479: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1480: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1481: /* static double dsqrarg; */
1482: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1483: static double sqrarg;
1484: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1485: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1486: int agegomp= AGEGOMP;
1487:
1488: int imx;
1489: int stepm=1;
1490: /* Stepm, step in month: minimum step interpolation*/
1491:
1492: int estepm;
1493: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1494:
1495: int m,nb;
1496: long *num;
1.197 brouard 1497: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1498: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1499: covariate for which somebody answered excluding
1500: undefined. Usually 2: 0 and 1. */
1501: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1502: covariate for which somebody answered including
1503: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1504: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1505: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1506: double ***mobaverage, ***mobaverages; /* New global variable */
1.332 brouard 1507: 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 1508: double *ageexmed,*agecens;
1509: double dateintmean=0;
1.296 brouard 1510: double anprojd, mprojd, jprojd; /* For eventual projections */
1511: double anprojf, mprojf, jprojf;
1.126 brouard 1512:
1.296 brouard 1513: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1514: double anbackf, mbackf, jbackf;
1515: double jintmean,mintmean,aintmean;
1.126 brouard 1516: double *weight;
1517: int **s; /* Status */
1.141 brouard 1518: double *agedc;
1.145 brouard 1519: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1520: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1521: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1522: double **coqvar; /* Fixed quantitative covariate nqv */
1.341 brouard 1523: double ***cotvar; /* Time varying covariate start at ncovcol + nqv + (1 to ntv) */
1.225 brouard 1524: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1525: double idx;
1526: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1527: /* Some documentation */
1528: /* Design original data
1529: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1530: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1531: * ntv=3 nqtv=1
1.330 brouard 1532: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1533: * For time varying covariate, quanti or dummies
1534: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1.341 brouard 1535: * cotvar[wav][ncovcol+nqv+ iv(1 to nqtv)][i]= [(1 to nqtv)][i]=(V12) quanti
1.319 brouard 1536: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1537: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1.332 brouard 1538: * covar[Vk,i], value of the Vkth fixed covariate dummy or quanti for individual i:
1.319 brouard 1539: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1540: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1541: * k= 1 2 3 4 5 6 7 8 9 10 11
1542: */
1543: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1544: /* 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
1545: # States 1=Coresidence, 2 Living alone, 3 Institution
1546: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1547: */
1.343 ! brouard 1548: /* V5+V4+ V3+V4*V3 +V5*age+V2 +V1*V2+V1*age+V1 */
! 1549: /* kmodel 1 2 3 4 5 6 7 8 9 */
1.319 brouard 1550: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1551: /* fixed or varying), 1 for age product, 2 for*/
1552: /* product */
1553: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1554: /*(single or product without age), 2 dummy*/
1555: /* with age product, 3 quant with age product*/
1556: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1557: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1.330 brouard 1558: /*TnsdVar[Tvar] 1 2 3 */
1.337 brouard 1559: /*Tvaraff[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1.319 brouard 1560: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1.338 brouard 1561: /*TvarsDind[nsd] 2 3 9 */ /* position K of single dummy cova */
1.319 brouard 1562: /* nsq 1 2 */ /* Counting single quantit tv */
1563: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1564: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1565: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1566: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1567: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1568: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 brouard 1569: /* 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 1570: /* 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 1571: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1572: /* Type */
1573: /* V 1 2 3 4 5 */
1574: /* F F V V V */
1575: /* D Q D D Q */
1576: /* */
1577: int *TvarsD;
1.330 brouard 1578: int *TnsdVar;
1.234 brouard 1579: int *TvarsDind;
1580: int *TvarsQ;
1581: int *TvarsQind;
1582:
1.318 brouard 1583: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1584: int nresult=0;
1.258 brouard 1585: int parameterline=0; /* # of the parameter (type) line */
1.334 brouard 1586: int TKresult[MAXRESULTLINESPONE]; /* TKresult[nres]=k for each resultline nres give the corresponding combination of dummies */
1587: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model corresponds to the k3 position in the resultline */
1588: int modelresult[MAXRESULTLINESPONE][NCOVMAX];/* modelresult[k3]=k1: k1th position in the model corresponds to the k3 position in the resultline */
1589: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1590: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1591: double TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.334 brouard 1592: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1593: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.318 brouard 1594: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1.332 brouard 1595: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.318 brouard 1596:
1597: /* 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
1598: # States 1=Coresidence, 2 Living alone, 3 Institution
1599: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1600: */
1.234 brouard 1601: /* 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 1602: 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 */
1603: 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 */
1604: 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 */
1605: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1606: 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 */
1607: 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 1608: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1609: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1610: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1611: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1612: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1613: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1614: 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 */
1615: 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 1616: int *TvarVV; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1617: int *TvarVVind; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1618: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
1619: /* model V1+V3+age*V1+age*V3+V1*V3 */
1620: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
1621: /* TvarVV={3,1,3}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1622: /* TvarVVind={2,5,5}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1.230 brouard 1623: int *Tvarsel; /**< Selected covariates for output */
1624: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1625: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1626: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1627: 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 1628: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1629: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1630: int *Tage;
1.227 brouard 1631: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1632: 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 1633: 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*/
1634: 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 1635: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1636: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1637: int **Tvard;
1.330 brouard 1638: int **Tvardk;
1.227 brouard 1639: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1640: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1641: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1642: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1643: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1644: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1645: double *lsurv, *lpop, *tpop;
1646:
1.231 brouard 1647: #define FD 1; /* Fixed dummy covariate */
1648: #define FQ 2; /* Fixed quantitative covariate */
1649: #define FP 3; /* Fixed product covariate */
1650: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1651: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1652: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1653: #define VD 10; /* Varying dummy covariate */
1654: #define VQ 11; /* Varying quantitative covariate */
1655: #define VP 12; /* Varying product covariate */
1656: #define VPDD 13; /* Varying product dummy*dummy covariate */
1657: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1658: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1659: #define APFD 16; /* Age product * fixed dummy covariate */
1660: #define APFQ 17; /* Age product * fixed quantitative covariate */
1661: #define APVD 18; /* Age product * varying dummy covariate */
1662: #define APVQ 19; /* Age product * varying quantitative covariate */
1663:
1664: #define FTYPE 1; /* Fixed covariate */
1665: #define VTYPE 2; /* Varying covariate (loop in wave) */
1666: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1667:
1668: struct kmodel{
1669: int maintype; /* main type */
1670: int subtype; /* subtype */
1671: };
1672: struct kmodel modell[NCOVMAX];
1673:
1.143 brouard 1674: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1675: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1676:
1677: /**************** split *************************/
1678: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1679: {
1680: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1681: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1682: */
1683: char *ss; /* pointer */
1.186 brouard 1684: int l1=0, l2=0; /* length counters */
1.126 brouard 1685:
1686: l1 = strlen(path ); /* length of path */
1687: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1688: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1689: if ( ss == NULL ) { /* no directory, so determine current directory */
1690: strcpy( name, path ); /* we got the fullname name because no directory */
1691: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1692: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1693: /* get current working directory */
1694: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1695: #ifdef WIN32
1696: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1697: #else
1698: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1699: #endif
1.126 brouard 1700: return( GLOCK_ERROR_GETCWD );
1701: }
1702: /* got dirc from getcwd*/
1703: printf(" DIRC = %s \n",dirc);
1.205 brouard 1704: } else { /* strip directory from path */
1.126 brouard 1705: ss++; /* after this, the filename */
1706: l2 = strlen( ss ); /* length of filename */
1707: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1708: strcpy( name, ss ); /* save file name */
1709: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1710: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1711: printf(" DIRC2 = %s \n",dirc);
1712: }
1713: /* We add a separator at the end of dirc if not exists */
1714: l1 = strlen( dirc ); /* length of directory */
1715: if( dirc[l1-1] != DIRSEPARATOR ){
1716: dirc[l1] = DIRSEPARATOR;
1717: dirc[l1+1] = 0;
1718: printf(" DIRC3 = %s \n",dirc);
1719: }
1720: ss = strrchr( name, '.' ); /* find last / */
1721: if (ss >0){
1722: ss++;
1723: strcpy(ext,ss); /* save extension */
1724: l1= strlen( name);
1725: l2= strlen(ss)+1;
1726: strncpy( finame, name, l1-l2);
1727: finame[l1-l2]= 0;
1728: }
1729:
1730: return( 0 ); /* we're done */
1731: }
1732:
1733:
1734: /******************************************/
1735:
1736: void replace_back_to_slash(char *s, char*t)
1737: {
1738: int i;
1739: int lg=0;
1740: i=0;
1741: lg=strlen(t);
1742: for(i=0; i<= lg; i++) {
1743: (s[i] = t[i]);
1744: if (t[i]== '\\') s[i]='/';
1745: }
1746: }
1747:
1.132 brouard 1748: char *trimbb(char *out, char *in)
1.137 brouard 1749: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1750: char *s;
1751: s=out;
1752: while (*in != '\0'){
1.137 brouard 1753: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1754: in++;
1755: }
1756: *out++ = *in++;
1757: }
1758: *out='\0';
1759: return s;
1760: }
1761:
1.187 brouard 1762: /* char *substrchaine(char *out, char *in, char *chain) */
1763: /* { */
1764: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1765: /* char *s, *t; */
1766: /* t=in;s=out; */
1767: /* while ((*in != *chain) && (*in != '\0')){ */
1768: /* *out++ = *in++; */
1769: /* } */
1770:
1771: /* /\* *in matches *chain *\/ */
1772: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1773: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1774: /* } */
1775: /* in--; chain--; */
1776: /* while ( (*in != '\0')){ */
1777: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1778: /* *out++ = *in++; */
1779: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1780: /* } */
1781: /* *out='\0'; */
1782: /* out=s; */
1783: /* return out; */
1784: /* } */
1785: char *substrchaine(char *out, char *in, char *chain)
1786: {
1787: /* Substract chain 'chain' from 'in', return and output 'out' */
1788: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1789:
1790: char *strloc;
1791:
1792: strcpy (out, in);
1793: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1794: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1795: if(strloc != NULL){
1796: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1797: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1798: /* strcpy (strloc, strloc +strlen(chain));*/
1799: }
1800: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1801: return out;
1802: }
1803:
1804:
1.145 brouard 1805: char *cutl(char *blocc, char *alocc, char *in, char occ)
1806: {
1.187 brouard 1807: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1808: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1809: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1810: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1811: */
1.160 brouard 1812: char *s, *t;
1.145 brouard 1813: t=in;s=in;
1814: while ((*in != occ) && (*in != '\0')){
1815: *alocc++ = *in++;
1816: }
1817: if( *in == occ){
1818: *(alocc)='\0';
1819: s=++in;
1820: }
1821:
1822: if (s == t) {/* occ not found */
1823: *(alocc-(in-s))='\0';
1824: in=s;
1825: }
1826: while ( *in != '\0'){
1827: *blocc++ = *in++;
1828: }
1829:
1830: *blocc='\0';
1831: return t;
1832: }
1.137 brouard 1833: char *cutv(char *blocc, char *alocc, char *in, char occ)
1834: {
1.187 brouard 1835: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1836: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1837: gives blocc="abcdef2ghi" and alocc="j".
1838: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1839: */
1840: char *s, *t;
1841: t=in;s=in;
1842: while (*in != '\0'){
1843: while( *in == occ){
1844: *blocc++ = *in++;
1845: s=in;
1846: }
1847: *blocc++ = *in++;
1848: }
1849: if (s == t) /* occ not found */
1850: *(blocc-(in-s))='\0';
1851: else
1852: *(blocc-(in-s)-1)='\0';
1853: in=s;
1854: while ( *in != '\0'){
1855: *alocc++ = *in++;
1856: }
1857:
1858: *alocc='\0';
1859: return s;
1860: }
1861:
1.126 brouard 1862: int nbocc(char *s, char occ)
1863: {
1864: int i,j=0;
1865: int lg=20;
1866: i=0;
1867: lg=strlen(s);
1868: for(i=0; i<= lg; i++) {
1.234 brouard 1869: if (s[i] == occ ) j++;
1.126 brouard 1870: }
1871: return j;
1872: }
1873:
1.137 brouard 1874: /* void cutv(char *u,char *v, char*t, char occ) */
1875: /* { */
1876: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1877: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1878: /* gives u="abcdef2ghi" and v="j" *\/ */
1879: /* int i,lg,j,p=0; */
1880: /* i=0; */
1881: /* lg=strlen(t); */
1882: /* for(j=0; j<=lg-1; j++) { */
1883: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1884: /* } */
1.126 brouard 1885:
1.137 brouard 1886: /* for(j=0; j<p; j++) { */
1887: /* (u[j] = t[j]); */
1888: /* } */
1889: /* u[p]='\0'; */
1.126 brouard 1890:
1.137 brouard 1891: /* for(j=0; j<= lg; j++) { */
1892: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1893: /* } */
1894: /* } */
1.126 brouard 1895:
1.160 brouard 1896: #ifdef _WIN32
1897: char * strsep(char **pp, const char *delim)
1898: {
1899: char *p, *q;
1900:
1901: if ((p = *pp) == NULL)
1902: return 0;
1903: if ((q = strpbrk (p, delim)) != NULL)
1904: {
1905: *pp = q + 1;
1906: *q = '\0';
1907: }
1908: else
1909: *pp = 0;
1910: return p;
1911: }
1912: #endif
1913:
1.126 brouard 1914: /********************** nrerror ********************/
1915:
1916: void nrerror(char error_text[])
1917: {
1918: fprintf(stderr,"ERREUR ...\n");
1919: fprintf(stderr,"%s\n",error_text);
1920: exit(EXIT_FAILURE);
1921: }
1922: /*********************** vector *******************/
1923: double *vector(int nl, int nh)
1924: {
1925: double *v;
1926: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1927: if (!v) nrerror("allocation failure in vector");
1928: return v-nl+NR_END;
1929: }
1930:
1931: /************************ free vector ******************/
1932: void free_vector(double*v, int nl, int nh)
1933: {
1934: free((FREE_ARG)(v+nl-NR_END));
1935: }
1936:
1937: /************************ivector *******************************/
1938: int *ivector(long nl,long nh)
1939: {
1940: int *v;
1941: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1942: if (!v) nrerror("allocation failure in ivector");
1943: return v-nl+NR_END;
1944: }
1945:
1946: /******************free ivector **************************/
1947: void free_ivector(int *v, long nl, long nh)
1948: {
1949: free((FREE_ARG)(v+nl-NR_END));
1950: }
1951:
1952: /************************lvector *******************************/
1953: long *lvector(long nl,long nh)
1954: {
1955: long *v;
1956: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1957: if (!v) nrerror("allocation failure in ivector");
1958: return v-nl+NR_END;
1959: }
1960:
1961: /******************free lvector **************************/
1962: void free_lvector(long *v, long nl, long nh)
1963: {
1964: free((FREE_ARG)(v+nl-NR_END));
1965: }
1966:
1967: /******************* imatrix *******************************/
1968: int **imatrix(long nrl, long nrh, long ncl, long nch)
1969: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1970: {
1971: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1972: int **m;
1973:
1974: /* allocate pointers to rows */
1975: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1976: if (!m) nrerror("allocation failure 1 in matrix()");
1977: m += NR_END;
1978: m -= nrl;
1979:
1980:
1981: /* allocate rows and set pointers to them */
1982: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1983: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1984: m[nrl] += NR_END;
1985: m[nrl] -= ncl;
1986:
1987: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1988:
1989: /* return pointer to array of pointers to rows */
1990: return m;
1991: }
1992:
1993: /****************** free_imatrix *************************/
1994: void free_imatrix(m,nrl,nrh,ncl,nch)
1995: int **m;
1996: long nch,ncl,nrh,nrl;
1997: /* free an int matrix allocated by imatrix() */
1998: {
1999: free((FREE_ARG) (m[nrl]+ncl-NR_END));
2000: free((FREE_ARG) (m+nrl-NR_END));
2001: }
2002:
2003: /******************* matrix *******************************/
2004: double **matrix(long nrl, long nrh, long ncl, long nch)
2005: {
2006: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
2007: double **m;
2008:
2009: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2010: if (!m) nrerror("allocation failure 1 in matrix()");
2011: m += NR_END;
2012: m -= nrl;
2013:
2014: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2015: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2016: m[nrl] += NR_END;
2017: m[nrl] -= ncl;
2018:
2019: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2020: return m;
1.145 brouard 2021: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
2022: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
2023: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 2024: */
2025: }
2026:
2027: /*************************free matrix ************************/
2028: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
2029: {
2030: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2031: free((FREE_ARG)(m+nrl-NR_END));
2032: }
2033:
2034: /******************* ma3x *******************************/
2035: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
2036: {
2037: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
2038: double ***m;
2039:
2040: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2041: if (!m) nrerror("allocation failure 1 in matrix()");
2042: m += NR_END;
2043: m -= nrl;
2044:
2045: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2046: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2047: m[nrl] += NR_END;
2048: m[nrl] -= ncl;
2049:
2050: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2051:
2052: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
2053: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
2054: m[nrl][ncl] += NR_END;
2055: m[nrl][ncl] -= nll;
2056: for (j=ncl+1; j<=nch; j++)
2057: m[nrl][j]=m[nrl][j-1]+nlay;
2058:
2059: for (i=nrl+1; i<=nrh; i++) {
2060: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
2061: for (j=ncl+1; j<=nch; j++)
2062: m[i][j]=m[i][j-1]+nlay;
2063: }
2064: return m;
2065: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
2066: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
2067: */
2068: }
2069:
2070: /*************************free ma3x ************************/
2071: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
2072: {
2073: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
2074: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2075: free((FREE_ARG)(m+nrl-NR_END));
2076: }
2077:
2078: /*************** function subdirf ***********/
2079: char *subdirf(char fileres[])
2080: {
2081: /* Caution optionfilefiname is hidden */
2082: strcpy(tmpout,optionfilefiname);
2083: strcat(tmpout,"/"); /* Add to the right */
2084: strcat(tmpout,fileres);
2085: return tmpout;
2086: }
2087:
2088: /*************** function subdirf2 ***********/
2089: char *subdirf2(char fileres[], char *preop)
2090: {
1.314 brouard 2091: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
2092: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 2093: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 2094: /* Caution optionfilefiname is hidden */
2095: strcpy(tmpout,optionfilefiname);
2096: strcat(tmpout,"/");
2097: strcat(tmpout,preop);
2098: strcat(tmpout,fileres);
2099: return tmpout;
2100: }
2101:
2102: /*************** function subdirf3 ***********/
2103: char *subdirf3(char fileres[], char *preop, char *preop2)
2104: {
2105:
2106: /* Caution optionfilefiname is hidden */
2107: strcpy(tmpout,optionfilefiname);
2108: strcat(tmpout,"/");
2109: strcat(tmpout,preop);
2110: strcat(tmpout,preop2);
2111: strcat(tmpout,fileres);
2112: return tmpout;
2113: }
1.213 brouard 2114:
2115: /*************** function subdirfext ***********/
2116: char *subdirfext(char fileres[], char *preop, char *postop)
2117: {
2118:
2119: strcpy(tmpout,preop);
2120: strcat(tmpout,fileres);
2121: strcat(tmpout,postop);
2122: return tmpout;
2123: }
1.126 brouard 2124:
1.213 brouard 2125: /*************** function subdirfext3 ***********/
2126: char *subdirfext3(char fileres[], char *preop, char *postop)
2127: {
2128:
2129: /* Caution optionfilefiname is hidden */
2130: strcpy(tmpout,optionfilefiname);
2131: strcat(tmpout,"/");
2132: strcat(tmpout,preop);
2133: strcat(tmpout,fileres);
2134: strcat(tmpout,postop);
2135: return tmpout;
2136: }
2137:
1.162 brouard 2138: char *asc_diff_time(long time_sec, char ascdiff[])
2139: {
2140: long sec_left, days, hours, minutes;
2141: days = (time_sec) / (60*60*24);
2142: sec_left = (time_sec) % (60*60*24);
2143: hours = (sec_left) / (60*60) ;
2144: sec_left = (sec_left) %(60*60);
2145: minutes = (sec_left) /60;
2146: sec_left = (sec_left) % (60);
2147: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2148: return ascdiff;
2149: }
2150:
1.126 brouard 2151: /***************** f1dim *************************/
2152: extern int ncom;
2153: extern double *pcom,*xicom;
2154: extern double (*nrfunc)(double []);
2155:
2156: double f1dim(double x)
2157: {
2158: int j;
2159: double f;
2160: double *xt;
2161:
2162: xt=vector(1,ncom);
2163: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2164: f=(*nrfunc)(xt);
2165: free_vector(xt,1,ncom);
2166: return f;
2167: }
2168:
2169: /*****************brent *************************/
2170: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2171: {
2172: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2173: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2174: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2175: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2176: * returned function value.
2177: */
1.126 brouard 2178: int iter;
2179: double a,b,d,etemp;
1.159 brouard 2180: double fu=0,fv,fw,fx;
1.164 brouard 2181: double ftemp=0.;
1.126 brouard 2182: double p,q,r,tol1,tol2,u,v,w,x,xm;
2183: double e=0.0;
2184:
2185: a=(ax < cx ? ax : cx);
2186: b=(ax > cx ? ax : cx);
2187: x=w=v=bx;
2188: fw=fv=fx=(*f)(x);
2189: for (iter=1;iter<=ITMAX;iter++) {
2190: xm=0.5*(a+b);
2191: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2192: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2193: printf(".");fflush(stdout);
2194: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2195: #ifdef DEBUGBRENT
1.126 brouard 2196: 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);
2197: 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);
2198: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2199: #endif
2200: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2201: *xmin=x;
2202: return fx;
2203: }
2204: ftemp=fu;
2205: if (fabs(e) > tol1) {
2206: r=(x-w)*(fx-fv);
2207: q=(x-v)*(fx-fw);
2208: p=(x-v)*q-(x-w)*r;
2209: q=2.0*(q-r);
2210: if (q > 0.0) p = -p;
2211: q=fabs(q);
2212: etemp=e;
2213: e=d;
2214: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2215: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2216: else {
1.224 brouard 2217: d=p/q;
2218: u=x+d;
2219: if (u-a < tol2 || b-u < tol2)
2220: d=SIGN(tol1,xm-x);
1.126 brouard 2221: }
2222: } else {
2223: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2224: }
2225: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2226: fu=(*f)(u);
2227: if (fu <= fx) {
2228: if (u >= x) a=x; else b=x;
2229: SHFT(v,w,x,u)
1.183 brouard 2230: SHFT(fv,fw,fx,fu)
2231: } else {
2232: if (u < x) a=u; else b=u;
2233: if (fu <= fw || w == x) {
1.224 brouard 2234: v=w;
2235: w=u;
2236: fv=fw;
2237: fw=fu;
1.183 brouard 2238: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2239: v=u;
2240: fv=fu;
1.183 brouard 2241: }
2242: }
1.126 brouard 2243: }
2244: nrerror("Too many iterations in brent");
2245: *xmin=x;
2246: return fx;
2247: }
2248:
2249: /****************** mnbrak ***********************/
2250:
2251: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2252: double (*func)(double))
1.183 brouard 2253: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2254: the downhill direction (defined by the function as evaluated at the initial points) and returns
2255: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2256: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2257: */
1.126 brouard 2258: double ulim,u,r,q, dum;
2259: double fu;
1.187 brouard 2260:
2261: double scale=10.;
2262: int iterscale=0;
2263:
2264: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2265: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2266:
2267:
2268: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2269: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2270: /* *bx = *ax - (*ax - *bx)/scale; */
2271: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2272: /* } */
2273:
1.126 brouard 2274: if (*fb > *fa) {
2275: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2276: SHFT(dum,*fb,*fa,dum)
2277: }
1.126 brouard 2278: *cx=(*bx)+GOLD*(*bx-*ax);
2279: *fc=(*func)(*cx);
1.183 brouard 2280: #ifdef DEBUG
1.224 brouard 2281: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2282: 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 2283: #endif
1.224 brouard 2284: 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 2285: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2286: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2287: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2288: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2289: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2290: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2291: fu=(*func)(u);
1.163 brouard 2292: #ifdef DEBUG
2293: /* f(x)=A(x-u)**2+f(u) */
2294: double A, fparabu;
2295: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2296: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2297: 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);
2298: 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 2299: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2300: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2301: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2302: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2303: #endif
1.184 brouard 2304: #ifdef MNBRAKORIGINAL
1.183 brouard 2305: #else
1.191 brouard 2306: /* if (fu > *fc) { */
2307: /* #ifdef DEBUG */
2308: /* printf("mnbrak4 fu > fc \n"); */
2309: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2310: /* #endif */
2311: /* /\* 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 *\\/ *\/ */
2312: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2313: /* dum=u; /\* Shifting c and u *\/ */
2314: /* u = *cx; */
2315: /* *cx = dum; */
2316: /* dum = fu; */
2317: /* fu = *fc; */
2318: /* *fc =dum; */
2319: /* } else { /\* end *\/ */
2320: /* #ifdef DEBUG */
2321: /* printf("mnbrak3 fu < fc \n"); */
2322: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2323: /* #endif */
2324: /* dum=u; /\* Shifting c and u *\/ */
2325: /* u = *cx; */
2326: /* *cx = dum; */
2327: /* dum = fu; */
2328: /* fu = *fc; */
2329: /* *fc =dum; */
2330: /* } */
1.224 brouard 2331: #ifdef DEBUGMNBRAK
2332: double A, fparabu;
2333: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2334: fparabu= *fa - A*(*ax-u)*(*ax-u);
2335: 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);
2336: 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 2337: #endif
1.191 brouard 2338: dum=u; /* Shifting c and u */
2339: u = *cx;
2340: *cx = dum;
2341: dum = fu;
2342: fu = *fc;
2343: *fc =dum;
1.183 brouard 2344: #endif
1.162 brouard 2345: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2346: #ifdef DEBUG
1.224 brouard 2347: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2348: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2349: #endif
1.126 brouard 2350: fu=(*func)(u);
2351: if (fu < *fc) {
1.183 brouard 2352: #ifdef DEBUG
1.224 brouard 2353: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2354: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2355: #endif
2356: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2357: SHFT(*fb,*fc,fu,(*func)(u))
2358: #ifdef DEBUG
2359: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2360: #endif
2361: }
1.162 brouard 2362: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2363: #ifdef DEBUG
1.224 brouard 2364: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2365: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2366: #endif
1.126 brouard 2367: u=ulim;
2368: fu=(*func)(u);
1.183 brouard 2369: } else { /* u could be left to b (if r > q parabola has a maximum) */
2370: #ifdef DEBUG
1.224 brouard 2371: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2372: 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 2373: #endif
1.126 brouard 2374: u=(*cx)+GOLD*(*cx-*bx);
2375: fu=(*func)(u);
1.224 brouard 2376: #ifdef DEBUG
2377: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2378: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2379: #endif
1.183 brouard 2380: } /* end tests */
1.126 brouard 2381: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2382: SHFT(*fa,*fb,*fc,fu)
2383: #ifdef DEBUG
1.224 brouard 2384: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2385: 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 2386: #endif
2387: } /* 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 2388: }
2389:
2390: /*************** linmin ************************/
1.162 brouard 2391: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2392: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2393: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2394: the value of func at the returned location p . This is actually all accomplished by calling the
2395: routines mnbrak and brent .*/
1.126 brouard 2396: int ncom;
2397: double *pcom,*xicom;
2398: double (*nrfunc)(double []);
2399:
1.224 brouard 2400: #ifdef LINMINORIGINAL
1.126 brouard 2401: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2402: #else
2403: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2404: #endif
1.126 brouard 2405: {
2406: double brent(double ax, double bx, double cx,
2407: double (*f)(double), double tol, double *xmin);
2408: double f1dim(double x);
2409: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2410: double *fc, double (*func)(double));
2411: int j;
2412: double xx,xmin,bx,ax;
2413: double fx,fb,fa;
1.187 brouard 2414:
1.203 brouard 2415: #ifdef LINMINORIGINAL
2416: #else
2417: double scale=10., axs, xxs; /* Scale added for infinity */
2418: #endif
2419:
1.126 brouard 2420: ncom=n;
2421: pcom=vector(1,n);
2422: xicom=vector(1,n);
2423: nrfunc=func;
2424: for (j=1;j<=n;j++) {
2425: pcom[j]=p[j];
1.202 brouard 2426: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2427: }
1.187 brouard 2428:
1.203 brouard 2429: #ifdef LINMINORIGINAL
2430: xx=1.;
2431: #else
2432: axs=0.0;
2433: xxs=1.;
2434: do{
2435: xx= xxs;
2436: #endif
1.187 brouard 2437: ax=0.;
2438: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2439: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2440: /* 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)) */
2441: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2442: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2443: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2444: /* 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 2445: #ifdef LINMINORIGINAL
2446: #else
2447: if (fx != fx){
1.224 brouard 2448: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2449: printf("|");
2450: fprintf(ficlog,"|");
1.203 brouard 2451: #ifdef DEBUGLINMIN
1.224 brouard 2452: 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 2453: #endif
2454: }
1.224 brouard 2455: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2456: #endif
2457:
1.191 brouard 2458: #ifdef DEBUGLINMIN
2459: 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 2460: 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 2461: #endif
1.224 brouard 2462: #ifdef LINMINORIGINAL
2463: #else
1.317 brouard 2464: if(fb == fx){ /* Flat function in the direction */
2465: xmin=xx;
1.224 brouard 2466: *flat=1;
1.317 brouard 2467: }else{
1.224 brouard 2468: *flat=0;
2469: #endif
2470: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2471: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2472: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2473: /* fmin = f(p[j] + xmin * xi[j]) */
2474: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2475: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2476: #ifdef DEBUG
1.224 brouard 2477: 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);
2478: 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);
2479: #endif
2480: #ifdef LINMINORIGINAL
2481: #else
2482: }
1.126 brouard 2483: #endif
1.191 brouard 2484: #ifdef DEBUGLINMIN
2485: printf("linmin end ");
1.202 brouard 2486: fprintf(ficlog,"linmin end ");
1.191 brouard 2487: #endif
1.126 brouard 2488: for (j=1;j<=n;j++) {
1.203 brouard 2489: #ifdef LINMINORIGINAL
2490: xi[j] *= xmin;
2491: #else
2492: #ifdef DEBUGLINMIN
2493: if(xxs <1.0)
2494: printf(" before xi[%d]=%12.8f", j,xi[j]);
2495: #endif
2496: 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) */
2497: #ifdef DEBUGLINMIN
2498: if(xxs <1.0)
2499: 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 );
2500: #endif
2501: #endif
1.187 brouard 2502: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2503: }
1.191 brouard 2504: #ifdef DEBUGLINMIN
1.203 brouard 2505: printf("\n");
1.191 brouard 2506: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2507: 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 2508: for (j=1;j<=n;j++) {
1.202 brouard 2509: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2510: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2511: if(j % ncovmodel == 0){
1.191 brouard 2512: printf("\n");
1.202 brouard 2513: fprintf(ficlog,"\n");
2514: }
1.191 brouard 2515: }
1.203 brouard 2516: #else
1.191 brouard 2517: #endif
1.126 brouard 2518: free_vector(xicom,1,n);
2519: free_vector(pcom,1,n);
2520: }
2521:
2522:
2523: /*************** powell ************************/
1.162 brouard 2524: /*
1.317 brouard 2525: Minimization of a function func of n variables. Input consists in an initial starting point
2526: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2527: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2528: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2529: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2530: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2531: */
1.224 brouard 2532: #ifdef LINMINORIGINAL
2533: #else
2534: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2535: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2536: #endif
1.126 brouard 2537: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2538: double (*func)(double []))
2539: {
1.224 brouard 2540: #ifdef LINMINORIGINAL
2541: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2542: double (*func)(double []));
1.224 brouard 2543: #else
1.241 brouard 2544: void linmin(double p[], double xi[], int n, double *fret,
2545: double (*func)(double []),int *flat);
1.224 brouard 2546: #endif
1.239 brouard 2547: int i,ibig,j,jk,k;
1.126 brouard 2548: double del,t,*pt,*ptt,*xit;
1.181 brouard 2549: double directest;
1.126 brouard 2550: double fp,fptt;
2551: double *xits;
2552: int niterf, itmp;
2553:
2554: pt=vector(1,n);
2555: ptt=vector(1,n);
2556: xit=vector(1,n);
2557: xits=vector(1,n);
2558: *fret=(*func)(p);
2559: for (j=1;j<=n;j++) pt[j]=p[j];
1.338 brouard 2560: rcurr_time = time(NULL);
2561: fp=(*fret); /* Initialisation */
1.126 brouard 2562: for (*iter=1;;++(*iter)) {
2563: ibig=0;
2564: del=0.0;
1.157 brouard 2565: rlast_time=rcurr_time;
2566: /* (void) gettimeofday(&curr_time,&tzp); */
2567: rcurr_time = time(NULL);
2568: curr_time = *localtime(&rcurr_time);
1.337 brouard 2569: /* 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); */
2570: /* 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); */
2571: 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);
2572: 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 2573: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2574: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2575: for (i=1;i<=n;i++) {
1.126 brouard 2576: fprintf(ficrespow," %.12lf", p[i]);
2577: }
1.239 brouard 2578: fprintf(ficrespow,"\n");fflush(ficrespow);
2579: printf("\n#model= 1 + age ");
2580: fprintf(ficlog,"\n#model= 1 + age ");
2581: if(nagesqr==1){
1.241 brouard 2582: printf(" + age*age ");
2583: fprintf(ficlog," + age*age ");
1.239 brouard 2584: }
2585: for(j=1;j <=ncovmodel-2;j++){
2586: if(Typevar[j]==0) {
2587: printf(" + V%d ",Tvar[j]);
2588: fprintf(ficlog," + V%d ",Tvar[j]);
2589: }else if(Typevar[j]==1) {
2590: printf(" + V%d*age ",Tvar[j]);
2591: fprintf(ficlog," + V%d*age ",Tvar[j]);
2592: }else if(Typevar[j]==2) {
2593: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2594: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2595: }
2596: }
1.126 brouard 2597: printf("\n");
1.239 brouard 2598: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2599: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2600: fprintf(ficlog,"\n");
1.239 brouard 2601: for(i=1,jk=1; i <=nlstate; i++){
2602: for(k=1; k <=(nlstate+ndeath); k++){
2603: if (k != i) {
2604: printf("%d%d ",i,k);
2605: fprintf(ficlog,"%d%d ",i,k);
2606: for(j=1; j <=ncovmodel; j++){
2607: printf("%12.7f ",p[jk]);
2608: fprintf(ficlog,"%12.7f ",p[jk]);
2609: jk++;
2610: }
2611: printf("\n");
2612: fprintf(ficlog,"\n");
2613: }
2614: }
2615: }
1.241 brouard 2616: if(*iter <=3 && *iter >1){
1.157 brouard 2617: tml = *localtime(&rcurr_time);
2618: strcpy(strcurr,asctime(&tml));
2619: rforecast_time=rcurr_time;
1.126 brouard 2620: itmp = strlen(strcurr);
2621: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2622: strcurr[itmp-1]='\0';
1.162 brouard 2623: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2624: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2625: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2626: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2627: forecast_time = *localtime(&rforecast_time);
2628: strcpy(strfor,asctime(&forecast_time));
2629: itmp = strlen(strfor);
2630: if(strfor[itmp-1]=='\n')
2631: strfor[itmp-1]='\0';
2632: 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);
2633: 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 2634: }
2635: }
1.187 brouard 2636: for (i=1;i<=n;i++) { /* For each direction i */
2637: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2638: fptt=(*fret);
2639: #ifdef DEBUG
1.203 brouard 2640: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2641: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2642: #endif
1.203 brouard 2643: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2644: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2645: #ifdef LINMINORIGINAL
1.188 brouard 2646: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2647: #else
2648: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2649: flatdir[i]=flat; /* Function is vanishing in that direction i */
2650: #endif
2651: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2652: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2653: /* because that direction will be replaced unless the gain del is small */
2654: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2655: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2656: /* with the new direction. */
2657: del=fabs(fptt-(*fret));
2658: ibig=i;
1.126 brouard 2659: }
2660: #ifdef DEBUG
2661: printf("%d %.12e",i,(*fret));
2662: fprintf(ficlog,"%d %.12e",i,(*fret));
2663: for (j=1;j<=n;j++) {
1.224 brouard 2664: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2665: printf(" x(%d)=%.12e",j,xit[j]);
2666: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2667: }
2668: for(j=1;j<=n;j++) {
1.225 brouard 2669: printf(" p(%d)=%.12e",j,p[j]);
2670: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2671: }
2672: printf("\n");
2673: fprintf(ficlog,"\n");
2674: #endif
1.187 brouard 2675: } /* end loop on each direction i */
2676: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2677: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2678: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2679: for(j=1;j<=n;j++) {
2680: if(flatdir[j] >0){
2681: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2682: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2683: }
1.319 brouard 2684: /* printf("\n"); */
2685: /* fprintf(ficlog,"\n"); */
2686: }
1.243 brouard 2687: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2688: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2689: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2690: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2691: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2692: /* decreased of more than 3.84 */
2693: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2694: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2695: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2696:
1.188 brouard 2697: /* Starting the program with initial values given by a former maximization will simply change */
2698: /* the scales of the directions and the directions, because the are reset to canonical directions */
2699: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2700: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2701: #ifdef DEBUG
2702: int k[2],l;
2703: k[0]=1;
2704: k[1]=-1;
2705: printf("Max: %.12e",(*func)(p));
2706: fprintf(ficlog,"Max: %.12e",(*func)(p));
2707: for (j=1;j<=n;j++) {
2708: printf(" %.12e",p[j]);
2709: fprintf(ficlog," %.12e",p[j]);
2710: }
2711: printf("\n");
2712: fprintf(ficlog,"\n");
2713: for(l=0;l<=1;l++) {
2714: for (j=1;j<=n;j++) {
2715: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2716: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2717: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2718: }
2719: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2720: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2721: }
2722: #endif
2723:
2724: free_vector(xit,1,n);
2725: free_vector(xits,1,n);
2726: free_vector(ptt,1,n);
2727: free_vector(pt,1,n);
2728: return;
1.192 brouard 2729: } /* enough precision */
1.240 brouard 2730: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2731: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2732: ptt[j]=2.0*p[j]-pt[j];
2733: xit[j]=p[j]-pt[j];
2734: pt[j]=p[j];
2735: }
1.181 brouard 2736: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2737: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2738: if (*iter <=4) {
1.225 brouard 2739: #else
2740: #endif
1.224 brouard 2741: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2742: #else
1.161 brouard 2743: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2744: #endif
1.162 brouard 2745: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2746: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2747: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2748: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2749: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2750: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2751: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2752: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2753: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2754: /* Even if f3 <f1, directest can be negative and t >0 */
2755: /* mu² and del² are equal when f3=f1 */
2756: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2757: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2758: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2759: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2760: #ifdef NRCORIGINAL
2761: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2762: #else
2763: 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 2764: t= t- del*SQR(fp-fptt);
1.183 brouard 2765: #endif
1.202 brouard 2766: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2767: #ifdef DEBUG
1.181 brouard 2768: 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);
2769: 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 2770: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2771: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2772: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2773: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2774: 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);
2775: 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);
2776: #endif
1.183 brouard 2777: #ifdef POWELLORIGINAL
2778: if (t < 0.0) { /* Then we use it for new direction */
2779: #else
1.182 brouard 2780: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2781: 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 2782: 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 2783: 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 2784: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2785: }
1.181 brouard 2786: if (directest < 0.0) { /* Then we use it for new direction */
2787: #endif
1.191 brouard 2788: #ifdef DEBUGLINMIN
1.234 brouard 2789: printf("Before linmin in direction P%d-P0\n",n);
2790: for (j=1;j<=n;j++) {
2791: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2792: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2793: if(j % ncovmodel == 0){
2794: printf("\n");
2795: fprintf(ficlog,"\n");
2796: }
2797: }
1.224 brouard 2798: #endif
2799: #ifdef LINMINORIGINAL
1.234 brouard 2800: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2801: #else
1.234 brouard 2802: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2803: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2804: #endif
1.234 brouard 2805:
1.191 brouard 2806: #ifdef DEBUGLINMIN
1.234 brouard 2807: for (j=1;j<=n;j++) {
2808: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2809: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2810: if(j % ncovmodel == 0){
2811: printf("\n");
2812: fprintf(ficlog,"\n");
2813: }
2814: }
1.224 brouard 2815: #endif
1.234 brouard 2816: for (j=1;j<=n;j++) {
2817: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2818: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2819: }
1.224 brouard 2820: #ifdef LINMINORIGINAL
2821: #else
1.234 brouard 2822: for (j=1, flatd=0;j<=n;j++) {
2823: if(flatdir[j]>0)
2824: flatd++;
2825: }
2826: if(flatd >0){
1.255 brouard 2827: printf("%d flat directions: ",flatd);
2828: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2829: for (j=1;j<=n;j++) {
2830: if(flatdir[j]>0){
2831: printf("%d ",j);
2832: fprintf(ficlog,"%d ",j);
2833: }
2834: }
2835: printf("\n");
2836: fprintf(ficlog,"\n");
1.319 brouard 2837: #ifdef FLATSUP
2838: free_vector(xit,1,n);
2839: free_vector(xits,1,n);
2840: free_vector(ptt,1,n);
2841: free_vector(pt,1,n);
2842: return;
2843: #endif
1.234 brouard 2844: }
1.191 brouard 2845: #endif
1.234 brouard 2846: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2847: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2848:
1.126 brouard 2849: #ifdef DEBUG
1.234 brouard 2850: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2851: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2852: for(j=1;j<=n;j++){
2853: printf(" %lf",xit[j]);
2854: fprintf(ficlog," %lf",xit[j]);
2855: }
2856: printf("\n");
2857: fprintf(ficlog,"\n");
1.126 brouard 2858: #endif
1.192 brouard 2859: } /* end of t or directest negative */
1.224 brouard 2860: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2861: #else
1.234 brouard 2862: } /* end if (fptt < fp) */
1.192 brouard 2863: #endif
1.225 brouard 2864: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2865: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2866: #else
1.224 brouard 2867: #endif
1.234 brouard 2868: } /* loop iteration */
1.126 brouard 2869: }
1.234 brouard 2870:
1.126 brouard 2871: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2872:
1.235 brouard 2873: 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 2874: {
1.338 brouard 2875: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij . Nicely done
1.279 brouard 2876: * (and selected quantitative values in nres)
2877: * by left multiplying the unit
2878: * matrix by transitions matrix until convergence is reached with precision ftolpl
2879: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2880: * Wx is row vector: population in state 1, population in state 2, population dead
2881: * or prevalence in state 1, prevalence in state 2, 0
2882: * newm is the matrix after multiplications, its rows are identical at a factor.
2883: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2884: * Output is prlim.
2885: * Initial matrix pimij
2886: */
1.206 brouard 2887: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2888: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2889: /* 0, 0 , 1} */
2890: /*
2891: * and after some iteration: */
2892: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2893: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2894: /* 0, 0 , 1} */
2895: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2896: /* {0.51571254859325999, 0.4842874514067399, */
2897: /* 0.51326036147820708, 0.48673963852179264} */
2898: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2899:
1.332 brouard 2900: int i, ii,j,k, k1;
1.209 brouard 2901: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2902: /* double **matprod2(); */ /* test */
1.218 brouard 2903: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2904: double **newm;
1.209 brouard 2905: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2906: int ncvloop=0;
1.288 brouard 2907: int first=0;
1.169 brouard 2908:
1.209 brouard 2909: min=vector(1,nlstate);
2910: max=vector(1,nlstate);
2911: meandiff=vector(1,nlstate);
2912:
1.218 brouard 2913: /* Starting with matrix unity */
1.126 brouard 2914: for (ii=1;ii<=nlstate+ndeath;ii++)
2915: for (j=1;j<=nlstate+ndeath;j++){
2916: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2917: }
1.169 brouard 2918:
2919: cov[1]=1.;
2920:
2921: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2922: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2923: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2924: ncvloop++;
1.126 brouard 2925: newm=savm;
2926: /* Covariates have to be included here again */
1.138 brouard 2927: cov[2]=agefin;
1.319 brouard 2928: if(nagesqr==1){
2929: cov[3]= agefin*agefin;
2930: }
1.332 brouard 2931: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
2932: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
2933: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
2934: if(Typevar[k1]==1){ /* A product with age */
2935: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
2936: }else{
2937: cov[2+nagesqr+k1]=precov[nres][k1];
2938: }
2939: }/* End of loop on model equation */
2940:
2941: /* Start of old code (replaced by a loop on position in the model equation */
2942: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only of the model *\/ */
2943: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
2944: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; *\/ */
2945: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])]; */
2946: /* /\* model = 1 +age + V1*V3 + age*V1 + V2 + V1 + age*V2 + V3 + V3*age + V1*V2 */
2947: /* * k 1 2 3 4 5 6 7 8 */
2948: /* *cov[] 1 2 3 4 5 6 7 8 9 10 */
2949: /* *TypeVar[k] 2 1 0 0 1 0 1 2 */
2950: /* *Dummy[k] 0 2 0 0 2 0 2 0 */
2951: /* *Tvar[k] 4 1 2 1 2 3 3 5 */
2952: /* *nsd=3 (1) (2) (3) */
2953: /* *TvarsD[nsd] [1]=2 1 3 */
2954: /* *TnsdVar [2]=2 [1]=1 [3]=3 */
2955: /* *TvarsDind[nsd](=k) [1]=3 [2]=4 [3]=6 */
2956: /* *Tage[] [1]=1 [2]=2 [3]=3 */
2957: /* *Tvard[] [1][1]=1 [2][1]=1 */
2958: /* * [1][2]=3 [2][2]=2 */
2959: /* *Tprod[](=k) [1]=1 [2]=8 */
2960: /* *TvarsDp(=Tvar) [1]=1 [2]=2 [3]=3 [4]=5 */
2961: /* *TvarD (=k) [1]=1 [2]=3 [3]=4 [3]=6 [4]=6 */
2962: /* *TvarsDpType */
2963: /* *si model= 1 + age + V3 + V2*age + V2 + V3*age */
2964: /* * nsd=1 (1) (2) */
2965: /* *TvarsD[nsd] 3 2 */
2966: /* *TnsdVar (3)=1 (2)=2 */
2967: /* *TvarsDind[nsd](=k) [1]=1 [2]=3 */
2968: /* *Tage[] [1]=2 [2]= 3 */
2969: /* *\/ */
2970: /* /\* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; *\/ */
2971: /* /\* 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)); *\/ */
2972: /* } */
2973: /* for (k=1; k<=nsq;k++) { /\* For single quantitative varying covariates only of the model *\/ */
2974: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
2975: /* /\* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline *\/ */
2976: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
2977: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][resultmodel[nres][k1]] */
2978: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2979: /* /\* 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]); *\/ */
2980: /* } */
2981: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
2982: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
2983: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
2984: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
2985: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
2986: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
2987: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2988: /* } */
2989: /* /\* 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]); *\/ */
2990: /* } */
2991: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
2992: /* /\* 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]); *\/ */
2993: /* if(Dummy[Tvard[k][1]]==0){ */
2994: /* if(Dummy[Tvard[k][2]]==0){ */
2995: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
2996: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
2997: /* }else{ */
2998: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
2999: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
3000: /* } */
3001: /* }else{ */
3002: /* if(Dummy[Tvard[k][2]]==0){ */
3003: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3004: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
3005: /* }else{ */
3006: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3007: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
3008: /* } */
3009: /* } */
3010: /* } /\* End product without age *\/ */
3011: /* ENd of old code */
1.138 brouard 3012: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3013: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3014: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 3015: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3016: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 3017: /* age and covariate values of ij are in 'cov' */
1.142 brouard 3018: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 3019:
1.126 brouard 3020: savm=oldm;
3021: oldm=newm;
1.209 brouard 3022:
3023: for(j=1; j<=nlstate; j++){
3024: max[j]=0.;
3025: min[j]=1.;
3026: }
3027: for(i=1;i<=nlstate;i++){
3028: sumnew=0;
3029: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
3030: for(j=1; j<=nlstate; j++){
3031: prlim[i][j]= newm[i][j]/(1-sumnew);
3032: max[j]=FMAX(max[j],prlim[i][j]);
3033: min[j]=FMIN(min[j],prlim[i][j]);
3034: }
3035: }
3036:
1.126 brouard 3037: maxmax=0.;
1.209 brouard 3038: for(j=1; j<=nlstate; j++){
3039: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
3040: maxmax=FMAX(maxmax,meandiff[j]);
3041: /* 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 3042: } /* j loop */
1.203 brouard 3043: *ncvyear= (int)age- (int)agefin;
1.208 brouard 3044: /* 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 3045: if(maxmax < ftolpl){
1.209 brouard 3046: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
3047: free_vector(min,1,nlstate);
3048: free_vector(max,1,nlstate);
3049: free_vector(meandiff,1,nlstate);
1.126 brouard 3050: return prlim;
3051: }
1.288 brouard 3052: } /* agefin loop */
1.208 brouard 3053: /* After some age loop it doesn't converge */
1.288 brouard 3054: if(!first){
3055: first=1;
3056: 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 3057: 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);
3058: }else if (first >=1 && first <10){
3059: 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);
3060: first++;
3061: }else if (first ==10){
3062: 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);
3063: 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");
3064: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
3065: first++;
1.288 brouard 3066: }
3067:
1.209 brouard 3068: /* 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); */
3069: free_vector(min,1,nlstate);
3070: free_vector(max,1,nlstate);
3071: free_vector(meandiff,1,nlstate);
1.208 brouard 3072:
1.169 brouard 3073: return prlim; /* should not reach here */
1.126 brouard 3074: }
3075:
1.217 brouard 3076:
3077: /**** Back Prevalence limit (stable or period prevalence) ****************/
3078:
1.218 brouard 3079: /* 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) */
3080: /* 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 3081: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 3082: {
1.264 brouard 3083: /* 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 3084: matrix by transitions matrix until convergence is reached with precision ftolpl */
3085: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
3086: /* Wx is row vector: population in state 1, population in state 2, population dead */
3087: /* or prevalence in state 1, prevalence in state 2, 0 */
3088: /* newm is the matrix after multiplications, its rows are identical at a factor */
3089: /* Initial matrix pimij */
3090: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
3091: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
3092: /* 0, 0 , 1} */
3093: /*
3094: * and after some iteration: */
3095: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
3096: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
3097: /* 0, 0 , 1} */
3098: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
3099: /* {0.51571254859325999, 0.4842874514067399, */
3100: /* 0.51326036147820708, 0.48673963852179264} */
3101: /* If we start from prlim again, prlim tends to a constant matrix */
3102:
1.332 brouard 3103: int i, ii,j,k, k1;
1.247 brouard 3104: int first=0;
1.217 brouard 3105: double *min, *max, *meandiff, maxmax,sumnew=0.;
3106: /* double **matprod2(); */ /* test */
3107: double **out, cov[NCOVMAX+1], **bmij();
3108: double **newm;
1.218 brouard 3109: double **dnewm, **doldm, **dsavm; /* for use */
3110: double **oldm, **savm; /* for use */
3111:
1.217 brouard 3112: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
3113: int ncvloop=0;
3114:
3115: min=vector(1,nlstate);
3116: max=vector(1,nlstate);
3117: meandiff=vector(1,nlstate);
3118:
1.266 brouard 3119: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
3120: oldm=oldms; savm=savms;
3121:
3122: /* Starting with matrix unity */
3123: for (ii=1;ii<=nlstate+ndeath;ii++)
3124: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 3125: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3126: }
3127:
3128: cov[1]=1.;
3129:
3130: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3131: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 3132: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 3133: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
3134: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 3135: ncvloop++;
1.218 brouard 3136: newm=savm; /* oldm should be kept from previous iteration or unity at start */
3137: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 3138: /* Covariates have to be included here again */
3139: cov[2]=agefin;
1.319 brouard 3140: if(nagesqr==1){
1.217 brouard 3141: cov[3]= agefin*agefin;;
1.319 brouard 3142: }
1.332 brouard 3143: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3144: if(Typevar[k1]==1){ /* A product with age */
3145: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.242 brouard 3146: }else{
1.332 brouard 3147: cov[2+nagesqr+k1]=precov[nres][k1];
1.242 brouard 3148: }
1.332 brouard 3149: }/* End of loop on model equation */
3150:
3151: /* Old code */
3152:
3153: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
3154: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
3155: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; */
3156: /* /\* 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)); *\/ */
3157: /* } */
3158: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
3159: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3160: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3161: /* /\* /\\* 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])]); *\\/ *\/ */
3162: /* /\* } *\/ */
3163: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3164: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3165: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3166: /* /\* 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]); *\/ */
3167: /* } */
3168: /* /\* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; *\/ */
3169: /* /\* for (k=1; k<=cptcovprod;k++) /\\* Useless *\\/ *\/ */
3170: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\\/ *\/ */
3171: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3172: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
3173: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ ERROR ???*\/ */
3174: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3175: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3176: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3177: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3178: /* } */
3179: /* /\* 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]); *\/ */
3180: /* } */
3181: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
3182: /* /\* 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]); *\/ */
3183: /* if(Dummy[Tvard[k][1]]==0){ */
3184: /* if(Dummy[Tvard[k][2]]==0){ */
3185: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3186: /* }else{ */
3187: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3188: /* } */
3189: /* }else{ */
3190: /* if(Dummy[Tvard[k][2]]==0){ */
3191: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3192: /* }else{ */
3193: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3194: /* } */
3195: /* } */
3196: /* } */
1.217 brouard 3197:
3198: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3199: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3200: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3201: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3202: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3203: /* ij should be linked to the correct index of cov */
3204: /* age and covariate values ij are in 'cov', but we need to pass
3205: * ij for the observed prevalence at age and status and covariate
3206: * number: prevacurrent[(int)agefin][ii][ij]
3207: */
3208: /* 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 *\/ */
3209: /* 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 *\/ */
3210: 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 3211: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3212: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3213: /* for(i=1; i<=nlstate+ndeath; i++) { */
3214: /* printf("%d newm= ",i); */
3215: /* for(j=1;j<=nlstate+ndeath;j++) { */
3216: /* printf("%f ",newm[i][j]); */
3217: /* } */
3218: /* printf("oldm * "); */
3219: /* for(j=1;j<=nlstate+ndeath;j++) { */
3220: /* printf("%f ",oldm[i][j]); */
3221: /* } */
1.268 brouard 3222: /* printf(" bmmij "); */
1.266 brouard 3223: /* for(j=1;j<=nlstate+ndeath;j++) { */
3224: /* printf("%f ",pmmij[i][j]); */
3225: /* } */
3226: /* printf("\n"); */
3227: /* } */
3228: /* } */
1.217 brouard 3229: savm=oldm;
3230: oldm=newm;
1.266 brouard 3231:
1.217 brouard 3232: for(j=1; j<=nlstate; j++){
3233: max[j]=0.;
3234: min[j]=1.;
3235: }
3236: for(j=1; j<=nlstate; j++){
3237: for(i=1;i<=nlstate;i++){
1.234 brouard 3238: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3239: bprlim[i][j]= newm[i][j];
3240: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3241: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3242: }
3243: }
1.218 brouard 3244:
1.217 brouard 3245: maxmax=0.;
3246: for(i=1; i<=nlstate; i++){
1.318 brouard 3247: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3248: maxmax=FMAX(maxmax,meandiff[i]);
3249: /* 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 3250: } /* i loop */
1.217 brouard 3251: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3252: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3253: if(maxmax < ftolpl){
1.220 brouard 3254: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3255: free_vector(min,1,nlstate);
3256: free_vector(max,1,nlstate);
3257: free_vector(meandiff,1,nlstate);
3258: return bprlim;
3259: }
1.288 brouard 3260: } /* agefin loop */
1.217 brouard 3261: /* After some age loop it doesn't converge */
1.288 brouard 3262: if(!first){
1.247 brouard 3263: first=1;
3264: 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\
3265: 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);
3266: }
3267: 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 3268: 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);
3269: /* 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); */
3270: free_vector(min,1,nlstate);
3271: free_vector(max,1,nlstate);
3272: free_vector(meandiff,1,nlstate);
3273:
3274: return bprlim; /* should not reach here */
3275: }
3276:
1.126 brouard 3277: /*************** transition probabilities ***************/
3278:
3279: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3280: {
1.138 brouard 3281: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3282: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3283: model to the ncovmodel covariates (including constant and age).
3284: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3285: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3286: ncth covariate in the global vector x is given by the formula:
3287: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3288: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3289: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3290: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3291: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3292: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3293: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3294: */
3295: double s1, lnpijopii;
1.126 brouard 3296: /*double t34;*/
1.164 brouard 3297: int i,j, nc, ii, jj;
1.126 brouard 3298:
1.223 brouard 3299: for(i=1; i<= nlstate; i++){
3300: for(j=1; j<i;j++){
3301: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3302: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3303: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3304: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3305: }
3306: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3307: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3308: }
3309: for(j=i+1; j<=nlstate+ndeath;j++){
3310: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3311: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3312: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3313: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3314: }
3315: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3316: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3317: }
3318: }
1.218 brouard 3319:
1.223 brouard 3320: for(i=1; i<= nlstate; i++){
3321: s1=0;
3322: for(j=1; j<i; j++){
1.339 brouard 3323: /* printf("debug1 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 3324: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3325: }
3326: for(j=i+1; j<=nlstate+ndeath; j++){
1.339 brouard 3327: /* printf("debug2 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 3328: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3329: }
3330: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3331: ps[i][i]=1./(s1+1.);
3332: /* Computing other pijs */
3333: for(j=1; j<i; j++)
1.325 brouard 3334: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3335: for(j=i+1; j<=nlstate+ndeath; j++)
3336: ps[i][j]= exp(ps[i][j])*ps[i][i];
3337: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3338: } /* end i */
1.218 brouard 3339:
1.223 brouard 3340: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3341: for(jj=1; jj<= nlstate+ndeath; jj++){
3342: ps[ii][jj]=0;
3343: ps[ii][ii]=1;
3344: }
3345: }
1.294 brouard 3346:
3347:
1.223 brouard 3348: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3349: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3350: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3351: /* } */
3352: /* printf("\n "); */
3353: /* } */
3354: /* printf("\n ");printf("%lf ",cov[2]);*/
3355: /*
3356: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3357: goto end;*/
1.266 brouard 3358: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3359: }
3360:
1.218 brouard 3361: /*************** backward transition probabilities ***************/
3362:
3363: /* 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 ) */
3364: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3365: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3366: {
1.302 brouard 3367: /* 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 3368: * 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 3369: */
1.218 brouard 3370: int i, ii, j,k;
1.222 brouard 3371:
3372: double **out, **pmij();
3373: double sumnew=0.;
1.218 brouard 3374: double agefin;
1.292 brouard 3375: 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 3376: double **dnewm, **dsavm, **doldm;
3377: double **bbmij;
3378:
1.218 brouard 3379: doldm=ddoldms; /* global pointers */
1.222 brouard 3380: dnewm=ddnewms;
3381: dsavm=ddsavms;
1.318 brouard 3382:
3383: /* Debug */
3384: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3385: agefin=cov[2];
1.268 brouard 3386: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3387: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3388: the observed prevalence (with this covariate ij) at beginning of transition */
3389: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3390:
3391: /* P_x */
1.325 brouard 3392: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3393: /* outputs pmmij which is a stochastic matrix in row */
3394:
3395: /* Diag(w_x) */
1.292 brouard 3396: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3397: sumnew=0.;
1.269 brouard 3398: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3399: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3400: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3401: sumnew+=prevacurrent[(int)agefin][ii][ij];
3402: }
3403: if(sumnew >0.01){ /* At least some value in the prevalence */
3404: for (ii=1;ii<=nlstate+ndeath;ii++){
3405: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3406: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3407: }
3408: }else{
3409: for (ii=1;ii<=nlstate+ndeath;ii++){
3410: for (j=1;j<=nlstate+ndeath;j++)
3411: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3412: }
3413: /* if(sumnew <0.9){ */
3414: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3415: /* } */
3416: }
3417: k3=0.0; /* We put the last diagonal to 0 */
3418: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3419: doldm[ii][ii]= k3;
3420: }
3421: /* End doldm, At the end doldm is diag[(w_i)] */
3422:
1.292 brouard 3423: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3424: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3425:
1.292 brouard 3426: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3427: /* 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 3428: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3429: sumnew=0.;
1.222 brouard 3430: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3431: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3432: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3433: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3434: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3435: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3436: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3437: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3438: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3439: /* }else */
1.268 brouard 3440: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3441: } /*End ii */
3442: } /* 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 */
3443:
1.292 brouard 3444: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3445: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3446: /* end bmij */
1.266 brouard 3447: return ps; /*pointer is unchanged */
1.218 brouard 3448: }
1.217 brouard 3449: /*************** transition probabilities ***************/
3450:
1.218 brouard 3451: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3452: {
3453: /* According to parameters values stored in x and the covariate's values stored in cov,
3454: computes the probability to be observed in state j being in state i by appying the
3455: model to the ncovmodel covariates (including constant and age).
3456: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3457: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3458: ncth covariate in the global vector x is given by the formula:
3459: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3460: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3461: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3462: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3463: Outputs ps[i][j] the probability to be observed in j being in j according to
3464: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3465: */
3466: double s1, lnpijopii;
3467: /*double t34;*/
3468: int i,j, nc, ii, jj;
3469:
1.234 brouard 3470: for(i=1; i<= nlstate; i++){
3471: for(j=1; j<i;j++){
3472: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3473: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3474: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3475: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3476: }
3477: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3478: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3479: }
3480: for(j=i+1; j<=nlstate+ndeath;j++){
3481: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3482: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3483: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3484: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3485: }
3486: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3487: }
3488: }
3489:
3490: for(i=1; i<= nlstate; i++){
3491: s1=0;
3492: for(j=1; j<i; j++){
3493: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3494: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3495: }
3496: for(j=i+1; j<=nlstate+ndeath; j++){
3497: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3498: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3499: }
3500: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3501: ps[i][i]=1./(s1+1.);
3502: /* Computing other pijs */
3503: for(j=1; j<i; j++)
3504: ps[i][j]= exp(ps[i][j])*ps[i][i];
3505: for(j=i+1; j<=nlstate+ndeath; j++)
3506: ps[i][j]= exp(ps[i][j])*ps[i][i];
3507: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3508: } /* end i */
3509:
3510: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3511: for(jj=1; jj<= nlstate+ndeath; jj++){
3512: ps[ii][jj]=0;
3513: ps[ii][ii]=1;
3514: }
3515: }
1.296 brouard 3516: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3517: for(jj=1; jj<= nlstate+ndeath; jj++){
3518: s1=0.;
3519: for(ii=1; ii<= nlstate+ndeath; ii++){
3520: s1+=ps[ii][jj];
3521: }
3522: for(ii=1; ii<= nlstate; ii++){
3523: ps[ii][jj]=ps[ii][jj]/s1;
3524: }
3525: }
3526: /* Transposition */
3527: for(jj=1; jj<= nlstate+ndeath; jj++){
3528: for(ii=jj; ii<= nlstate+ndeath; ii++){
3529: s1=ps[ii][jj];
3530: ps[ii][jj]=ps[jj][ii];
3531: ps[jj][ii]=s1;
3532: }
3533: }
3534: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3535: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3536: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3537: /* } */
3538: /* printf("\n "); */
3539: /* } */
3540: /* printf("\n ");printf("%lf ",cov[2]);*/
3541: /*
3542: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3543: goto end;*/
3544: return ps;
1.217 brouard 3545: }
3546:
3547:
1.126 brouard 3548: /**************** Product of 2 matrices ******************/
3549:
1.145 brouard 3550: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3551: {
3552: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3553: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3554: /* in, b, out are matrice of pointers which should have been initialized
3555: before: only the contents of out is modified. The function returns
3556: a pointer to pointers identical to out */
1.145 brouard 3557: int i, j, k;
1.126 brouard 3558: for(i=nrl; i<= nrh; i++)
1.145 brouard 3559: for(k=ncolol; k<=ncoloh; k++){
3560: out[i][k]=0.;
3561: for(j=ncl; j<=nch; j++)
3562: out[i][k] +=in[i][j]*b[j][k];
3563: }
1.126 brouard 3564: return out;
3565: }
3566:
3567:
3568: /************* Higher Matrix Product ***************/
3569:
1.235 brouard 3570: 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 3571: {
1.336 brouard 3572: /* Already optimized with precov.
3573: 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 3574: 'nhstepm*hstepm*stepm' months (i.e. until
3575: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3576: nhstepm*hstepm matrices.
3577: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3578: (typically every 2 years instead of every month which is too big
3579: for the memory).
3580: Model is determined by parameters x and covariates have to be
3581: included manually here.
3582:
3583: */
3584:
1.330 brouard 3585: int i, j, d, h, k, k1;
1.131 brouard 3586: double **out, cov[NCOVMAX+1];
1.126 brouard 3587: double **newm;
1.187 brouard 3588: double agexact;
1.214 brouard 3589: double agebegin, ageend;
1.126 brouard 3590:
3591: /* Hstepm could be zero and should return the unit matrix */
3592: for (i=1;i<=nlstate+ndeath;i++)
3593: for (j=1;j<=nlstate+ndeath;j++){
3594: oldm[i][j]=(i==j ? 1.0 : 0.0);
3595: po[i][j][0]=(i==j ? 1.0 : 0.0);
3596: }
3597: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3598: for(h=1; h <=nhstepm; h++){
3599: for(d=1; d <=hstepm; d++){
3600: newm=savm;
3601: /* Covariates have to be included here again */
3602: cov[1]=1.;
1.214 brouard 3603: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3604: cov[2]=agexact;
1.319 brouard 3605: if(nagesqr==1){
1.227 brouard 3606: cov[3]= agexact*agexact;
1.319 brouard 3607: }
1.330 brouard 3608: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
3609: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
3610: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.332 brouard 3611: if(Typevar[k1]==1){ /* A product with age */
3612: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
3613: }else{
3614: cov[2+nagesqr+k1]=precov[nres][k1];
3615: }
3616: }/* End of loop on model equation */
3617: /* Old code */
3618: /* if( Dummy[k1]==0 && Typevar[k1]==0 ){ /\* Single dummy *\/ */
3619: /* /\* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) *\/ */
3620: /* /\* for (k=1; k<=nsd;k++) { /\\* For single dummy covariates only *\\/ *\/ */
3621: /* /\* /\\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\\/ *\/ */
3622: /* /\* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 *\/ */
3623: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3624: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3625: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
3626: /* /\* nsd 1 2 3 *\/ /\* Counting single dummies covar fixed or tv *\/ */
3627: /* /\*TvarsD[nsd] 4 3 1 *\/ /\* ID of single dummy cova fixed or timevary*\/ */
3628: /* /\*TvarsDind[k] 2 3 9 *\/ /\* position K of single dummy cova *\/ */
3629: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] *\/ */
3630: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3631: /* /\* 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]])); *\/ */
3632: /* 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); */
3633: /* printf("hpxij new Dummy precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3634: /* }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /\* Single quantitative variables *\/ */
3635: /* /\* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline *\/ */
3636: /* cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]]; */
3637: /* /\* for (k=1; k<=nsq;k++) { /\\* For single varying covariates only *\\/ *\/ */
3638: /* /\* /\\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\\/ *\/ */
3639: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
3640: /* 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]]); */
3641: /* printf("hpxij new Quanti precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3642: /* }else if( Dummy[k1]==2 ){ /\* For dummy with age product *\/ */
3643: /* /\* Tvar[k1] Variable in the age product age*V1 is 1 *\/ */
3644: /* /\* [Tinvresult[nres][V1] is its value in the resultline nres *\/ */
3645: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvar[k1]]*cov[2]; */
3646: /* 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]); */
3647: /* printf("hpxij new Dummy with age product precov[nres=%d][k1=%d]=%.4f * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3648:
3649: /* /\* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; *\/ */
3650: /* /\* for (k=1; k<=cptcovage;k++){ /\\* For product with age V1+V1*age +V4 +age*V3 *\\/ *\/ */
3651: /* /\* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*\/ */
3652: /* /\* *\/ */
1.330 brouard 3653: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3654: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3655: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
1.332 brouard 3656: /* /\*cptcovage=2 1 2 *\/ */
3657: /* /\*Tage[k]= 5 8 *\/ */
3658: /* }else if( Dummy[k1]==3 ){ /\* For quant with age product *\/ */
3659: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3660: /* 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]]); */
3661: /* printf("hpxij new Quanti with age product precov[nres=%d][k1=%d] * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3662: /* /\* if(Dummy[Tage[k]]== 2){ /\\* dummy with age *\\/ *\/ */
3663: /* /\* /\\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\\* dummy with age *\\\/ *\\/ *\/ */
3664: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\\/ *\/ */
3665: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\\/ *\/ */
3666: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
3667: /* /\* 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); *\/ */
3668: /* /\* } else if(Dummy[Tage[k]]== 3){ /\\* quantitative with age *\\/ *\/ */
3669: /* /\* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; *\/ */
3670: /* /\* } *\/ */
3671: /* /\* 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]); *\/ */
3672: /* }else if(Typevar[k1]==2 ){ /\* For product (not with age) *\/ */
3673: /* /\* for (k=1; k<=cptcovprod;k++){ /\\* For product without age *\\/ *\/ */
3674: /* /\* /\\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\\/ *\/ */
3675: /* /\* /\\* k 1 2 3 4 5 6 7 8 9 *\\/ *\/ */
3676: /* /\* /\\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\\/ *\/ */
3677: /* /\* /\\*cptcovprod=1 1 2 *\\/ *\/ */
3678: /* /\* /\\*Tprod[]= 4 7 *\\/ *\/ */
3679: /* /\* /\\*Tvard[][1] 4 1 *\\/ *\/ */
3680: /* /\* /\\*Tvard[][2] 3 2 *\\/ *\/ */
1.330 brouard 3681:
1.332 brouard 3682: /* /\* 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])]); *\/ */
3683: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3684: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]]; */
3685: /* 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]]); */
3686: /* printf("hpxij new Product no age product precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3687:
3688: /* /\* if(Dummy[Tvardk[k1][1]]==0){ *\/ */
3689: /* /\* if(Dummy[Tvardk[k1][2]]==0){ /\\* Product of dummies *\\/ *\/ */
3690: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3691: /* /\* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; *\/ */
3692: /* /\* 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]])]; *\/ */
3693: /* /\* }else{ /\\* Product of dummy by quantitative *\\/ *\/ */
3694: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; *\/ */
3695: /* /\* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; *\/ */
3696: /* /\* } *\/ */
3697: /* /\* }else{ /\\* Product of quantitative by...*\\/ *\/ */
3698: /* /\* if(Dummy[Tvard[k][2]]==0){ /\\* quant by dummy *\\/ *\/ */
3699: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\\/ *\/ */
3700: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3701: /* /\* }else{ /\\* Product of two quant *\\/ *\/ */
3702: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\\/ *\/ */
3703: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3704: /* /\* } *\/ */
3705: /* /\* }/\\*end of products quantitative *\\/ *\/ */
3706: /* }/\*end of products *\/ */
3707: /* } /\* End of loop on model equation *\/ */
1.235 brouard 3708: /* for (k=1; k<=cptcovn;k++) */
3709: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3710: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3711: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3712: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3713: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3714:
3715:
1.126 brouard 3716: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3717: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3718: /* right multiplication of oldm by the current matrix */
1.126 brouard 3719: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3720: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3721: /* if((int)age == 70){ */
3722: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3723: /* for(i=1; i<=nlstate+ndeath; i++) { */
3724: /* printf("%d pmmij ",i); */
3725: /* for(j=1;j<=nlstate+ndeath;j++) { */
3726: /* printf("%f ",pmmij[i][j]); */
3727: /* } */
3728: /* printf(" oldm "); */
3729: /* for(j=1;j<=nlstate+ndeath;j++) { */
3730: /* printf("%f ",oldm[i][j]); */
3731: /* } */
3732: /* printf("\n"); */
3733: /* } */
3734: /* } */
1.126 brouard 3735: savm=oldm;
3736: oldm=newm;
3737: }
3738: for(i=1; i<=nlstate+ndeath; i++)
3739: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3740: po[i][j][h]=newm[i][j];
3741: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3742: }
1.128 brouard 3743: /*printf("h=%d ",h);*/
1.126 brouard 3744: } /* end h */
1.267 brouard 3745: /* printf("\n H=%d \n",h); */
1.126 brouard 3746: return po;
3747: }
3748:
1.217 brouard 3749: /************* Higher Back Matrix Product ***************/
1.218 brouard 3750: /* 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 3751: 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 3752: {
1.332 brouard 3753: /* For dummy covariates given in each resultline (for historical, computes the corresponding combination ij),
3754: computes the transition matrix starting at age 'age' over
1.217 brouard 3755: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3756: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3757: nhstepm*hstepm matrices.
3758: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3759: (typically every 2 years instead of every month which is too big
1.217 brouard 3760: for the memory).
1.218 brouard 3761: Model is determined by parameters x and covariates have to be
1.266 brouard 3762: included manually here. Then we use a call to bmij(x and cov)
3763: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3764: */
1.217 brouard 3765:
1.332 brouard 3766: int i, j, d, h, k, k1;
1.266 brouard 3767: double **out, cov[NCOVMAX+1], **bmij();
3768: double **newm, ***newmm;
1.217 brouard 3769: double agexact;
3770: double agebegin, ageend;
1.222 brouard 3771: double **oldm, **savm;
1.217 brouard 3772:
1.266 brouard 3773: newmm=po; /* To be saved */
3774: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3775: /* Hstepm could be zero and should return the unit matrix */
3776: for (i=1;i<=nlstate+ndeath;i++)
3777: for (j=1;j<=nlstate+ndeath;j++){
3778: oldm[i][j]=(i==j ? 1.0 : 0.0);
3779: po[i][j][0]=(i==j ? 1.0 : 0.0);
3780: }
3781: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3782: for(h=1; h <=nhstepm; h++){
3783: for(d=1; d <=hstepm; d++){
3784: newm=savm;
3785: /* Covariates have to be included here again */
3786: cov[1]=1.;
1.271 brouard 3787: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3788: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3789: /* Debug */
3790: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3791: cov[2]=agexact;
1.332 brouard 3792: if(nagesqr==1){
1.222 brouard 3793: cov[3]= agexact*agexact;
1.332 brouard 3794: }
3795: /** New code */
3796: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3797: if(Typevar[k1]==1){ /* A product with age */
3798: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.325 brouard 3799: }else{
1.332 brouard 3800: cov[2+nagesqr+k1]=precov[nres][k1];
1.325 brouard 3801: }
1.332 brouard 3802: }/* End of loop on model equation */
3803: /** End of new code */
3804: /** This was old code */
3805: /* for (k=1; k<=nsd;k++){ /\* For single dummy covariates only *\//\* cptcovn error *\/ */
3806: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3807: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3808: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/\* Bug valgrind *\/ */
3809: /* /\* 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)); *\/ */
3810: /* } */
3811: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3812: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3813: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3814: /* /\* 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]); *\/ */
3815: /* } */
3816: /* for (k=1; k<=cptcovage;k++){ /\* Should start at cptcovn+1 *\//\* For product with age *\/ */
3817: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age error!!!*\\/ *\/ */
3818: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3819: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3820: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3821: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3822: /* } */
3823: /* /\* 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]); *\/ */
3824: /* } */
3825: /* for (k=1; k<=cptcovprod;k++){ /\* Useless because included in cptcovn *\/ */
3826: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3827: /* if(Dummy[Tvard[k][1]]==0){ */
3828: /* if(Dummy[Tvard[k][2]]==0){ */
3829: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]; */
3830: /* }else{ */
3831: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3832: /* } */
3833: /* }else{ */
3834: /* if(Dummy[Tvard[k][2]]==0){ */
3835: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3836: /* }else{ */
3837: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3838: /* } */
3839: /* } */
3840: /* } */
3841: /* /\*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*\/ */
3842: /* /\*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*\/ */
3843: /** End of old code */
3844:
1.218 brouard 3845: /* Careful transposed matrix */
1.266 brouard 3846: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3847: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3848: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3849: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3850: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3851: /* if((int)age == 70){ */
3852: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3853: /* for(i=1; i<=nlstate+ndeath; i++) { */
3854: /* printf("%d pmmij ",i); */
3855: /* for(j=1;j<=nlstate+ndeath;j++) { */
3856: /* printf("%f ",pmmij[i][j]); */
3857: /* } */
3858: /* printf(" oldm "); */
3859: /* for(j=1;j<=nlstate+ndeath;j++) { */
3860: /* printf("%f ",oldm[i][j]); */
3861: /* } */
3862: /* printf("\n"); */
3863: /* } */
3864: /* } */
3865: savm=oldm;
3866: oldm=newm;
3867: }
3868: for(i=1; i<=nlstate+ndeath; i++)
3869: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3870: po[i][j][h]=newm[i][j];
1.268 brouard 3871: /* if(h==nhstepm) */
3872: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3873: }
1.268 brouard 3874: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3875: } /* end h */
1.268 brouard 3876: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3877: return po;
3878: }
3879:
3880:
1.162 brouard 3881: #ifdef NLOPT
3882: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3883: double fret;
3884: double *xt;
3885: int j;
3886: myfunc_data *d2 = (myfunc_data *) pd;
3887: /* xt = (p1-1); */
3888: xt=vector(1,n);
3889: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3890:
3891: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3892: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3893: printf("Function = %.12lf ",fret);
3894: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3895: printf("\n");
3896: free_vector(xt,1,n);
3897: return fret;
3898: }
3899: #endif
1.126 brouard 3900:
3901: /*************** log-likelihood *************/
3902: double func( double *x)
3903: {
1.336 brouard 3904: int i, ii, j, k, mi, d, kk, kf=0;
1.226 brouard 3905: int ioffset=0;
1.339 brouard 3906: int ipos=0,iposold=0,ncovv=0;
3907:
1.340 brouard 3908: double cotvarv, cotvarvold;
1.226 brouard 3909: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3910: double **out;
3911: double lli; /* Individual log likelihood */
3912: int s1, s2;
1.228 brouard 3913: 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 3914:
1.226 brouard 3915: double bbh, survp;
3916: double agexact;
1.336 brouard 3917: double agebegin, ageend;
1.226 brouard 3918: /*extern weight */
3919: /* We are differentiating ll according to initial status */
3920: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3921: /*for(i=1;i<imx;i++)
3922: printf(" %d\n",s[4][i]);
3923: */
1.162 brouard 3924:
1.226 brouard 3925: ++countcallfunc;
1.162 brouard 3926:
1.226 brouard 3927: cov[1]=1.;
1.126 brouard 3928:
1.226 brouard 3929: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3930: ioffset=0;
1.226 brouard 3931: if(mle==1){
3932: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3933: /* Computes the values of the ncovmodel covariates of the model
3934: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3935: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3936: to be observed in j being in i according to the model.
3937: */
1.243 brouard 3938: ioffset=2+nagesqr ;
1.233 brouard 3939: /* Fixed */
1.336 brouard 3940: for (kf=1; kf<=ncovf;kf++){ /* For each fixed covariate dummu or quant or prod */
1.319 brouard 3941: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3942: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3943: /* 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 3944: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.336 brouard 3945: 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 3946: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3947: }
1.226 brouard 3948: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3949: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3950: has been calculated etc */
3951: /* For an individual i, wav[i] gives the number of effective waves */
3952: /* We compute the contribution to Likelihood of each effective transition
3953: mw[mi][i] is real wave of the mi th effectve wave */
3954: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3955: s2=s[mw[mi+1][i]][i];
1.341 brouard 3956: 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 3957: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3958: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3959: */
1.336 brouard 3960: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
3961: /* Wave varying (but not age varying) */
1.339 brouard 3962: /* 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*\/ */
3963: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? *\/ */
3964: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
3965: /* } */
1.340 brouard 3966: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age )*/
3967: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
3968: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
3969: if(TvarFind[itv]==0){ /* Not a fixed covariate */
1.341 brouard 3970: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* cotvar[wav][ncovcol+nqv+iv][i] */
1.340 brouard 3971: }else{ /* fixed covariate */
3972: cotvarv=covar[Tvar[TvarFind[itv]]][i];
3973: }
1.339 brouard 3974: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 3975: cotvarvold=cotvarv;
3976: }else{ /* A second product */
3977: cotvarv=cotvarv*cotvarvold;
1.339 brouard 3978: }
3979: iposold=ipos;
1.340 brouard 3980: cov[ioffset+ipos]=cotvarv;
1.234 brouard 3981: }
1.339 brouard 3982: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
3983: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3984: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3985: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3986: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3987: /* 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]); */
3988: /* } */
3989: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
3990: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3991: /* /\* 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]); *\/ */
3992: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
3993: /* } */
3994: /* for products of time varying to be done */
1.234 brouard 3995: for (ii=1;ii<=nlstate+ndeath;ii++)
3996: for (j=1;j<=nlstate+ndeath;j++){
3997: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3998: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3999: }
1.336 brouard 4000:
4001: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4002: 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 4003: for(d=0; d<dh[mi][i]; d++){
4004: newm=savm;
4005: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4006: cov[2]=agexact;
4007: if(nagesqr==1)
4008: cov[3]= agexact*agexact; /* Should be changed here */
4009: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 4010: if(!FixedV[Tvar[Tage[kk]]])
4011: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4012: else
1.341 brouard 4013: 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 4014: }
4015: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4016: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4017: savm=oldm;
4018: oldm=newm;
4019: } /* end mult */
4020:
4021: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
4022: /* But now since version 0.9 we anticipate for bias at large stepm.
4023: * If stepm is larger than one month (smallest stepm) and if the exact delay
4024: * (in months) between two waves is not a multiple of stepm, we rounded to
4025: * the nearest (and in case of equal distance, to the lowest) interval but now
4026: * we keep into memory the bias bh[mi][i] and also the previous matrix product
4027: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
4028: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 4029: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
4030: * -stepm/2 to stepm/2 .
4031: * For stepm=1 the results are the same as for previous versions of Imach.
4032: * For stepm > 1 the results are less biased than in previous versions.
4033: */
1.234 brouard 4034: s1=s[mw[mi][i]][i];
4035: s2=s[mw[mi+1][i]][i];
4036: bbh=(double)bh[mi][i]/(double)stepm;
4037: /* bias bh is positive if real duration
4038: * is higher than the multiple of stepm and negative otherwise.
4039: */
4040: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
4041: if( s2 > nlstate){
4042: /* i.e. if s2 is a death state and if the date of death is known
4043: then the contribution to the likelihood is the probability to
4044: die between last step unit time and current step unit time,
4045: which is also equal to probability to die before dh
4046: minus probability to die before dh-stepm .
4047: In version up to 0.92 likelihood was computed
4048: as if date of death was unknown. Death was treated as any other
4049: health state: the date of the interview describes the actual state
4050: and not the date of a change in health state. The former idea was
4051: to consider that at each interview the state was recorded
4052: (healthy, disable or death) and IMaCh was corrected; but when we
4053: introduced the exact date of death then we should have modified
4054: the contribution of an exact death to the likelihood. This new
4055: contribution is smaller and very dependent of the step unit
4056: stepm. It is no more the probability to die between last interview
4057: and month of death but the probability to survive from last
4058: interview up to one month before death multiplied by the
4059: probability to die within a month. Thanks to Chris
4060: Jackson for correcting this bug. Former versions increased
4061: mortality artificially. The bad side is that we add another loop
4062: which slows down the processing. The difference can be up to 10%
4063: lower mortality.
4064: */
4065: /* If, at the beginning of the maximization mostly, the
4066: cumulative probability or probability to be dead is
4067: constant (ie = 1) over time d, the difference is equal to
4068: 0. out[s1][3] = savm[s1][3]: probability, being at state
4069: s1 at precedent wave, to be dead a month before current
4070: wave is equal to probability, being at state s1 at
4071: precedent wave, to be dead at mont of the current
4072: wave. Then the observed probability (that this person died)
4073: is null according to current estimated parameter. In fact,
4074: it should be very low but not zero otherwise the log go to
4075: infinity.
4076: */
1.183 brouard 4077: /* #ifdef INFINITYORIGINAL */
4078: /* lli=log(out[s1][s2] - savm[s1][s2]); */
4079: /* #else */
4080: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
4081: /* lli=log(mytinydouble); */
4082: /* else */
4083: /* lli=log(out[s1][s2] - savm[s1][s2]); */
4084: /* #endif */
1.226 brouard 4085: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4086:
1.226 brouard 4087: } else if ( s2==-1 ) { /* alive */
4088: for (j=1,survp=0. ; j<=nlstate; j++)
4089: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4090: /*survp += out[s1][j]; */
4091: lli= log(survp);
4092: }
1.336 brouard 4093: /* else if (s2==-4) { */
4094: /* for (j=3,survp=0. ; j<=nlstate; j++) */
4095: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4096: /* lli= log(survp); */
4097: /* } */
4098: /* else if (s2==-5) { */
4099: /* for (j=1,survp=0. ; j<=2; j++) */
4100: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4101: /* lli= log(survp); */
4102: /* } */
1.226 brouard 4103: else{
4104: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4105: /* 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 */
4106: }
4107: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
4108: /*if(lli ==000.0)*/
1.340 brouard 4109: /* 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 4110: ipmx +=1;
4111: sw += weight[i];
4112: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4113: /* if (lli < log(mytinydouble)){ */
4114: /* 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); */
4115: /* 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]); */
4116: /* } */
4117: } /* end of wave */
4118: } /* end of individual */
4119: } else if(mle==2){
4120: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 4121: ioffset=2+nagesqr ;
4122: for (k=1; k<=ncovf;k++)
4123: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 4124: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 4125: for(k=1; k <= ncovv ; k++){
1.341 brouard 4126: 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 4127: }
1.226 brouard 4128: for (ii=1;ii<=nlstate+ndeath;ii++)
4129: for (j=1;j<=nlstate+ndeath;j++){
4130: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4131: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4132: }
4133: for(d=0; d<=dh[mi][i]; d++){
4134: newm=savm;
4135: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4136: cov[2]=agexact;
4137: if(nagesqr==1)
4138: cov[3]= agexact*agexact;
4139: for (kk=1; kk<=cptcovage;kk++) {
4140: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4141: }
4142: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4143: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4144: savm=oldm;
4145: oldm=newm;
4146: } /* end mult */
4147:
4148: s1=s[mw[mi][i]][i];
4149: s2=s[mw[mi+1][i]][i];
4150: bbh=(double)bh[mi][i]/(double)stepm;
4151: 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 */
4152: ipmx +=1;
4153: sw += weight[i];
4154: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4155: } /* end of wave */
4156: } /* end of individual */
4157: } else if(mle==3){ /* exponential inter-extrapolation */
4158: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4159: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4160: for(mi=1; mi<= wav[i]-1; mi++){
4161: for (ii=1;ii<=nlstate+ndeath;ii++)
4162: for (j=1;j<=nlstate+ndeath;j++){
4163: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4164: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4165: }
4166: for(d=0; d<dh[mi][i]; d++){
4167: newm=savm;
4168: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4169: cov[2]=agexact;
4170: if(nagesqr==1)
4171: cov[3]= agexact*agexact;
4172: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 4173: if(!FixedV[Tvar[Tage[kk]]])
4174: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4175: else
1.341 brouard 4176: 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 4177: }
4178: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4179: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4180: savm=oldm;
4181: oldm=newm;
4182: } /* end mult */
4183:
4184: s1=s[mw[mi][i]][i];
4185: s2=s[mw[mi+1][i]][i];
4186: bbh=(double)bh[mi][i]/(double)stepm;
4187: 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 */
4188: ipmx +=1;
4189: sw += weight[i];
4190: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4191: } /* end of wave */
4192: } /* end of individual */
4193: }else if (mle==4){ /* ml=4 no inter-extrapolation */
4194: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4195: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4196: for(mi=1; mi<= wav[i]-1; mi++){
4197: for (ii=1;ii<=nlstate+ndeath;ii++)
4198: for (j=1;j<=nlstate+ndeath;j++){
4199: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4200: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4201: }
4202: for(d=0; d<dh[mi][i]; d++){
4203: newm=savm;
4204: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4205: cov[2]=agexact;
4206: if(nagesqr==1)
4207: cov[3]= agexact*agexact;
4208: for (kk=1; kk<=cptcovage;kk++) {
4209: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4210: }
1.126 brouard 4211:
1.226 brouard 4212: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4213: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4214: savm=oldm;
4215: oldm=newm;
4216: } /* end mult */
4217:
4218: s1=s[mw[mi][i]][i];
4219: s2=s[mw[mi+1][i]][i];
4220: if( s2 > nlstate){
4221: lli=log(out[s1][s2] - savm[s1][s2]);
4222: } else if ( s2==-1 ) { /* alive */
4223: for (j=1,survp=0. ; j<=nlstate; j++)
4224: survp += out[s1][j];
4225: lli= log(survp);
4226: }else{
4227: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4228: }
4229: ipmx +=1;
4230: sw += weight[i];
4231: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.343 ! brouard 4232: /* 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 4233: } /* end of wave */
4234: } /* end of individual */
4235: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
4236: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4237: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4238: for(mi=1; mi<= wav[i]-1; mi++){
4239: for (ii=1;ii<=nlstate+ndeath;ii++)
4240: for (j=1;j<=nlstate+ndeath;j++){
4241: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4242: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4243: }
4244: for(d=0; d<dh[mi][i]; d++){
4245: newm=savm;
4246: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4247: cov[2]=agexact;
4248: if(nagesqr==1)
4249: cov[3]= agexact*agexact;
4250: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 4251: if(!FixedV[Tvar[Tage[kk]]])
4252: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4253: else
1.341 brouard 4254: 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 4255: }
1.126 brouard 4256:
1.226 brouard 4257: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4258: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4259: savm=oldm;
4260: oldm=newm;
4261: } /* end mult */
4262:
4263: s1=s[mw[mi][i]][i];
4264: s2=s[mw[mi+1][i]][i];
4265: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4266: ipmx +=1;
4267: sw += weight[i];
4268: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4269: /*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]);*/
4270: } /* end of wave */
4271: } /* end of individual */
4272: } /* End of if */
4273: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4274: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4275: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4276: return -l;
1.126 brouard 4277: }
4278:
4279: /*************** log-likelihood *************/
4280: double funcone( double *x)
4281: {
1.228 brouard 4282: /* Same as func but slower because of a lot of printf and if */
1.335 brouard 4283: int i, ii, j, k, mi, d, kk, kf=0;
1.228 brouard 4284: int ioffset=0;
1.339 brouard 4285: int ipos=0,iposold=0,ncovv=0;
4286:
1.340 brouard 4287: double cotvarv, cotvarvold;
1.131 brouard 4288: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 4289: double **out;
4290: double lli; /* Individual log likelihood */
4291: double llt;
4292: int s1, s2;
1.228 brouard 4293: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
4294:
1.126 brouard 4295: double bbh, survp;
1.187 brouard 4296: double agexact;
1.214 brouard 4297: double agebegin, ageend;
1.126 brouard 4298: /*extern weight */
4299: /* We are differentiating ll according to initial status */
4300: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4301: /*for(i=1;i<imx;i++)
4302: printf(" %d\n",s[4][i]);
4303: */
4304: cov[1]=1.;
4305:
4306: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4307: ioffset=0;
4308: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.336 brouard 4309: /* Computes the values of the ncovmodel covariates of the model
4310: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
4311: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
4312: to be observed in j being in i according to the model.
4313: */
1.243 brouard 4314: /* ioffset=2+nagesqr+cptcovage; */
4315: ioffset=2+nagesqr;
1.232 brouard 4316: /* Fixed */
1.224 brouard 4317: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4318: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.335 brouard 4319: 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 4320: /* printf("Debug3 TvarFind[%d]=%d",kf, TvarFind[kf]); */
4321: /* printf(" Tvar[TvarFind[kf]]=%d", Tvar[TvarFind[kf]]); */
4322: /* printf(" i=%d covar[Tvar[TvarFind[kf]]][i]=%f\n",i,covar[Tvar[TvarFind[kf]]][i]); */
1.335 brouard 4323: 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 4324: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4325: /* cov[2+6]=covar[Tvar[6]][i]; */
4326: /* cov[2+6]=covar[2][i]; V2 */
4327: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4328: /* cov[2+7]=covar[Tvar[7]][i]; */
4329: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4330: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4331: /* cov[2+9]=covar[Tvar[9]][i]; */
4332: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4333: }
1.336 brouard 4334: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
4335: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
4336: has been calculated etc */
4337: /* For an individual i, wav[i] gives the number of effective waves */
4338: /* We compute the contribution to Likelihood of each effective transition
4339: mw[mi][i] is real wave of the mi th effectve wave */
4340: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
4341: s2=s[mw[mi+1][i]][i];
1.341 brouard 4342: And the iv th varying covariate in the DATA is the cotvar[mw[mi+1][i]][ncovcol+nqv+iv][i]
1.336 brouard 4343: */
4344: /* This part may be useless now because everythin should be in covar */
1.232 brouard 4345: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4346: /* 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?)*\/ */
4347: /* } */
1.231 brouard 4348: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4349: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4350: /* } */
1.225 brouard 4351:
1.233 brouard 4352:
4353: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.339 brouard 4354: /* 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 */
4355: /* for(k=1; k <= ncovv ; k++){ /\* Varying covariates (single and product but no age )*\/ */
4356: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; *\/ */
4357: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
4358: /* } */
4359:
4360: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
4361: /* model V1+V3+age*V1+age*V3+V1*V3 */
4362: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
4363: /* TvarVV[1]=V3 (first time varying in the model equation, TvarVV[2]=V1 (in V1*V3) TvarVV[3]=3(V3) */
4364: /* We need the position of the time varying or product in the model */
4365: /* 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 */
4366: /* TvarVV gives the variable name */
1.340 brouard 4367: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
4368: * k= 1 2 3 4 5 6 7 8 9
4369: * varying 1 2 3 4 5
4370: * ncovv 1 2 3 4 5 6 7 8
1.343 ! brouard 4371: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
1.340 brouard 4372: * TvarVVind 2 3 7 7 8 8 9 9
4373: * TvarFind[k] 1 0 0 0 0 0 0 0 0
4374: */
4375: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
4376: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
4377: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
4378: if(TvarFind[itv]==0){ /* Not a fixed covariate */
1.341 brouard 4379: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.340 brouard 4380: }else{ /* fixed covariate */
4381: cotvarv=covar[Tvar[TvarFind[itv]]][i];
4382: }
1.339 brouard 4383: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 4384: cotvarvold=cotvarv;
4385: }else{ /* A second product */
4386: cotvarv=cotvarv*cotvarvold;
1.339 brouard 4387: }
4388: iposold=ipos;
1.340 brouard 4389: cov[ioffset+ipos]=cotvarv;
1.339 brouard 4390: /* For products */
4391: }
4392: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates single *\/ */
4393: /* iv=TvarVDind[itv]; /\* iv, position in the model equation of time varying covariate itv *\/ */
4394: /* /\* "V1+V3+age*V1+age*V3+V1*V3" with V3 time varying *\/ */
4395: /* /\* 1 2 3 4 5 *\/ */
4396: /* /\*itv 1 *\/ */
4397: /* /\* TvarVInd[1]= 2 *\/ */
4398: /* /\* iv= Tvar[Tmodelind[itv]]-ncovcol-nqv; /\\* Counting the # varying covariate from 1 to ntveff *\\/ *\/ */
4399: /* /\* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; *\/ */
4400: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; *\/ */
4401: /* /\* k=ioffset-2-nagesqr-cptcovage+itv; /\\* position in simple model *\\/ *\/ */
4402: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; *\/ */
4403: /* cov[ioffset+iv]=cotvar[mw[mi][i]][itv][i]; */
4404: /* /\* 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]); *\/ */
4405: /* } */
1.232 brouard 4406: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4407: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4408: /* /\* 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]); *\/ */
4409: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4410: /* } */
1.126 brouard 4411: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4412: for (j=1;j<=nlstate+ndeath;j++){
4413: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4414: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4415: }
1.214 brouard 4416:
4417: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4418: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4419: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4420: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4421: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4422: and mw[mi+1][i]. dh depends on stepm.*/
4423: newm=savm;
1.247 brouard 4424: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4425: cov[2]=agexact;
4426: if(nagesqr==1)
4427: cov[3]= agexact*agexact;
4428: for (kk=1; kk<=cptcovage;kk++) {
4429: if(!FixedV[Tvar[Tage[kk]]])
4430: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4431: else
1.341 brouard 4432: 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 4433: }
4434: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4435: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4436: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4437: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4438: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4439: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4440: savm=oldm;
4441: oldm=newm;
1.126 brouard 4442: } /* end mult */
1.336 brouard 4443: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
4444: /* But now since version 0.9 we anticipate for bias at large stepm.
4445: * If stepm is larger than one month (smallest stepm) and if the exact delay
4446: * (in months) between two waves is not a multiple of stepm, we rounded to
4447: * the nearest (and in case of equal distance, to the lowest) interval but now
4448: * we keep into memory the bias bh[mi][i] and also the previous matrix product
4449: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
4450: * probability in order to take into account the bias as a fraction of the way
4451: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
4452: * -stepm/2 to stepm/2 .
4453: * For stepm=1 the results are the same as for previous versions of Imach.
4454: * For stepm > 1 the results are less biased than in previous versions.
4455: */
1.126 brouard 4456: s1=s[mw[mi][i]][i];
4457: s2=s[mw[mi+1][i]][i];
1.217 brouard 4458: /* if(s2==-1){ */
1.268 brouard 4459: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4460: /* /\* exit(1); *\/ */
4461: /* } */
1.126 brouard 4462: bbh=(double)bh[mi][i]/(double)stepm;
4463: /* bias is positive if real duration
4464: * is higher than the multiple of stepm and negative otherwise.
4465: */
4466: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4467: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4468: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4469: for (j=1,survp=0. ; j<=nlstate; j++)
4470: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4471: lli= log(survp);
1.126 brouard 4472: }else if (mle==1){
1.242 brouard 4473: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4474: } else if(mle==2){
1.242 brouard 4475: 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 4476: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4477: 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 4478: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4479: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4480: } else{ /* mle=0 back to 1 */
1.242 brouard 4481: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4482: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4483: } /* End of if */
4484: ipmx +=1;
4485: sw += weight[i];
4486: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.342 brouard 4487: /* Printing covariates values for each contribution for checking */
1.343 ! brouard 4488: /* 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 4489: if(globpr){
1.246 brouard 4490: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4491: %11.6f %11.6f %11.6f ", \
1.242 brouard 4492: 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 4493: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.343 ! brouard 4494: /* printf("%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\ */
! 4495: /* %11.6f %11.6f %11.6f ", \ */
! 4496: /* num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw, */
! 4497: /* 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.242 brouard 4498: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4499: llt +=ll[k]*gipmx/gsw;
4500: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
1.335 brouard 4501: /* printf(" %10.6f",-ll[k]*gipmx/gsw); */
1.242 brouard 4502: }
1.343 ! brouard 4503: fprintf(ficresilk," %10.6f ", -llt);
1.335 brouard 4504: /* printf(" %10.6f\n", -llt); */
1.342 brouard 4505: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
1.343 ! brouard 4506: /* fprintf(ficresilk,"%09ld ", num[i]); */ /* not necessary */
! 4507: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
! 4508: fprintf(ficresilk," %g",covar[Tvar[TvarFind[kf]]][i]);
! 4509: }
! 4510: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
! 4511: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
! 4512: if(ipos!=iposold){ /* Not a product or first of a product */
! 4513: fprintf(ficresilk," %g",cov[ioffset+ipos]);
! 4514: /* printf(" %g",cov[ioffset+ipos]); */
! 4515: }else{
! 4516: fprintf(ficresilk,"*");
! 4517: /* printf("*"); */
1.342 brouard 4518: }
1.343 ! brouard 4519: iposold=ipos;
! 4520: }
! 4521: for (kk=1; kk<=cptcovage;kk++) {
! 4522: if(!FixedV[Tvar[Tage[kk]]]){
! 4523: fprintf(ficresilk," %g*age",covar[Tvar[Tage[kk]]][i]);
! 4524: /* printf(" %g*age",covar[Tvar[Tage[kk]]][i]); */
! 4525: }else{
! 4526: fprintf(ficresilk," %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
! 4527: /* 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 4528: }
1.343 ! brouard 4529: }
! 4530: /* printf("\n"); */
1.342 brouard 4531: /* } /\* End debugILK *\/ */
4532: fprintf(ficresilk,"\n");
4533: } /* End if globpr */
1.335 brouard 4534: } /* end of wave */
4535: } /* end of individual */
4536: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
1.232 brouard 4537: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
1.335 brouard 4538: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4539: if(globpr==0){ /* First time we count the contributions and weights */
4540: gipmx=ipmx;
4541: gsw=sw;
4542: }
1.343 ! brouard 4543: return -l;
1.126 brouard 4544: }
4545:
4546:
4547: /*************** function likelione ***********/
1.292 brouard 4548: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4549: {
4550: /* This routine should help understanding what is done with
4551: the selection of individuals/waves and
4552: to check the exact contribution to the likelihood.
4553: Plotting could be done.
1.342 brouard 4554: */
4555: void pstamp(FILE *ficres);
1.343 ! brouard 4556: int k, kf, kk, kvar, ncovv, iposold, ipos;
1.126 brouard 4557:
4558: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4559: strcpy(fileresilk,"ILK_");
1.202 brouard 4560: strcat(fileresilk,fileresu);
1.126 brouard 4561: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4562: printf("Problem with resultfile: %s\n", fileresilk);
4563: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4564: }
1.342 brouard 4565: pstamp(ficresilk);fprintf(ficresilk,"# model=1+age+%s\n",model);
1.214 brouard 4566: 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");
4567: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4568: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4569: for(k=1; k<=nlstate; k++)
4570: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
1.342 brouard 4571: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total) ");
4572:
4573: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
4574: for(kf=1;kf <= ncovf; kf++){
4575: fprintf(ficresilk,"V%d",Tvar[TvarFind[kf]]);
4576: /* printf("V%d",Tvar[TvarFind[kf]]); */
4577: }
4578: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){
1.343 ! brouard 4579: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
1.342 brouard 4580: if(ipos!=iposold){ /* Not a product or first of a product */
4581: /* printf(" %d",ipos); */
4582: fprintf(ficresilk," V%d",TvarVV[ncovv]);
4583: }else{
4584: /* printf("*"); */
4585: fprintf(ficresilk,"*");
1.343 ! brouard 4586: }
1.342 brouard 4587: iposold=ipos;
4588: }
4589: for (kk=1; kk<=cptcovage;kk++) {
4590: if(!FixedV[Tvar[Tage[kk]]]){
4591: /* printf(" %d*age(Fixed)",Tvar[Tage[kk]]); */
4592: fprintf(ficresilk," %d*age(Fixed)",Tvar[Tage[kk]]);
4593: }else{
4594: fprintf(ficresilk," %d*age(Varying)",Tvar[Tage[kk]]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
4595: /* printf(" %d*age(Varying)",Tvar[Tage[kk]]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
4596: }
4597: }
4598: /* } /\* End if debugILK *\/ */
4599: /* printf("\n"); */
4600: fprintf(ficresilk,"\n");
4601: } /* End glogpri */
1.126 brouard 4602:
1.292 brouard 4603: *fretone=(*func)(p);
1.126 brouard 4604: if(*globpri !=0){
4605: fclose(ficresilk);
1.205 brouard 4606: if (mle ==0)
4607: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4608: else if(mle >=1)
4609: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4610: 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 4611: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4612:
1.207 brouard 4613: 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 4614: <img src=\"%s-ori.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4615: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.343 ! brouard 4616: <img src=\"%s-dest.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
! 4617:
! 4618: for (k=1; k<= nlstate ; k++) {
! 4619: 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 \
! 4620: <img src=\"%s-p%dj.png\">\n",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
! 4621: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
! 4622: /* kvar=Tvar[TvarFind[kf]]; */ /* variable */
! 4623: 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> \
! 4624: <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]]);
! 4625: }
! 4626: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Loop on the time varying extended covariates (with extension of Vn*Vm */
! 4627: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
! 4628: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
! 4629: /* 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]); */
! 4630: if(ipos!=iposold){ /* Not a product or first of a product */
! 4631: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
! 4632: /* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); */
! 4633: 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) */
! 4634: 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> \
! 4635: <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);
! 4636: } /* End only for dummies time varying (single?) */
! 4637: }else{ /* Useless product */
! 4638: /* printf("*"); */
! 4639: /* fprintf(ficresilk,"*"); */
! 4640: }
! 4641: iposold=ipos;
! 4642: } /* For each time varying covariate */
! 4643: } /* End loop on states */
! 4644:
! 4645: /* if(debugILK){ */
! 4646: /* for(kf=1; kf <= ncovf; kf++){ /\* For each simple dummy covariate of the model *\/ */
! 4647: /* /\* kvar=Tvar[TvarFind[kf]]; *\/ /\* variable *\/ */
! 4648: /* for (k=1; k<= nlstate ; k++) { */
! 4649: /* 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> \ */
! 4650: /* <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]]); */
! 4651: /* } */
! 4652: /* } */
! 4653: /* for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /\* Loop on the time varying extended covariates (with extension of Vn*Vm *\/ */
! 4654: /* ipos=TvarVVind[ncovv]; /\* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate *\/ */
! 4655: /* kvar=TvarVV[ncovv]; /\* TvarVV={3, 1, 3} gives the name of each varying covariate *\/ */
! 4656: /* /\* 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]); *\/ */
! 4657: /* if(ipos!=iposold){ /\* Not a product or first of a product *\/ */
! 4658: /* /\* fprintf(ficresilk," V%d",TvarVV[ncovv]); *\/ */
! 4659: /* /\* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); *\/ */
! 4660: /* 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) *\/ */
! 4661: /* for (k=1; k<= nlstate ; k++) { */
! 4662: /* 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> \ */
! 4663: /* <img src=\"%s-p%dj-%d.png\">",k,k,kvar,kvar,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,kvar); */
! 4664: /* } /\* End state *\/ */
! 4665: /* } /\* End only for dummies time varying (single?) *\/ */
! 4666: /* }else{ /\* Useless product *\/ */
! 4667: /* /\* printf("*"); *\/ */
! 4668: /* /\* fprintf(ficresilk,"*"); *\/ */
! 4669: /* } */
! 4670: /* iposold=ipos; */
! 4671: /* } /\* For each time varying covariate *\/ */
! 4672: /* }/\* End debugILK *\/ */
1.207 brouard 4673: fflush(fichtm);
1.343 ! brouard 4674: }/* End globpri */
1.126 brouard 4675: return;
4676: }
4677:
4678:
4679: /*********** Maximum Likelihood Estimation ***************/
4680:
4681: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4682: {
1.319 brouard 4683: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4684: double **xi;
4685: double fret;
4686: double fretone; /* Only one call to likelihood */
4687: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4688:
4689: #ifdef NLOPT
4690: int creturn;
4691: nlopt_opt opt;
4692: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4693: double *lb;
4694: double minf; /* the minimum objective value, upon return */
4695: double * p1; /* Shifted parameters from 0 instead of 1 */
4696: myfunc_data dinst, *d = &dinst;
4697: #endif
4698:
4699:
1.126 brouard 4700: xi=matrix(1,npar,1,npar);
4701: for (i=1;i<=npar;i++)
4702: for (j=1;j<=npar;j++)
4703: xi[i][j]=(i==j ? 1.0 : 0.0);
4704: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4705: strcpy(filerespow,"POW_");
1.126 brouard 4706: strcat(filerespow,fileres);
4707: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4708: printf("Problem with resultfile: %s\n", filerespow);
4709: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4710: }
4711: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4712: for (i=1;i<=nlstate;i++)
4713: for(j=1;j<=nlstate+ndeath;j++)
4714: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4715: fprintf(ficrespow,"\n");
1.162 brouard 4716: #ifdef POWELL
1.319 brouard 4717: #ifdef LINMINORIGINAL
4718: #else /* LINMINORIGINAL */
4719:
4720: flatdir=ivector(1,npar);
4721: for (j=1;j<=npar;j++) flatdir[j]=0;
4722: #endif /*LINMINORIGINAL */
4723:
4724: #ifdef FLATSUP
4725: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4726: /* reorganizing p by suppressing flat directions */
4727: for(i=1, jk=1; i <=nlstate; i++){
4728: for(k=1; k <=(nlstate+ndeath); k++){
4729: if (k != i) {
4730: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4731: if(flatdir[jk]==1){
4732: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4733: }
4734: for(j=1; j <=ncovmodel; j++){
4735: printf("%12.7f ",p[jk]);
4736: jk++;
4737: }
4738: printf("\n");
4739: }
4740: }
4741: }
4742: /* skipping */
4743: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4744: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4745: for(k=1; k <=(nlstate+ndeath); k++){
4746: if (k != i) {
4747: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4748: if(flatdir[jk]==1){
4749: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4750: for(j=1; j <=ncovmodel; jk++,j++){
4751: printf(" p[%d]=%12.7f",jk, p[jk]);
4752: /*q[jjk]=p[jk];*/
4753: }
4754: }else{
4755: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4756: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4757: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4758: /*q[jjk]=p[jk];*/
4759: }
4760: }
4761: printf("\n");
4762: }
4763: fflush(stdout);
4764: }
4765: }
4766: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4767: #else /* FLATSUP */
1.126 brouard 4768: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4769: #endif /* FLATSUP */
4770:
4771: #ifdef LINMINORIGINAL
4772: #else
4773: free_ivector(flatdir,1,npar);
4774: #endif /* LINMINORIGINAL*/
4775: #endif /* POWELL */
1.126 brouard 4776:
1.162 brouard 4777: #ifdef NLOPT
4778: #ifdef NEWUOA
4779: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4780: #else
4781: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4782: #endif
4783: lb=vector(0,npar-1);
4784: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4785: nlopt_set_lower_bounds(opt, lb);
4786: nlopt_set_initial_step1(opt, 0.1);
4787:
4788: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4789: d->function = func;
4790: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4791: nlopt_set_min_objective(opt, myfunc, d);
4792: nlopt_set_xtol_rel(opt, ftol);
4793: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4794: printf("nlopt failed! %d\n",creturn);
4795: }
4796: else {
4797: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4798: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4799: iter=1; /* not equal */
4800: }
4801: nlopt_destroy(opt);
4802: #endif
1.319 brouard 4803: #ifdef FLATSUP
4804: /* npared = npar -flatd/ncovmodel; */
4805: /* xired= matrix(1,npared,1,npared); */
4806: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4807: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4808: /* free_matrix(xire,1,npared,1,npared); */
4809: #else /* FLATSUP */
4810: #endif /* FLATSUP */
1.126 brouard 4811: free_matrix(xi,1,npar,1,npar);
4812: fclose(ficrespow);
1.203 brouard 4813: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4814: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4815: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4816:
4817: }
4818:
4819: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4820: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4821: {
4822: double **a,**y,*x,pd;
1.203 brouard 4823: /* double **hess; */
1.164 brouard 4824: int i, j;
1.126 brouard 4825: int *indx;
4826:
4827: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4828: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4829: void lubksb(double **a, int npar, int *indx, double b[]) ;
4830: void ludcmp(double **a, int npar, int *indx, double *d) ;
4831: double gompertz(double p[]);
1.203 brouard 4832: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4833:
4834: printf("\nCalculation of the hessian matrix. Wait...\n");
4835: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4836: for (i=1;i<=npar;i++){
1.203 brouard 4837: printf("%d-",i);fflush(stdout);
4838: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4839:
4840: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4841:
4842: /* printf(" %f ",p[i]);
4843: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4844: }
4845:
4846: for (i=1;i<=npar;i++) {
4847: for (j=1;j<=npar;j++) {
4848: if (j>i) {
1.203 brouard 4849: printf(".%d-%d",i,j);fflush(stdout);
4850: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4851: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4852:
4853: hess[j][i]=hess[i][j];
4854: /*printf(" %lf ",hess[i][j]);*/
4855: }
4856: }
4857: }
4858: printf("\n");
4859: fprintf(ficlog,"\n");
4860:
4861: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4862: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4863:
4864: a=matrix(1,npar,1,npar);
4865: y=matrix(1,npar,1,npar);
4866: x=vector(1,npar);
4867: indx=ivector(1,npar);
4868: for (i=1;i<=npar;i++)
4869: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4870: ludcmp(a,npar,indx,&pd);
4871:
4872: for (j=1;j<=npar;j++) {
4873: for (i=1;i<=npar;i++) x[i]=0;
4874: x[j]=1;
4875: lubksb(a,npar,indx,x);
4876: for (i=1;i<=npar;i++){
4877: matcov[i][j]=x[i];
4878: }
4879: }
4880:
4881: printf("\n#Hessian matrix#\n");
4882: fprintf(ficlog,"\n#Hessian matrix#\n");
4883: for (i=1;i<=npar;i++) {
4884: for (j=1;j<=npar;j++) {
1.203 brouard 4885: printf("%.6e ",hess[i][j]);
4886: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4887: }
4888: printf("\n");
4889: fprintf(ficlog,"\n");
4890: }
4891:
1.203 brouard 4892: /* printf("\n#Covariance matrix#\n"); */
4893: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4894: /* for (i=1;i<=npar;i++) { */
4895: /* for (j=1;j<=npar;j++) { */
4896: /* printf("%.6e ",matcov[i][j]); */
4897: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4898: /* } */
4899: /* printf("\n"); */
4900: /* fprintf(ficlog,"\n"); */
4901: /* } */
4902:
1.126 brouard 4903: /* Recompute Inverse */
1.203 brouard 4904: /* for (i=1;i<=npar;i++) */
4905: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4906: /* ludcmp(a,npar,indx,&pd); */
4907:
4908: /* printf("\n#Hessian matrix recomputed#\n"); */
4909:
4910: /* for (j=1;j<=npar;j++) { */
4911: /* for (i=1;i<=npar;i++) x[i]=0; */
4912: /* x[j]=1; */
4913: /* lubksb(a,npar,indx,x); */
4914: /* for (i=1;i<=npar;i++){ */
4915: /* y[i][j]=x[i]; */
4916: /* printf("%.3e ",y[i][j]); */
4917: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4918: /* } */
4919: /* printf("\n"); */
4920: /* fprintf(ficlog,"\n"); */
4921: /* } */
4922:
4923: /* Verifying the inverse matrix */
4924: #ifdef DEBUGHESS
4925: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4926:
1.203 brouard 4927: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4928: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4929:
4930: for (j=1;j<=npar;j++) {
4931: for (i=1;i<=npar;i++){
1.203 brouard 4932: printf("%.2f ",y[i][j]);
4933: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4934: }
4935: printf("\n");
4936: fprintf(ficlog,"\n");
4937: }
1.203 brouard 4938: #endif
1.126 brouard 4939:
4940: free_matrix(a,1,npar,1,npar);
4941: free_matrix(y,1,npar,1,npar);
4942: free_vector(x,1,npar);
4943: free_ivector(indx,1,npar);
1.203 brouard 4944: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4945:
4946:
4947: }
4948:
4949: /*************** hessian matrix ****************/
4950: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4951: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4952: int i;
4953: int l=1, lmax=20;
1.203 brouard 4954: double k1,k2, res, fx;
1.132 brouard 4955: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4956: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4957: int k=0,kmax=10;
4958: double l1;
4959:
4960: fx=func(x);
4961: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4962: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4963: l1=pow(10,l);
4964: delts=delt;
4965: for(k=1 ; k <kmax; k=k+1){
4966: delt = delta*(l1*k);
4967: p2[theta]=x[theta] +delt;
1.145 brouard 4968: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4969: p2[theta]=x[theta]-delt;
4970: k2=func(p2)-fx;
4971: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4972: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4973:
1.203 brouard 4974: #ifdef DEBUGHESSII
1.126 brouard 4975: 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);
4976: 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);
4977: #endif
4978: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4979: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4980: k=kmax;
4981: }
4982: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4983: k=kmax; l=lmax*10;
1.126 brouard 4984: }
4985: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4986: delts=delt;
4987: }
1.203 brouard 4988: } /* End loop k */
1.126 brouard 4989: }
4990: delti[theta]=delts;
4991: return res;
4992:
4993: }
4994:
1.203 brouard 4995: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4996: {
4997: int i;
1.164 brouard 4998: int l=1, lmax=20;
1.126 brouard 4999: double k1,k2,k3,k4,res,fx;
1.132 brouard 5000: double p2[MAXPARM+1];
1.203 brouard 5001: int k, kmax=1;
5002: double v1, v2, cv12, lc1, lc2;
1.208 brouard 5003:
5004: int firstime=0;
1.203 brouard 5005:
1.126 brouard 5006: fx=func(x);
1.203 brouard 5007: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 5008: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 5009: p2[thetai]=x[thetai]+delti[thetai]*k;
5010: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 5011: k1=func(p2)-fx;
5012:
1.203 brouard 5013: p2[thetai]=x[thetai]+delti[thetai]*k;
5014: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 5015: k2=func(p2)-fx;
5016:
1.203 brouard 5017: p2[thetai]=x[thetai]-delti[thetai]*k;
5018: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 5019: k3=func(p2)-fx;
5020:
1.203 brouard 5021: p2[thetai]=x[thetai]-delti[thetai]*k;
5022: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 5023: k4=func(p2)-fx;
1.203 brouard 5024: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
5025: if(k1*k2*k3*k4 <0.){
1.208 brouard 5026: firstime=1;
1.203 brouard 5027: kmax=kmax+10;
1.208 brouard 5028: }
5029: if(kmax >=10 || firstime ==1){
1.246 brouard 5030: 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);
5031: 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 5032: 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);
5033: 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);
5034: }
5035: #ifdef DEBUGHESSIJ
5036: v1=hess[thetai][thetai];
5037: v2=hess[thetaj][thetaj];
5038: cv12=res;
5039: /* Computing eigen value of Hessian matrix */
5040: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
5041: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
5042: if ((lc2 <0) || (lc1 <0) ){
5043: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
5044: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
5045: 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);
5046: 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);
5047: }
1.126 brouard 5048: #endif
5049: }
5050: return res;
5051: }
5052:
1.203 brouard 5053: /* Not done yet: Was supposed to fix if not exactly at the maximum */
5054: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
5055: /* { */
5056: /* int i; */
5057: /* int l=1, lmax=20; */
5058: /* double k1,k2,k3,k4,res,fx; */
5059: /* double p2[MAXPARM+1]; */
5060: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
5061: /* int k=0,kmax=10; */
5062: /* double l1; */
5063:
5064: /* fx=func(x); */
5065: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
5066: /* l1=pow(10,l); */
5067: /* delts=delt; */
5068: /* for(k=1 ; k <kmax; k=k+1){ */
5069: /* delt = delti*(l1*k); */
5070: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
5071: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
5072: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
5073: /* k1=func(p2)-fx; */
5074:
5075: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
5076: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
5077: /* k2=func(p2)-fx; */
5078:
5079: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
5080: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
5081: /* k3=func(p2)-fx; */
5082:
5083: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
5084: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
5085: /* k4=func(p2)-fx; */
5086: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
5087: /* #ifdef DEBUGHESSIJ */
5088: /* 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); */
5089: /* 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); */
5090: /* #endif */
5091: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
5092: /* k=kmax; */
5093: /* } */
5094: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
5095: /* k=kmax; l=lmax*10; */
5096: /* } */
5097: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
5098: /* delts=delt; */
5099: /* } */
5100: /* } /\* End loop k *\/ */
5101: /* } */
5102: /* delti[theta]=delts; */
5103: /* return res; */
5104: /* } */
5105:
5106:
1.126 brouard 5107: /************** Inverse of matrix **************/
5108: void ludcmp(double **a, int n, int *indx, double *d)
5109: {
5110: int i,imax,j,k;
5111: double big,dum,sum,temp;
5112: double *vv;
5113:
5114: vv=vector(1,n);
5115: *d=1.0;
5116: for (i=1;i<=n;i++) {
5117: big=0.0;
5118: for (j=1;j<=n;j++)
5119: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 5120: if (big == 0.0){
5121: printf(" Singular Hessian matrix at row %d:\n",i);
5122: for (j=1;j<=n;j++) {
5123: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
5124: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
5125: }
5126: fflush(ficlog);
5127: fclose(ficlog);
5128: nrerror("Singular matrix in routine ludcmp");
5129: }
1.126 brouard 5130: vv[i]=1.0/big;
5131: }
5132: for (j=1;j<=n;j++) {
5133: for (i=1;i<j;i++) {
5134: sum=a[i][j];
5135: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
5136: a[i][j]=sum;
5137: }
5138: big=0.0;
5139: for (i=j;i<=n;i++) {
5140: sum=a[i][j];
5141: for (k=1;k<j;k++)
5142: sum -= a[i][k]*a[k][j];
5143: a[i][j]=sum;
5144: if ( (dum=vv[i]*fabs(sum)) >= big) {
5145: big=dum;
5146: imax=i;
5147: }
5148: }
5149: if (j != imax) {
5150: for (k=1;k<=n;k++) {
5151: dum=a[imax][k];
5152: a[imax][k]=a[j][k];
5153: a[j][k]=dum;
5154: }
5155: *d = -(*d);
5156: vv[imax]=vv[j];
5157: }
5158: indx[j]=imax;
5159: if (a[j][j] == 0.0) a[j][j]=TINY;
5160: if (j != n) {
5161: dum=1.0/(a[j][j]);
5162: for (i=j+1;i<=n;i++) a[i][j] *= dum;
5163: }
5164: }
5165: free_vector(vv,1,n); /* Doesn't work */
5166: ;
5167: }
5168:
5169: void lubksb(double **a, int n, int *indx, double b[])
5170: {
5171: int i,ii=0,ip,j;
5172: double sum;
5173:
5174: for (i=1;i<=n;i++) {
5175: ip=indx[i];
5176: sum=b[ip];
5177: b[ip]=b[i];
5178: if (ii)
5179: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
5180: else if (sum) ii=i;
5181: b[i]=sum;
5182: }
5183: for (i=n;i>=1;i--) {
5184: sum=b[i];
5185: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
5186: b[i]=sum/a[i][i];
5187: }
5188: }
5189:
5190: void pstamp(FILE *fichier)
5191: {
1.196 brouard 5192: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 5193: }
5194:
1.297 brouard 5195: void date2dmy(double date,double *day, double *month, double *year){
5196: double yp=0., yp1=0., yp2=0.;
5197:
5198: yp1=modf(date,&yp);/* extracts integral of date in yp and
5199: fractional in yp1 */
5200: *year=yp;
5201: yp2=modf((yp1*12),&yp);
5202: *month=yp;
5203: yp1=modf((yp2*30.5),&yp);
5204: *day=yp;
5205: if(*day==0) *day=1;
5206: if(*month==0) *month=1;
5207: }
5208:
1.253 brouard 5209:
5210:
1.126 brouard 5211: /************ Frequencies ********************/
1.251 brouard 5212: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 5213: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
5214: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 5215: { /* Some frequencies as well as proposing some starting values */
1.332 brouard 5216: /* Frequencies of any combination of dummy covariate used in the model equation */
1.265 brouard 5217: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 5218: int iind=0, iage=0;
5219: int mi; /* Effective wave */
5220: int first;
5221: double ***freq; /* Frequencies */
1.268 brouard 5222: 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 */
5223: 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 5224: double *meanq, *stdq, *idq;
1.226 brouard 5225: double **meanqt;
5226: double *pp, **prop, *posprop, *pospropt;
5227: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
5228: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
5229: double agebegin, ageend;
5230:
5231: pp=vector(1,nlstate);
1.251 brouard 5232: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5233: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
5234: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
5235: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
5236: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 5237: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 5238: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 5239: meanqt=matrix(1,lastpass,1,nqtveff);
5240: strcpy(fileresp,"P_");
5241: strcat(fileresp,fileresu);
5242: /*strcat(fileresphtm,fileresu);*/
5243: if((ficresp=fopen(fileresp,"w"))==NULL) {
5244: printf("Problem with prevalence resultfile: %s\n", fileresp);
5245: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
5246: exit(0);
5247: }
1.240 brouard 5248:
1.226 brouard 5249: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
5250: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
5251: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5252: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5253: fflush(ficlog);
5254: exit(70);
5255: }
5256: else{
5257: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 5258: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5259: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5260: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5261: }
1.319 brouard 5262: 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 5263:
1.226 brouard 5264: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
5265: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
5266: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5267: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5268: fflush(ficlog);
5269: exit(70);
1.240 brouard 5270: } else{
1.226 brouard 5271: 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 5272: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5273: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5274: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5275: }
1.319 brouard 5276: 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 5277:
1.253 brouard 5278: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
5279: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 5280: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5281: j1=0;
1.126 brouard 5282:
1.227 brouard 5283: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
1.335 brouard 5284: j=cptcoveff; /* Only simple dummy covariates used in the model */
1.330 brouard 5285: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 5286: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 5287:
5288:
1.226 brouard 5289: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
5290: reference=low_education V1=0,V2=0
5291: med_educ V1=1 V2=0,
5292: high_educ V1=0 V2=1
1.330 brouard 5293: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 5294: */
1.249 brouard 5295: dateintsum=0;
5296: k2cpt=0;
5297:
1.253 brouard 5298: if(cptcoveff == 0 )
1.265 brouard 5299: nl=1; /* Constant and age model only */
1.253 brouard 5300: else
5301: nl=2;
1.265 brouard 5302:
5303: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
5304: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.335 brouard 5305: * Loop on j1(1 to 2**cptcoveff) covariate combination
1.265 brouard 5306: * freq[s1][s2][iage] =0.
5307: * Loop on iind
5308: * ++freq[s1][s2][iage] weighted
5309: * end iind
5310: * if covariate and j!0
5311: * headers Variable on one line
5312: * endif cov j!=0
5313: * header of frequency table by age
5314: * Loop on age
5315: * pp[s1]+=freq[s1][s2][iage] weighted
5316: * pos+=freq[s1][s2][iage] weighted
5317: * Loop on s1 initial state
5318: * fprintf(ficresp
5319: * end s1
5320: * end age
5321: * if j!=0 computes starting values
5322: * end compute starting values
5323: * end j1
5324: * end nl
5325: */
1.253 brouard 5326: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
5327: if(nj==1)
5328: j=0; /* First pass for the constant */
1.265 brouard 5329: else{
1.335 brouard 5330: 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 5331: }
1.251 brouard 5332: first=1;
1.332 brouard 5333: 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 5334: posproptt=0.;
1.330 brouard 5335: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 5336: scanf("%d", i);*/
5337: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 5338: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 5339: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 5340: freq[i][s2][m]=0;
1.251 brouard 5341:
5342: for (i=1; i<=nlstate; i++) {
1.240 brouard 5343: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 5344: prop[i][m]=0;
5345: posprop[i]=0;
5346: pospropt[i]=0;
5347: }
1.283 brouard 5348: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 5349: idq[z1]=0.;
5350: meanq[z1]=0.;
5351: stdq[z1]=0.;
1.283 brouard 5352: }
5353: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 5354: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 5355: /* meanqt[m][z1]=0.; */
5356: /* } */
5357: /* } */
1.251 brouard 5358: /* dateintsum=0; */
5359: /* k2cpt=0; */
5360:
1.265 brouard 5361: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 5362: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
5363: bool=1;
5364: if(j !=0){
5365: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.335 brouard 5366: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
5367: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
1.251 brouard 5368: /* if(Tvaraff[z1] ==-20){ */
5369: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
5370: /* }else if(Tvaraff[z1] ==-10){ */
5371: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 brouard 5372: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
1.335 brouard 5373: /* 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); */
5374: if(Tvaraff[z1]<1 || Tvaraff[z1]>=NCOVMAX)
1.338 brouard 5375: printf("Error Tvaraff[z1]=%d<1 or >=%d, cptcoveff=%d model=1+age+%s\n",Tvaraff[z1],NCOVMAX, cptcoveff, model);
1.332 brouard 5376: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]){ /* for combination j1 of covariates */
1.265 brouard 5377: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 5378: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.332 brouard 5379: /* 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", */
5380: /* bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),*/
5381: /* j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
1.251 brouard 5382: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
5383: } /* Onlyf fixed */
5384: } /* end z1 */
1.335 brouard 5385: } /* cptcoveff > 0 */
1.251 brouard 5386: } /* end any */
5387: }/* end j==0 */
1.265 brouard 5388: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 5389: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 5390: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 5391: m=mw[mi][iind];
5392: if(j!=0){
5393: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.335 brouard 5394: for (z1=1; z1<=cptcoveff; z1++) {
1.251 brouard 5395: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 5396: /* iv= Tvar[Tmodelind[z1]]-ncovcol-nqv; /\* Good *\/ */
5397: iv= Tvar[Tmodelind[z1]]; /* Good *//* because cotvar starts now at first at ncovcol+nqv+ntv */
1.332 brouard 5398: 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 5399: value is -1, we don't select. It differs from the
5400: constant and age model which counts them. */
5401: bool=0; /* not selected */
5402: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.334 brouard 5403: /* i1=Tvaraff[z1]; */
5404: /* i2=TnsdVar[i1]; */
5405: /* i3=nbcode[i1][i2]; */
5406: /* i4=covar[i1][iind]; */
5407: /* if(i4 != i3){ */
5408: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) { /* Bug valgrind */
1.251 brouard 5409: bool=0;
5410: }
5411: }
5412: }
5413: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
5414: } /* end j==0 */
5415: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 5416: if(bool==1){ /*Selected */
1.251 brouard 5417: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
5418: and mw[mi+1][iind]. dh depends on stepm. */
5419: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
5420: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
5421: if(m >=firstpass && m <=lastpass){
5422: k2=anint[m][iind]+(mint[m][iind]/12.);
5423: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
5424: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
5425: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
5426: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
5427: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
5428: if (m<lastpass) {
5429: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
5430: /* 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]); */
5431: if(s[m][iind]==-1)
5432: 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.));
5433: 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 5434: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
5435: if(!isnan(covar[ncovcol+z1][iind])){
1.332 brouard 5436: idq[z1]=idq[z1]+weight[iind];
5437: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
5438: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
5439: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
1.311 brouard 5440: }
1.284 brouard 5441: }
1.251 brouard 5442: /* if((int)agev[m][iind] == 55) */
5443: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
5444: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
5445: 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 5446: }
1.251 brouard 5447: } /* end if between passes */
5448: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
5449: dateintsum=dateintsum+k2; /* on all covariates ?*/
5450: k2cpt++;
5451: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 5452: }
1.251 brouard 5453: }else{
5454: bool=1;
5455: }/* end bool 2 */
5456: } /* end m */
1.284 brouard 5457: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
5458: /* idq[z1]=idq[z1]+weight[iind]; */
5459: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
5460: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
5461: /* } */
1.251 brouard 5462: } /* end bool */
5463: } /* end iind = 1 to imx */
1.319 brouard 5464: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 5465: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
5466:
5467:
5468: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.335 brouard 5469: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
1.265 brouard 5470: pstamp(ficresp);
1.335 brouard 5471: if (cptcoveff>0 && j!=0){
1.265 brouard 5472: pstamp(ficresp);
1.251 brouard 5473: printf( "\n#********** Variable ");
5474: fprintf(ficresp, "\n#********** Variable ");
5475: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
5476: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
5477: fprintf(ficlog, "\n#********** Variable ");
1.340 brouard 5478: for (z1=1; z1<=cptcoveff; z1++){
1.251 brouard 5479: if(!FixedV[Tvaraff[z1]]){
1.330 brouard 5480: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5481: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5482: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5483: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5484: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 5485: }else{
1.330 brouard 5486: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5487: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5488: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5489: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5490: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 5491: }
5492: }
5493: printf( "**********\n#");
5494: fprintf(ficresp, "**********\n#");
5495: fprintf(ficresphtm, "**********</h3>\n");
5496: fprintf(ficresphtmfr, "**********</h3>\n");
5497: fprintf(ficlog, "**********\n");
5498: }
1.284 brouard 5499: /*
5500: Printing means of quantitative variables if any
5501: */
5502: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5503: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5504: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5505: if(weightopt==1){
5506: printf(" Weighted mean and standard deviation of");
5507: fprintf(ficlog," Weighted mean and standard deviation of");
5508: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5509: }
1.311 brouard 5510: /* mu = \frac{w x}{\sum w}
5511: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5512: */
5513: 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]));
5514: 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]));
5515: 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 5516: }
5517: /* for (z1=1; z1<= nqtveff; z1++) { */
5518: /* for(m=1;m<=lastpass;m++){ */
5519: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5520: /* } */
5521: /* } */
1.283 brouard 5522:
1.251 brouard 5523: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.335 brouard 5524: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 5525: fprintf(ficresp, " Age");
1.335 brouard 5526: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) {
5527: 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]]);
5528: fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5529: }
1.251 brouard 5530: for(i=1; i<=nlstate;i++) {
1.335 brouard 5531: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5532: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5533: }
1.335 brouard 5534: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5535: fprintf(ficresphtm, "\n");
5536:
5537: /* Header of frequency table by age */
5538: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5539: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5540: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5541: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5542: if(s2!=0 && m!=0)
5543: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5544: }
1.226 brouard 5545: }
1.251 brouard 5546: fprintf(ficresphtmfr, "\n");
5547:
5548: /* For each age */
5549: for(iage=iagemin; iage <= iagemax+3; iage++){
5550: fprintf(ficresphtm,"<tr>");
5551: if(iage==iagemax+1){
5552: fprintf(ficlog,"1");
5553: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5554: }else if(iage==iagemax+2){
5555: fprintf(ficlog,"0");
5556: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5557: }else if(iage==iagemax+3){
5558: fprintf(ficlog,"Total");
5559: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5560: }else{
1.240 brouard 5561: if(first==1){
1.251 brouard 5562: first=0;
5563: printf("See log file for details...\n");
5564: }
5565: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5566: fprintf(ficlog,"Age %d", iage);
5567: }
1.265 brouard 5568: for(s1=1; s1 <=nlstate ; s1++){
5569: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5570: pp[s1] += freq[s1][m][iage];
1.251 brouard 5571: }
1.265 brouard 5572: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5573: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5574: pos += freq[s1][m][iage];
5575: if(pp[s1]>=1.e-10){
1.251 brouard 5576: if(first==1){
1.265 brouard 5577: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5578: }
1.265 brouard 5579: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5580: }else{
5581: if(first==1)
1.265 brouard 5582: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5583: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5584: }
5585: }
5586:
1.265 brouard 5587: for(s1=1; s1 <=nlstate ; s1++){
5588: /* posprop[s1]=0; */
5589: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5590: pp[s1] += freq[s1][m][iage];
5591: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5592:
5593: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5594: pos += pp[s1]; /* pos is the total number of transitions until this age */
5595: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5596: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5597: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5598: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5599: }
5600:
5601: /* Writing ficresp */
1.335 brouard 5602: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5603: if( iage <= iagemax){
5604: fprintf(ficresp," %d",iage);
5605: }
5606: }else if( nj==2){
5607: if( iage <= iagemax){
5608: fprintf(ficresp," %d",iage);
1.335 brouard 5609: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.265 brouard 5610: }
1.240 brouard 5611: }
1.265 brouard 5612: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5613: if(pos>=1.e-5){
1.251 brouard 5614: if(first==1)
1.265 brouard 5615: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5616: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5617: }else{
5618: if(first==1)
1.265 brouard 5619: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5620: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5621: }
5622: if( iage <= iagemax){
5623: if(pos>=1.e-5){
1.335 brouard 5624: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5625: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5626: }else if( nj==2){
5627: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5628: }
5629: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5630: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5631: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5632: } else{
1.335 brouard 5633: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 5634: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5635: }
1.240 brouard 5636: }
1.265 brouard 5637: pospropt[s1] +=posprop[s1];
5638: } /* end loop s1 */
1.251 brouard 5639: /* pospropt=0.; */
1.265 brouard 5640: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5641: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5642: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5643: if(first==1){
1.265 brouard 5644: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5645: }
1.265 brouard 5646: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5647: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5648: }
1.265 brouard 5649: if(s1!=0 && m!=0)
5650: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5651: }
1.265 brouard 5652: } /* end loop s1 */
1.251 brouard 5653: posproptt=0.;
1.265 brouard 5654: for(s1=1; s1 <=nlstate; s1++){
5655: posproptt += pospropt[s1];
1.251 brouard 5656: }
5657: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5658: fprintf(ficresphtm,"</tr>\n");
1.335 brouard 5659: if((cptcoveff==0 && nj==1)|| nj==2 ) {
1.265 brouard 5660: if(iage <= iagemax)
5661: fprintf(ficresp,"\n");
1.240 brouard 5662: }
1.251 brouard 5663: if(first==1)
5664: printf("Others in log...\n");
5665: fprintf(ficlog,"\n");
5666: } /* end loop age iage */
1.265 brouard 5667:
1.251 brouard 5668: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5669: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5670: if(posproptt < 1.e-5){
1.265 brouard 5671: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5672: }else{
1.265 brouard 5673: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5674: }
1.226 brouard 5675: }
1.251 brouard 5676: fprintf(ficresphtm,"</tr>\n");
5677: fprintf(ficresphtm,"</table>\n");
5678: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5679: if(posproptt < 1.e-5){
1.251 brouard 5680: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5681: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5682: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5683: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5684: invalidvarcomb[j1]=1;
1.226 brouard 5685: }else{
1.338 brouard 5686: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced (or no resultline).</p>",j1);
1.251 brouard 5687: invalidvarcomb[j1]=0;
1.226 brouard 5688: }
1.251 brouard 5689: fprintf(ficresphtmfr,"</table>\n");
5690: fprintf(ficlog,"\n");
5691: if(j!=0){
5692: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5693: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5694: for(k=1; k <=(nlstate+ndeath); k++){
5695: if (k != i) {
1.265 brouard 5696: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5697: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5698: if(j1==1){ /* All dummy covariates to zero */
5699: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5700: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5701: printf("%d%d ",i,k);
5702: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5703: 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]));
5704: 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]));
5705: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5706: }
1.253 brouard 5707: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5708: for(iage=iagemin; iage <= iagemax+3; iage++){
5709: x[iage]= (double)iage;
5710: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5711: /* 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 5712: }
1.268 brouard 5713: /* Some are not finite, but linreg will ignore these ages */
5714: no=0;
1.253 brouard 5715: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5716: pstart[s1]=b;
5717: pstart[s1-1]=a;
1.252 brouard 5718: }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 */
5719: 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]);
5720: 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 5721: 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 5722: printf("%d%d ",i,k);
5723: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5724: 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 5725: }else{ /* Other cases, like quantitative fixed or varying covariates */
5726: ;
5727: }
5728: /* printf("%12.7f )", param[i][jj][k]); */
5729: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5730: s1++;
1.251 brouard 5731: } /* end jj */
5732: } /* end k!= i */
5733: } /* end k */
1.265 brouard 5734: } /* end i, s1 */
1.251 brouard 5735: } /* end j !=0 */
5736: } /* end selected combination of covariate j1 */
5737: if(j==0){ /* We can estimate starting values from the occurences in each case */
5738: printf("#Freqsummary: Starting values for the constants:\n");
5739: fprintf(ficlog,"\n");
1.265 brouard 5740: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5741: for(k=1; k <=(nlstate+ndeath); k++){
5742: if (k != i) {
5743: printf("%d%d ",i,k);
5744: fprintf(ficlog,"%d%d ",i,k);
5745: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5746: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5747: if(jj==1){ /* Age has to be done */
1.265 brouard 5748: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5749: 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]));
5750: 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 5751: }
5752: /* printf("%12.7f )", param[i][jj][k]); */
5753: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5754: s1++;
1.250 brouard 5755: }
1.251 brouard 5756: printf("\n");
5757: fprintf(ficlog,"\n");
1.250 brouard 5758: }
5759: }
1.284 brouard 5760: } /* end of state i */
1.251 brouard 5761: printf("#Freqsummary\n");
5762: fprintf(ficlog,"\n");
1.265 brouard 5763: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5764: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5765: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5766: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5767: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5768: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5769: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5770: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5771: /* } */
5772: }
1.265 brouard 5773: } /* end loop s1 */
1.251 brouard 5774:
5775: printf("\n");
5776: fprintf(ficlog,"\n");
5777: } /* end j=0 */
1.249 brouard 5778: } /* end j */
1.252 brouard 5779:
1.253 brouard 5780: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5781: for(i=1, jk=1; i <=nlstate; i++){
5782: for(j=1; j <=nlstate+ndeath; j++){
5783: if(j!=i){
5784: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5785: printf("%1d%1d",i,j);
5786: fprintf(ficparo,"%1d%1d",i,j);
5787: for(k=1; k<=ncovmodel;k++){
5788: /* printf(" %lf",param[i][j][k]); */
5789: /* fprintf(ficparo," %lf",param[i][j][k]); */
5790: p[jk]=pstart[jk];
5791: printf(" %f ",pstart[jk]);
5792: fprintf(ficparo," %f ",pstart[jk]);
5793: jk++;
5794: }
5795: printf("\n");
5796: fprintf(ficparo,"\n");
5797: }
5798: }
5799: }
5800: } /* end mle=-2 */
1.226 brouard 5801: dateintmean=dateintsum/k2cpt;
1.296 brouard 5802: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5803:
1.226 brouard 5804: fclose(ficresp);
5805: fclose(ficresphtm);
5806: fclose(ficresphtmfr);
1.283 brouard 5807: free_vector(idq,1,nqfveff);
1.226 brouard 5808: free_vector(meanq,1,nqfveff);
1.284 brouard 5809: free_vector(stdq,1,nqfveff);
1.226 brouard 5810: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5811: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5812: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5813: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5814: free_vector(pospropt,1,nlstate);
5815: free_vector(posprop,1,nlstate);
1.251 brouard 5816: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5817: free_vector(pp,1,nlstate);
5818: /* End of freqsummary */
5819: }
1.126 brouard 5820:
1.268 brouard 5821: /* Simple linear regression */
5822: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5823:
5824: /* y=a+bx regression */
5825: double sumx = 0.0; /* sum of x */
5826: double sumx2 = 0.0; /* sum of x**2 */
5827: double sumxy = 0.0; /* sum of x * y */
5828: double sumy = 0.0; /* sum of y */
5829: double sumy2 = 0.0; /* sum of y**2 */
5830: double sume2 = 0.0; /* sum of square or residuals */
5831: double yhat;
5832:
5833: double denom=0;
5834: int i;
5835: int ne=*no;
5836:
5837: for ( i=ifi, ne=0;i<=ila;i++) {
5838: if(!isfinite(x[i]) || !isfinite(y[i])){
5839: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5840: continue;
5841: }
5842: ne=ne+1;
5843: sumx += x[i];
5844: sumx2 += x[i]*x[i];
5845: sumxy += x[i] * y[i];
5846: sumy += y[i];
5847: sumy2 += y[i]*y[i];
5848: denom = (ne * sumx2 - sumx*sumx);
5849: /* 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); */
5850: }
5851:
5852: denom = (ne * sumx2 - sumx*sumx);
5853: if (denom == 0) {
5854: // vertical, slope m is infinity
5855: *b = INFINITY;
5856: *a = 0;
5857: if (r) *r = 0;
5858: return 1;
5859: }
5860:
5861: *b = (ne * sumxy - sumx * sumy) / denom;
5862: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5863: if (r!=NULL) {
5864: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5865: sqrt((sumx2 - sumx*sumx/ne) *
5866: (sumy2 - sumy*sumy/ne));
5867: }
5868: *no=ne;
5869: for ( i=ifi, ne=0;i<=ila;i++) {
5870: if(!isfinite(x[i]) || !isfinite(y[i])){
5871: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5872: continue;
5873: }
5874: ne=ne+1;
5875: yhat = y[i] - *a -*b* x[i];
5876: sume2 += yhat * yhat ;
5877:
5878: denom = (ne * sumx2 - sumx*sumx);
5879: /* 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); */
5880: }
5881: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5882: *sa= *sb * sqrt(sumx2/ne);
5883:
5884: return 0;
5885: }
5886:
1.126 brouard 5887: /************ Prevalence ********************/
1.227 brouard 5888: 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)
5889: {
5890: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5891: in each health status at the date of interview (if between dateprev1 and dateprev2).
5892: We still use firstpass and lastpass as another selection.
5893: */
1.126 brouard 5894:
1.227 brouard 5895: int i, m, jk, j1, bool, z1,j, iv;
5896: int mi; /* Effective wave */
5897: int iage;
5898: double agebegin, ageend;
5899:
5900: double **prop;
5901: double posprop;
5902: double y2; /* in fractional years */
5903: int iagemin, iagemax;
5904: int first; /** to stop verbosity which is redirected to log file */
5905:
5906: iagemin= (int) agemin;
5907: iagemax= (int) agemax;
5908: /*pp=vector(1,nlstate);*/
1.251 brouard 5909: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5910: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5911: j1=0;
1.222 brouard 5912:
1.227 brouard 5913: /*j=cptcoveff;*/
5914: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5915:
1.288 brouard 5916: first=0;
1.335 brouard 5917: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of simple dummy covariates */
1.227 brouard 5918: for (i=1; i<=nlstate; i++)
1.251 brouard 5919: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5920: prop[i][iage]=0.0;
5921: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5922: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5923: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5924:
5925: for (i=1; i<=imx; i++) { /* Each individual */
5926: bool=1;
5927: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5928: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5929: m=mw[mi][i];
5930: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5931: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5932: for (z1=1; z1<=cptcoveff; z1++){
5933: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 5934: iv= Tvar[Tmodelind[z1]];/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.332 brouard 5935: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality */
1.227 brouard 5936: bool=0;
5937: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.332 brouard 5938: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.227 brouard 5939: bool=0;
5940: }
5941: }
5942: if(bool==1){ /* Otherwise we skip that wave/person */
5943: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5944: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5945: if(m >=firstpass && m <=lastpass){
5946: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5947: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5948: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5949: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5950: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5951: 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);
5952: exit(1);
5953: }
5954: if (s[m][i]>0 && s[m][i]<=nlstate) {
5955: /*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]]);*/
5956: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5957: prop[s[m][i]][iagemax+3] += weight[i];
5958: } /* end valid statuses */
5959: } /* end selection of dates */
5960: } /* end selection of waves */
5961: } /* end bool */
5962: } /* end wave */
5963: } /* end individual */
5964: for(i=iagemin; i <= iagemax+3; i++){
5965: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5966: posprop += prop[jk][i];
5967: }
5968:
5969: for(jk=1; jk <=nlstate ; jk++){
5970: if( i <= iagemax){
5971: if(posprop>=1.e-5){
5972: probs[i][jk][j1]= prop[jk][i]/posprop;
5973: } else{
1.288 brouard 5974: if(!first){
5975: first=1;
1.266 brouard 5976: 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]);
5977: }else{
1.288 brouard 5978: 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 5979: }
5980: }
5981: }
5982: }/* end jk */
5983: }/* end i */
1.222 brouard 5984: /*} *//* end i1 */
1.227 brouard 5985: } /* end j1 */
1.222 brouard 5986:
1.227 brouard 5987: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5988: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5989: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5990: } /* End of prevalence */
1.126 brouard 5991:
5992: /************* Waves Concatenation ***************/
5993:
5994: 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)
5995: {
1.298 brouard 5996: /* 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 5997: Death is a valid wave (if date is known).
5998: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5999: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 6000: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 6001: */
1.126 brouard 6002:
1.224 brouard 6003: int i=0, mi=0, m=0, mli=0;
1.126 brouard 6004: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
6005: double sum=0., jmean=0.;*/
1.224 brouard 6006: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 6007: int j, k=0,jk, ju, jl;
6008: double sum=0.;
6009: first=0;
1.214 brouard 6010: firstwo=0;
1.217 brouard 6011: firsthree=0;
1.218 brouard 6012: firstfour=0;
1.164 brouard 6013: jmin=100000;
1.126 brouard 6014: jmax=-1;
6015: jmean=0.;
1.224 brouard 6016:
6017: /* Treating live states */
1.214 brouard 6018: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 6019: mi=0; /* First valid wave */
1.227 brouard 6020: mli=0; /* Last valid wave */
1.309 brouard 6021: m=firstpass; /* Loop on waves */
6022: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 6023: 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 */
6024: mli=m-1;/* mw[++mi][i]=m-1; */
6025: }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 6026: 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 6027: mli=m;
1.224 brouard 6028: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
6029: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 6030: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 6031: }
1.309 brouard 6032: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 6033: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 6034: break;
1.224 brouard 6035: #else
1.317 brouard 6036: 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 6037: if(firsthree == 0){
1.302 brouard 6038: 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 6039: firsthree=1;
1.317 brouard 6040: }else if(firsthree >=1 && firsthree < 10){
6041: 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);
6042: firsthree++;
6043: }else if(firsthree == 10){
6044: printf("Information, too many Information flags: no more reported to log either\n");
6045: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
6046: firsthree++;
6047: }else{
6048: firsthree++;
1.227 brouard 6049: }
1.309 brouard 6050: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 6051: mli=m;
6052: }
6053: if(s[m][i]==-2){ /* Vital status is really unknown */
6054: nbwarn++;
1.309 brouard 6055: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 6056: 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);
6057: 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);
6058: }
6059: break;
6060: }
6061: break;
1.224 brouard 6062: #endif
1.227 brouard 6063: }/* End m >= lastpass */
1.126 brouard 6064: }/* end while */
1.224 brouard 6065:
1.227 brouard 6066: /* 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 6067: /* After last pass */
1.224 brouard 6068: /* Treating death states */
1.214 brouard 6069: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 6070: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
6071: /* } */
1.126 brouard 6072: mi++; /* Death is another wave */
6073: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 6074: /* Only death is a correct wave */
1.126 brouard 6075: mw[mi][i]=m;
1.257 brouard 6076: } /* else not in a death state */
1.224 brouard 6077: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 6078: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 6079: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 6080: 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 6081: nbwarn++;
6082: if(firstfiv==0){
1.309 brouard 6083: 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 6084: firstfiv=1;
6085: }else{
1.309 brouard 6086: 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 6087: }
1.309 brouard 6088: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
6089: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 6090: nberr++;
6091: if(firstwo==0){
1.309 brouard 6092: 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 6093: firstwo=1;
6094: }
1.309 brouard 6095: 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 6096: }
1.257 brouard 6097: }else{ /* if date of interview is unknown */
1.227 brouard 6098: /* death is known but not confirmed by death status at any wave */
6099: if(firstfour==0){
1.309 brouard 6100: 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 6101: firstfour=1;
6102: }
1.309 brouard 6103: 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 6104: }
1.224 brouard 6105: } /* end if date of death is known */
6106: #endif
1.309 brouard 6107: wav[i]=mi; /* mi should be the last effective wave (or mli), */
6108: /* wav[i]=mw[mi][i]; */
1.126 brouard 6109: if(mi==0){
6110: nbwarn++;
6111: if(first==0){
1.227 brouard 6112: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
6113: first=1;
1.126 brouard 6114: }
6115: if(first==1){
1.227 brouard 6116: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 6117: }
6118: } /* end mi==0 */
6119: } /* End individuals */
1.214 brouard 6120: /* wav and mw are no more changed */
1.223 brouard 6121:
1.317 brouard 6122: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
6123: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
6124:
6125:
1.126 brouard 6126: for(i=1; i<=imx; i++){
6127: for(mi=1; mi<wav[i];mi++){
6128: if (stepm <=0)
1.227 brouard 6129: dh[mi][i]=1;
1.126 brouard 6130: else{
1.260 brouard 6131: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 6132: if (agedc[i] < 2*AGESUP) {
6133: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
6134: if(j==0) j=1; /* Survives at least one month after exam */
6135: else if(j<0){
6136: nberr++;
6137: 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]);
6138: j=1; /* Temporary Dangerous patch */
6139: 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);
6140: 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]);
6141: 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);
6142: }
6143: k=k+1;
6144: if (j >= jmax){
6145: jmax=j;
6146: ijmax=i;
6147: }
6148: if (j <= jmin){
6149: jmin=j;
6150: ijmin=i;
6151: }
6152: sum=sum+j;
6153: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
6154: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
6155: }
6156: }
6157: else{
6158: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 6159: /* 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 6160:
1.227 brouard 6161: k=k+1;
6162: if (j >= jmax) {
6163: jmax=j;
6164: ijmax=i;
6165: }
6166: else if (j <= jmin){
6167: jmin=j;
6168: ijmin=i;
6169: }
6170: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
6171: /*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]);*/
6172: if(j<0){
6173: nberr++;
6174: 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]);
6175: 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]);
6176: }
6177: sum=sum+j;
6178: }
6179: jk= j/stepm;
6180: jl= j -jk*stepm;
6181: ju= j -(jk+1)*stepm;
6182: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
6183: if(jl==0){
6184: dh[mi][i]=jk;
6185: bh[mi][i]=0;
6186: }else{ /* We want a negative bias in order to only have interpolation ie
6187: * to avoid the price of an extra matrix product in likelihood */
6188: dh[mi][i]=jk+1;
6189: bh[mi][i]=ju;
6190: }
6191: }else{
6192: if(jl <= -ju){
6193: dh[mi][i]=jk;
6194: bh[mi][i]=jl; /* bias is positive if real duration
6195: * is higher than the multiple of stepm and negative otherwise.
6196: */
6197: }
6198: else{
6199: dh[mi][i]=jk+1;
6200: bh[mi][i]=ju;
6201: }
6202: if(dh[mi][i]==0){
6203: dh[mi][i]=1; /* At least one step */
6204: bh[mi][i]=ju; /* At least one step */
6205: /* 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);*/
6206: }
6207: } /* end if mle */
1.126 brouard 6208: }
6209: } /* end wave */
6210: }
6211: jmean=sum/k;
6212: 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 6213: 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 6214: }
1.126 brouard 6215:
6216: /*********** Tricode ****************************/
1.220 brouard 6217: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 6218: {
6219: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
6220: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
6221: * Boring subroutine which should only output nbcode[Tvar[j]][k]
6222: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
6223: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
6224: */
1.130 brouard 6225:
1.242 brouard 6226: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
6227: int modmaxcovj=0; /* Modality max of covariates j */
6228: int cptcode=0; /* Modality max of covariates j */
6229: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 6230:
6231:
1.242 brouard 6232: /* cptcoveff=0; */
6233: /* *cptcov=0; */
1.126 brouard 6234:
1.242 brouard 6235: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 6236: for (k=1; k <= maxncov; k++)
6237: for(j=1; j<=2; j++)
6238: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 6239:
1.242 brouard 6240: /* Loop on covariates without age and products and no quantitative variable */
1.335 brouard 6241: 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 6242: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
1.343 ! brouard 6243: /* printf("Testing k=%d, cptcovt=%d\n",k, cptcovt); */
1.339 brouard 6244: if(Dummy[k]==0 && Typevar[k] !=1 && Typevar[k] != 2){ /* Dummy covariate and not age product nor fixed product */
1.242 brouard 6245: switch(Fixed[k]) {
6246: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 6247: modmaxcovj=0;
6248: modmincovj=0;
1.242 brouard 6249: 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 6250: /* 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 6251: ij=(int)(covar[Tvar[k]][i]);
6252: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
6253: * If product of Vn*Vm, still boolean *:
6254: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
6255: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
6256: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
6257: modality of the nth covariate of individual i. */
6258: if (ij > modmaxcovj)
6259: modmaxcovj=ij;
6260: else if (ij < modmincovj)
6261: modmincovj=ij;
1.287 brouard 6262: if (ij <0 || ij >1 ){
1.311 brouard 6263: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6264: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6265: fflush(ficlog);
6266: exit(1);
1.287 brouard 6267: }
6268: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 6269: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
6270: exit(1);
6271: }else
6272: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
6273: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
6274: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
6275: /* getting the maximum value of the modality of the covariate
6276: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
6277: female ies 1, then modmaxcovj=1.
6278: */
6279: } /* end for loop on individuals i */
6280: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6281: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6282: cptcode=modmaxcovj;
6283: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
6284: /*for (i=0; i<=cptcode; i++) {*/
6285: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
6286: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6287: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6288: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
6289: if( j != -1){
6290: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
6291: covariate for which somebody answered excluding
6292: undefined. Usually 2: 0 and 1. */
6293: }
6294: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
6295: covariate for which somebody answered including
6296: undefined. Usually 3: -1, 0 and 1. */
6297: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
6298: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
6299: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 6300:
1.242 brouard 6301: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
6302: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
6303: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
6304: /* modmincovj=3; modmaxcovj = 7; */
6305: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
6306: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
6307: /* defining two dummy variables: variables V1_1 and V1_2.*/
6308: /* nbcode[Tvar[j]][ij]=k; */
6309: /* nbcode[Tvar[j]][1]=0; */
6310: /* nbcode[Tvar[j]][2]=1; */
6311: /* nbcode[Tvar[j]][3]=2; */
6312: /* To be continued (not working yet). */
6313: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 6314:
6315: /* 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*/
6316: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
6317: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
6318: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
6319: /*, could be restored in the future */
6320: 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 6321: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
6322: break;
6323: }
6324: ij++;
1.287 brouard 6325: 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 6326: cptcode = ij; /* New max modality for covar j */
6327: } /* end of loop on modality i=-1 to 1 or more */
6328: break;
6329: case 1: /* Testing on varying covariate, could be simple and
6330: * should look at waves or product of fixed *
6331: * varying. No time to test -1, assuming 0 and 1 only */
6332: ij=0;
6333: for(i=0; i<=1;i++){
6334: nbcode[Tvar[k]][++ij]=i;
6335: }
6336: break;
6337: default:
6338: break;
6339: } /* end switch */
6340: } /* end dummy test */
1.342 brouard 6341: if(Dummy[k]==1 && Typevar[k] !=1 && Fixed ==0){ /* Fixed Quantitative covariate and not age product */
1.311 brouard 6342: 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 6343: if(Tvar[k]<=0 || Tvar[k]>=NCOVMAX){
6344: printf("Error k=%d \n",k);
6345: exit(1);
6346: }
1.311 brouard 6347: if(isnan(covar[Tvar[k]][i])){
6348: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6349: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6350: fflush(ficlog);
6351: exit(1);
6352: }
6353: }
1.335 brouard 6354: } /* end Quanti */
1.287 brouard 6355: } /* 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 6356:
6357: for (k=-1; k< maxncov; k++) Ndum[k]=0;
6358: /* Look at fixed dummy (single or product) covariates to check empty modalities */
6359: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
6360: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
6361: 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 */
6362: 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 */
6363: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
6364: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
6365:
6366: ij=0;
6367: /* 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 6368: 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 */
6369: /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
1.242 brouard 6370: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
6371: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
1.335 brouard 6372: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy simple and non empty in the model */
6373: /* Typevar[k] =0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
6374: /* 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 6375: /* If product not in single variable we don't print results */
6376: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
1.335 brouard 6377: ++ij;/* V5 + V4 + V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V1, *//* V5 quanti, V2 quanti, V4, V3, V1 dummies */
6378: /* k= 1 2 3 4 5 6 7 8 9 */
6379: /* Tvar[k]= 5 4 3 6 5 2 7 1 1 */
6380: /* ij 1 2 3 */
6381: /* Tvaraff[ij]= 4 3 1 */
6382: /* Tmodelind[ij]=2 3 9 */
6383: /* TmodelInvind[ij]=2 1 1 */
1.242 brouard 6384: 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*/
6385: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
6386: 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 */
6387: if(Fixed[k]!=0)
6388: anyvaryingduminmodel=1;
6389: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
6390: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
6391: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
6392: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
6393: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
6394: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
6395: }
6396: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
6397: /* ij--; */
6398: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.335 brouard 6399: *cptcov=ij; /* cptcov= Number of total real effective simple dummies (fixed or time arying) effective (used as cptcoveff in other functions)
1.242 brouard 6400: * because they can be excluded from the model and real
6401: * if in the model but excluded because missing values, but how to get k from ij?*/
6402: for(j=ij+1; j<= cptcovt; j++){
6403: Tvaraff[j]=0;
6404: Tmodelind[j]=0;
6405: }
6406: for(j=ntveff+1; j<= cptcovt; j++){
6407: TmodelInvind[j]=0;
6408: }
6409: /* To be sorted */
6410: ;
6411: }
1.126 brouard 6412:
1.145 brouard 6413:
1.126 brouard 6414: /*********** Health Expectancies ****************/
6415:
1.235 brouard 6416: 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 6417:
6418: {
6419: /* Health expectancies, no variances */
1.329 brouard 6420: /* cij is the combination in the list of combination of dummy covariates */
6421: /* strstart is a string of time at start of computing */
1.164 brouard 6422: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 6423: int nhstepma, nstepma; /* Decreasing with age */
6424: double age, agelim, hf;
6425: double ***p3mat;
6426: double eip;
6427:
1.238 brouard 6428: /* pstamp(ficreseij); */
1.126 brouard 6429: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
6430: fprintf(ficreseij,"# Age");
6431: for(i=1; i<=nlstate;i++){
6432: for(j=1; j<=nlstate;j++){
6433: fprintf(ficreseij," e%1d%1d ",i,j);
6434: }
6435: fprintf(ficreseij," e%1d. ",i);
6436: }
6437: fprintf(ficreseij,"\n");
6438:
6439:
6440: if(estepm < stepm){
6441: printf ("Problem %d lower than %d\n",estepm, stepm);
6442: }
6443: else hstepm=estepm;
6444: /* We compute the life expectancy from trapezoids spaced every estepm months
6445: * This is mainly to measure the difference between two models: for example
6446: * if stepm=24 months pijx are given only every 2 years and by summing them
6447: * we are calculating an estimate of the Life Expectancy assuming a linear
6448: * progression in between and thus overestimating or underestimating according
6449: * to the curvature of the survival function. If, for the same date, we
6450: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6451: * to compare the new estimate of Life expectancy with the same linear
6452: * hypothesis. A more precise result, taking into account a more precise
6453: * curvature will be obtained if estepm is as small as stepm. */
6454:
6455: /* For example we decided to compute the life expectancy with the smallest unit */
6456: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6457: nhstepm is the number of hstepm from age to agelim
6458: nstepm is the number of stepm from age to agelin.
1.270 brouard 6459: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 6460: and note for a fixed period like estepm months */
6461: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6462: survival function given by stepm (the optimization length). Unfortunately it
6463: means that if the survival funtion is printed only each two years of age and if
6464: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6465: results. So we changed our mind and took the option of the best precision.
6466: */
6467: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6468:
6469: agelim=AGESUP;
6470: /* If stepm=6 months */
6471: /* Computed by stepm unit matrices, product of hstepm matrices, stored
6472: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
6473:
6474: /* nhstepm age range expressed in number of stepm */
6475: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6476: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6477: /* if (stepm >= YEARM) hstepm=1;*/
6478: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6479: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6480:
6481: for (age=bage; age<=fage; age ++){
6482: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6483: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6484: /* if (stepm >= YEARM) hstepm=1;*/
6485: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
6486:
6487: /* If stepm=6 months */
6488: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6489: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 brouard 6490: /* 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 6491: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 6492:
6493: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6494:
6495: printf("%d|",(int)age);fflush(stdout);
6496: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6497:
6498: /* Computing expectancies */
6499: for(i=1; i<=nlstate;i++)
6500: for(j=1; j<=nlstate;j++)
6501: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6502: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
6503:
6504: /* 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]);*/
6505:
6506: }
6507:
6508: fprintf(ficreseij,"%3.0f",age );
6509: for(i=1; i<=nlstate;i++){
6510: eip=0;
6511: for(j=1; j<=nlstate;j++){
6512: eip +=eij[i][j][(int)age];
6513: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
6514: }
6515: fprintf(ficreseij,"%9.4f", eip );
6516: }
6517: fprintf(ficreseij,"\n");
6518:
6519: }
6520: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6521: printf("\n");
6522: fprintf(ficlog,"\n");
6523:
6524: }
6525:
1.235 brouard 6526: 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 6527:
6528: {
6529: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6530: to initial status i, ei. .
1.126 brouard 6531: */
1.336 brouard 6532: /* Very time consuming function, but already optimized with precov */
1.126 brouard 6533: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6534: int nhstepma, nstepma; /* Decreasing with age */
6535: double age, agelim, hf;
6536: double ***p3matp, ***p3matm, ***varhe;
6537: double **dnewm,**doldm;
6538: double *xp, *xm;
6539: double **gp, **gm;
6540: double ***gradg, ***trgradg;
6541: int theta;
6542:
6543: double eip, vip;
6544:
6545: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6546: xp=vector(1,npar);
6547: xm=vector(1,npar);
6548: dnewm=matrix(1,nlstate*nlstate,1,npar);
6549: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6550:
6551: pstamp(ficresstdeij);
6552: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6553: fprintf(ficresstdeij,"# Age");
6554: for(i=1; i<=nlstate;i++){
6555: for(j=1; j<=nlstate;j++)
6556: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6557: fprintf(ficresstdeij," e%1d. ",i);
6558: }
6559: fprintf(ficresstdeij,"\n");
6560:
6561: pstamp(ficrescveij);
6562: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6563: fprintf(ficrescveij,"# Age");
6564: for(i=1; i<=nlstate;i++)
6565: for(j=1; j<=nlstate;j++){
6566: cptj= (j-1)*nlstate+i;
6567: for(i2=1; i2<=nlstate;i2++)
6568: for(j2=1; j2<=nlstate;j2++){
6569: cptj2= (j2-1)*nlstate+i2;
6570: if(cptj2 <= cptj)
6571: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6572: }
6573: }
6574: fprintf(ficrescveij,"\n");
6575:
6576: if(estepm < stepm){
6577: printf ("Problem %d lower than %d\n",estepm, stepm);
6578: }
6579: else hstepm=estepm;
6580: /* We compute the life expectancy from trapezoids spaced every estepm months
6581: * This is mainly to measure the difference between two models: for example
6582: * if stepm=24 months pijx are given only every 2 years and by summing them
6583: * we are calculating an estimate of the Life Expectancy assuming a linear
6584: * progression in between and thus overestimating or underestimating according
6585: * to the curvature of the survival function. If, for the same date, we
6586: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6587: * to compare the new estimate of Life expectancy with the same linear
6588: * hypothesis. A more precise result, taking into account a more precise
6589: * curvature will be obtained if estepm is as small as stepm. */
6590:
6591: /* For example we decided to compute the life expectancy with the smallest unit */
6592: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6593: nhstepm is the number of hstepm from age to agelim
6594: nstepm is the number of stepm from age to agelin.
6595: Look at hpijx to understand the reason of that which relies in memory size
6596: and note for a fixed period like estepm months */
6597: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6598: survival function given by stepm (the optimization length). Unfortunately it
6599: means that if the survival funtion is printed only each two years of age and if
6600: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6601: results. So we changed our mind and took the option of the best precision.
6602: */
6603: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6604:
6605: /* If stepm=6 months */
6606: /* nhstepm age range expressed in number of stepm */
6607: agelim=AGESUP;
6608: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6609: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6610: /* if (stepm >= YEARM) hstepm=1;*/
6611: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6612:
6613: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6614: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6615: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6616: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6617: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6618: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6619:
6620: for (age=bage; age<=fage; age ++){
6621: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6622: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6623: /* if (stepm >= YEARM) hstepm=1;*/
6624: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6625:
1.126 brouard 6626: /* If stepm=6 months */
6627: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6628: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6629:
6630: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6631:
1.126 brouard 6632: /* Computing Variances of health expectancies */
6633: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6634: decrease memory allocation */
6635: for(theta=1; theta <=npar; theta++){
6636: for(i=1; i<=npar; i++){
1.222 brouard 6637: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6638: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6639: }
1.235 brouard 6640: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6641: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6642:
1.126 brouard 6643: for(j=1; j<= nlstate; j++){
1.222 brouard 6644: for(i=1; i<=nlstate; i++){
6645: for(h=0; h<=nhstepm-1; h++){
6646: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6647: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6648: }
6649: }
1.126 brouard 6650: }
1.218 brouard 6651:
1.126 brouard 6652: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6653: for(h=0; h<=nhstepm-1; h++){
6654: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6655: }
1.126 brouard 6656: }/* End theta */
6657:
6658:
6659: for(h=0; h<=nhstepm-1; h++)
6660: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6661: for(theta=1; theta <=npar; theta++)
6662: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6663:
1.218 brouard 6664:
1.222 brouard 6665: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6666: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6667: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6668:
1.222 brouard 6669: printf("%d|",(int)age);fflush(stdout);
6670: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6671: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6672: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6673: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6674: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6675: for(ij=1;ij<=nlstate*nlstate;ij++)
6676: for(ji=1;ji<=nlstate*nlstate;ji++)
6677: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6678: }
6679: }
1.320 brouard 6680: /* if((int)age ==50){ */
6681: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6682: /* } */
1.126 brouard 6683: /* Computing expectancies */
1.235 brouard 6684: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6685: for(i=1; i<=nlstate;i++)
6686: for(j=1; j<=nlstate;j++)
1.222 brouard 6687: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6688: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6689:
1.222 brouard 6690: /* 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 6691:
1.222 brouard 6692: }
1.269 brouard 6693:
6694: /* Standard deviation of expectancies ij */
1.126 brouard 6695: fprintf(ficresstdeij,"%3.0f",age );
6696: for(i=1; i<=nlstate;i++){
6697: eip=0.;
6698: vip=0.;
6699: for(j=1; j<=nlstate;j++){
1.222 brouard 6700: eip += eij[i][j][(int)age];
6701: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6702: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6703: 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 6704: }
6705: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6706: }
6707: fprintf(ficresstdeij,"\n");
1.218 brouard 6708:
1.269 brouard 6709: /* Variance of expectancies ij */
1.126 brouard 6710: fprintf(ficrescveij,"%3.0f",age );
6711: for(i=1; i<=nlstate;i++)
6712: for(j=1; j<=nlstate;j++){
1.222 brouard 6713: cptj= (j-1)*nlstate+i;
6714: for(i2=1; i2<=nlstate;i2++)
6715: for(j2=1; j2<=nlstate;j2++){
6716: cptj2= (j2-1)*nlstate+i2;
6717: if(cptj2 <= cptj)
6718: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6719: }
1.126 brouard 6720: }
6721: fprintf(ficrescveij,"\n");
1.218 brouard 6722:
1.126 brouard 6723: }
6724: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6725: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6726: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6727: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6728: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6729: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6730: printf("\n");
6731: fprintf(ficlog,"\n");
1.218 brouard 6732:
1.126 brouard 6733: free_vector(xm,1,npar);
6734: free_vector(xp,1,npar);
6735: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6736: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6737: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6738: }
1.218 brouard 6739:
1.126 brouard 6740: /************ Variance ******************/
1.235 brouard 6741: 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 6742: {
1.279 brouard 6743: /** Variance of health expectancies
6744: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6745: * double **newm;
6746: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6747: */
1.218 brouard 6748:
6749: /* int movingaverage(); */
6750: double **dnewm,**doldm;
6751: double **dnewmp,**doldmp;
6752: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6753: int first=0;
1.218 brouard 6754: int k;
6755: double *xp;
1.279 brouard 6756: double **gp, **gm; /**< for var eij */
6757: double ***gradg, ***trgradg; /**< for var eij */
6758: double **gradgp, **trgradgp; /**< for var p point j */
6759: double *gpp, *gmp; /**< for var p point j */
6760: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6761: double ***p3mat;
6762: double age,agelim, hf;
6763: /* double ***mobaverage; */
6764: int theta;
6765: char digit[4];
6766: char digitp[25];
6767:
6768: char fileresprobmorprev[FILENAMELENGTH];
6769:
6770: if(popbased==1){
6771: if(mobilav!=0)
6772: strcpy(digitp,"-POPULBASED-MOBILAV_");
6773: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6774: }
6775: else
6776: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6777:
1.218 brouard 6778: /* if (mobilav!=0) { */
6779: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6780: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6781: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6782: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6783: /* } */
6784: /* } */
6785:
6786: strcpy(fileresprobmorprev,"PRMORPREV-");
6787: sprintf(digit,"%-d",ij);
6788: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6789: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6790: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6791: strcat(fileresprobmorprev,fileresu);
6792: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6793: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6794: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6795: }
6796: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6797: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6798: pstamp(ficresprobmorprev);
6799: 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 6800: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
1.337 brouard 6801:
6802: /* We use TinvDoQresult[nres][resultmodel[nres][j] we sort according to the equation model and the resultline: it is a choice */
6803: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ /\* To be done*\/ */
6804: /* fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
6805: /* } */
6806: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */ /* To be done*/
6807: fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 6808: }
1.337 brouard 6809: /* for(j=1;j<=cptcoveff;j++) */
6810: /* fprintf(ficresprobmorprev," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,TnsdVar[Tvaraff[j]])]); */
1.238 brouard 6811: fprintf(ficresprobmorprev,"\n");
6812:
1.218 brouard 6813: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6814: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6815: fprintf(ficresprobmorprev," p.%-d SE",j);
6816: for(i=1; i<=nlstate;i++)
6817: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6818: }
6819: fprintf(ficresprobmorprev,"\n");
6820:
6821: fprintf(ficgp,"\n# Routine varevsij");
6822: fprintf(ficgp,"\nunset title \n");
6823: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6824: 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");
6825: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6826:
1.218 brouard 6827: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6828: pstamp(ficresvij);
6829: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6830: if(popbased==1)
6831: 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);
6832: else
6833: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6834: fprintf(ficresvij,"# Age");
6835: for(i=1; i<=nlstate;i++)
6836: for(j=1; j<=nlstate;j++)
6837: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6838: fprintf(ficresvij,"\n");
6839:
6840: xp=vector(1,npar);
6841: dnewm=matrix(1,nlstate,1,npar);
6842: doldm=matrix(1,nlstate,1,nlstate);
6843: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6844: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6845:
6846: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6847: gpp=vector(nlstate+1,nlstate+ndeath);
6848: gmp=vector(nlstate+1,nlstate+ndeath);
6849: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6850:
1.218 brouard 6851: if(estepm < stepm){
6852: printf ("Problem %d lower than %d\n",estepm, stepm);
6853: }
6854: else hstepm=estepm;
6855: /* For example we decided to compute the life expectancy with the smallest unit */
6856: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6857: nhstepm is the number of hstepm from age to agelim
6858: nstepm is the number of stepm from age to agelim.
6859: Look at function hpijx to understand why because of memory size limitations,
6860: we decided (b) to get a life expectancy respecting the most precise curvature of the
6861: survival function given by stepm (the optimization length). Unfortunately it
6862: means that if the survival funtion is printed every two years of age and if
6863: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6864: results. So we changed our mind and took the option of the best precision.
6865: */
6866: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6867: agelim = AGESUP;
6868: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6869: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6870: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6871: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6872: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6873: gp=matrix(0,nhstepm,1,nlstate);
6874: gm=matrix(0,nhstepm,1,nlstate);
6875:
6876:
6877: for(theta=1; theta <=npar; theta++){
6878: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6879: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6880: }
1.279 brouard 6881: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6882: * returns into prlim .
1.288 brouard 6883: */
1.242 brouard 6884: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6885:
6886: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6887: if (popbased==1) {
6888: if(mobilav ==0){
6889: for(i=1; i<=nlstate;i++)
6890: prlim[i][i]=probs[(int)age][i][ij];
6891: }else{ /* mobilav */
6892: for(i=1; i<=nlstate;i++)
6893: prlim[i][i]=mobaverage[(int)age][i][ij];
6894: }
6895: }
1.295 brouard 6896: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6897: */
6898: 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 6899: /**< 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 6900: * at horizon h in state j including mortality.
6901: */
1.218 brouard 6902: for(j=1; j<= nlstate; j++){
6903: for(h=0; h<=nhstepm; h++){
6904: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6905: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6906: }
6907: }
1.279 brouard 6908: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6909: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6910: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6911: */
6912: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6913: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6914: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6915: }
6916:
6917: /* Again with minus shift */
1.218 brouard 6918:
6919: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6920: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6921:
1.242 brouard 6922: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6923:
6924: if (popbased==1) {
6925: if(mobilav ==0){
6926: for(i=1; i<=nlstate;i++)
6927: prlim[i][i]=probs[(int)age][i][ij];
6928: }else{ /* mobilav */
6929: for(i=1; i<=nlstate;i++)
6930: prlim[i][i]=mobaverage[(int)age][i][ij];
6931: }
6932: }
6933:
1.235 brouard 6934: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6935:
6936: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6937: for(h=0; h<=nhstepm; h++){
6938: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6939: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6940: }
6941: }
6942: /* This for computing probability of death (h=1 means
6943: computed over hstepm matrices product = hstepm*stepm months)
6944: as a weighted average of prlim.
6945: */
6946: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6947: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6948: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6949: }
1.279 brouard 6950: /* end shifting computations */
6951:
6952: /**< Computing gradient matrix at horizon h
6953: */
1.218 brouard 6954: for(j=1; j<= nlstate; j++) /* vareij */
6955: for(h=0; h<=nhstepm; h++){
6956: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6957: }
1.279 brouard 6958: /**< Gradient of overall mortality p.3 (or p.j)
6959: */
6960: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6961: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6962: }
6963:
6964: } /* End theta */
1.279 brouard 6965:
6966: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6967: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6968:
6969: for(h=0; h<=nhstepm; h++) /* veij */
6970: for(j=1; j<=nlstate;j++)
6971: for(theta=1; theta <=npar; theta++)
6972: trgradg[h][j][theta]=gradg[h][theta][j];
6973:
6974: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6975: for(theta=1; theta <=npar; theta++)
6976: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6977: /**< as well as its transposed matrix
6978: */
1.218 brouard 6979:
6980: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6981: for(i=1;i<=nlstate;i++)
6982: for(j=1;j<=nlstate;j++)
6983: vareij[i][j][(int)age] =0.;
1.279 brouard 6984:
6985: /* Computing trgradg by matcov by gradg at age and summing over h
6986: * and k (nhstepm) formula 15 of article
6987: * Lievre-Brouard-Heathcote
6988: */
6989:
1.218 brouard 6990: for(h=0;h<=nhstepm;h++){
6991: for(k=0;k<=nhstepm;k++){
6992: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6993: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6994: for(i=1;i<=nlstate;i++)
6995: for(j=1;j<=nlstate;j++)
6996: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6997: }
6998: }
6999:
1.279 brouard 7000: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
7001: * p.j overall mortality formula 49 but computed directly because
7002: * we compute the grad (wix pijx) instead of grad (pijx),even if
7003: * wix is independent of theta.
7004: */
1.218 brouard 7005: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
7006: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
7007: for(j=nlstate+1;j<=nlstate+ndeath;j++)
7008: for(i=nlstate+1;i<=nlstate+ndeath;i++)
7009: varppt[j][i]=doldmp[j][i];
7010: /* end ppptj */
7011: /* x centered again */
7012:
1.242 brouard 7013: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 7014:
7015: if (popbased==1) {
7016: if(mobilav ==0){
7017: for(i=1; i<=nlstate;i++)
7018: prlim[i][i]=probs[(int)age][i][ij];
7019: }else{ /* mobilav */
7020: for(i=1; i<=nlstate;i++)
7021: prlim[i][i]=mobaverage[(int)age][i][ij];
7022: }
7023: }
7024:
7025: /* This for computing probability of death (h=1 means
7026: computed over hstepm (estepm) matrices product = hstepm*stepm months)
7027: as a weighted average of prlim.
7028: */
1.235 brouard 7029: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 7030: for(j=nlstate+1;j<=nlstate+ndeath;j++){
7031: for(i=1,gmp[j]=0.;i<= nlstate; i++)
7032: gmp[j] += prlim[i][i]*p3mat[i][j][1];
7033: }
7034: /* end probability of death */
7035:
7036: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
7037: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
7038: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
7039: for(i=1; i<=nlstate;i++){
7040: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
7041: }
7042: }
7043: fprintf(ficresprobmorprev,"\n");
7044:
7045: fprintf(ficresvij,"%.0f ",age );
7046: for(i=1; i<=nlstate;i++)
7047: for(j=1; j<=nlstate;j++){
7048: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
7049: }
7050: fprintf(ficresvij,"\n");
7051: free_matrix(gp,0,nhstepm,1,nlstate);
7052: free_matrix(gm,0,nhstepm,1,nlstate);
7053: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
7054: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
7055: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7056: } /* End age */
7057: free_vector(gpp,nlstate+1,nlstate+ndeath);
7058: free_vector(gmp,nlstate+1,nlstate+ndeath);
7059: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
7060: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
7061: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
7062: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
7063: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
7064: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
7065: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
7066: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
7067: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
7068: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
7069: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
7070: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
7071: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
7072: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
7073: 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);
7074: /* 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 7075: */
1.218 brouard 7076: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
7077: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 7078:
1.218 brouard 7079: free_vector(xp,1,npar);
7080: free_matrix(doldm,1,nlstate,1,nlstate);
7081: free_matrix(dnewm,1,nlstate,1,npar);
7082: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
7083: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
7084: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
7085: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7086: fclose(ficresprobmorprev);
7087: fflush(ficgp);
7088: fflush(fichtm);
7089: } /* end varevsij */
1.126 brouard 7090:
7091: /************ Variance of prevlim ******************/
1.269 brouard 7092: 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 7093: {
1.205 brouard 7094: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 7095: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 7096:
1.268 brouard 7097: double **dnewmpar,**doldm;
1.126 brouard 7098: int i, j, nhstepm, hstepm;
7099: double *xp;
7100: double *gp, *gm;
7101: double **gradg, **trgradg;
1.208 brouard 7102: double **mgm, **mgp;
1.126 brouard 7103: double age,agelim;
7104: int theta;
7105:
7106: pstamp(ficresvpl);
1.288 brouard 7107: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 7108: fprintf(ficresvpl,"# Age ");
7109: if(nresult >=1)
7110: fprintf(ficresvpl," Result# ");
1.126 brouard 7111: for(i=1; i<=nlstate;i++)
7112: fprintf(ficresvpl," %1d-%1d",i,i);
7113: fprintf(ficresvpl,"\n");
7114:
7115: xp=vector(1,npar);
1.268 brouard 7116: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 7117: doldm=matrix(1,nlstate,1,nlstate);
7118:
7119: hstepm=1*YEARM; /* Every year of age */
7120: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
7121: agelim = AGESUP;
7122: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
7123: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
7124: if (stepm >= YEARM) hstepm=1;
7125: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
7126: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 7127: mgp=matrix(1,npar,1,nlstate);
7128: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 7129: gp=vector(1,nlstate);
7130: gm=vector(1,nlstate);
7131:
7132: for(theta=1; theta <=npar; theta++){
7133: for(i=1; i<=npar; i++){ /* Computes gradient */
7134: xp[i] = x[i] + (i==theta ?delti[theta]:0);
7135: }
1.288 brouard 7136: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
7137: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
7138: /* else */
7139: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 7140: for(i=1;i<=nlstate;i++){
1.126 brouard 7141: gp[i] = prlim[i][i];
1.208 brouard 7142: mgp[theta][i] = prlim[i][i];
7143: }
1.126 brouard 7144: for(i=1; i<=npar; i++) /* Computes gradient */
7145: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 7146: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
7147: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
7148: /* else */
7149: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 7150: for(i=1;i<=nlstate;i++){
1.126 brouard 7151: gm[i] = prlim[i][i];
1.208 brouard 7152: mgm[theta][i] = prlim[i][i];
7153: }
1.126 brouard 7154: for(i=1;i<=nlstate;i++)
7155: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 7156: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 7157: } /* End theta */
7158:
7159: trgradg =matrix(1,nlstate,1,npar);
7160:
7161: for(j=1; j<=nlstate;j++)
7162: for(theta=1; theta <=npar; theta++)
7163: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 7164: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7165: /* printf("\nmgm mgp %d ",(int)age); */
7166: /* for(j=1; j<=nlstate;j++){ */
7167: /* printf(" %d ",j); */
7168: /* for(theta=1; theta <=npar; theta++) */
7169: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7170: /* printf("\n "); */
7171: /* } */
7172: /* } */
7173: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7174: /* printf("\n gradg %d ",(int)age); */
7175: /* for(j=1; j<=nlstate;j++){ */
7176: /* printf("%d ",j); */
7177: /* for(theta=1; theta <=npar; theta++) */
7178: /* printf("%d %lf ",theta,gradg[theta][j]); */
7179: /* printf("\n "); */
7180: /* } */
7181: /* } */
1.126 brouard 7182:
7183: for(i=1;i<=nlstate;i++)
7184: varpl[i][(int)age] =0.;
1.209 brouard 7185: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 7186: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7187: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 7188: }else{
1.268 brouard 7189: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7190: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 7191: }
1.126 brouard 7192: for(i=1;i<=nlstate;i++)
7193: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7194:
7195: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 7196: if(nresult >=1)
7197: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 7198: for(i=1; i<=nlstate;i++){
1.126 brouard 7199: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 7200: /* for(j=1;j<=nlstate;j++) */
7201: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
7202: }
1.126 brouard 7203: fprintf(ficresvpl,"\n");
7204: free_vector(gp,1,nlstate);
7205: free_vector(gm,1,nlstate);
1.208 brouard 7206: free_matrix(mgm,1,npar,1,nlstate);
7207: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 7208: free_matrix(gradg,1,npar,1,nlstate);
7209: free_matrix(trgradg,1,nlstate,1,npar);
7210: } /* End age */
7211:
7212: free_vector(xp,1,npar);
7213: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 7214: free_matrix(dnewmpar,1,nlstate,1,nlstate);
7215:
7216: }
7217:
7218:
7219: /************ Variance of backprevalence limit ******************/
1.269 brouard 7220: 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 7221: {
7222: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
7223: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
7224:
7225: double **dnewmpar,**doldm;
7226: int i, j, nhstepm, hstepm;
7227: double *xp;
7228: double *gp, *gm;
7229: double **gradg, **trgradg;
7230: double **mgm, **mgp;
7231: double age,agelim;
7232: int theta;
7233:
7234: pstamp(ficresvbl);
7235: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
7236: fprintf(ficresvbl,"# Age ");
7237: if(nresult >=1)
7238: fprintf(ficresvbl," Result# ");
7239: for(i=1; i<=nlstate;i++)
7240: fprintf(ficresvbl," %1d-%1d",i,i);
7241: fprintf(ficresvbl,"\n");
7242:
7243: xp=vector(1,npar);
7244: dnewmpar=matrix(1,nlstate,1,npar);
7245: doldm=matrix(1,nlstate,1,nlstate);
7246:
7247: hstepm=1*YEARM; /* Every year of age */
7248: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
7249: agelim = AGEINF;
7250: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
7251: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
7252: if (stepm >= YEARM) hstepm=1;
7253: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
7254: gradg=matrix(1,npar,1,nlstate);
7255: mgp=matrix(1,npar,1,nlstate);
7256: mgm=matrix(1,npar,1,nlstate);
7257: gp=vector(1,nlstate);
7258: gm=vector(1,nlstate);
7259:
7260: for(theta=1; theta <=npar; theta++){
7261: for(i=1; i<=npar; i++){ /* Computes gradient */
7262: xp[i] = x[i] + (i==theta ?delti[theta]:0);
7263: }
7264: if(mobilavproj > 0 )
7265: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7266: else
7267: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7268: for(i=1;i<=nlstate;i++){
7269: gp[i] = bprlim[i][i];
7270: mgp[theta][i] = bprlim[i][i];
7271: }
7272: for(i=1; i<=npar; i++) /* Computes gradient */
7273: xp[i] = x[i] - (i==theta ?delti[theta]:0);
7274: if(mobilavproj > 0 )
7275: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7276: else
7277: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7278: for(i=1;i<=nlstate;i++){
7279: gm[i] = bprlim[i][i];
7280: mgm[theta][i] = bprlim[i][i];
7281: }
7282: for(i=1;i<=nlstate;i++)
7283: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
7284: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
7285: } /* End theta */
7286:
7287: trgradg =matrix(1,nlstate,1,npar);
7288:
7289: for(j=1; j<=nlstate;j++)
7290: for(theta=1; theta <=npar; theta++)
7291: trgradg[j][theta]=gradg[theta][j];
7292: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7293: /* printf("\nmgm mgp %d ",(int)age); */
7294: /* for(j=1; j<=nlstate;j++){ */
7295: /* printf(" %d ",j); */
7296: /* for(theta=1; theta <=npar; theta++) */
7297: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7298: /* printf("\n "); */
7299: /* } */
7300: /* } */
7301: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7302: /* printf("\n gradg %d ",(int)age); */
7303: /* for(j=1; j<=nlstate;j++){ */
7304: /* printf("%d ",j); */
7305: /* for(theta=1; theta <=npar; theta++) */
7306: /* printf("%d %lf ",theta,gradg[theta][j]); */
7307: /* printf("\n "); */
7308: /* } */
7309: /* } */
7310:
7311: for(i=1;i<=nlstate;i++)
7312: varbpl[i][(int)age] =0.;
7313: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
7314: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7315: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7316: }else{
7317: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7318: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7319: }
7320: for(i=1;i<=nlstate;i++)
7321: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7322:
7323: fprintf(ficresvbl,"%.0f ",age );
7324: if(nresult >=1)
7325: fprintf(ficresvbl,"%d ",nres );
7326: for(i=1; i<=nlstate;i++)
7327: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
7328: fprintf(ficresvbl,"\n");
7329: free_vector(gp,1,nlstate);
7330: free_vector(gm,1,nlstate);
7331: free_matrix(mgm,1,npar,1,nlstate);
7332: free_matrix(mgp,1,npar,1,nlstate);
7333: free_matrix(gradg,1,npar,1,nlstate);
7334: free_matrix(trgradg,1,nlstate,1,npar);
7335: } /* End age */
7336:
7337: free_vector(xp,1,npar);
7338: free_matrix(doldm,1,nlstate,1,npar);
7339: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 7340:
7341: }
7342:
7343: /************ Variance of one-step probabilities ******************/
7344: 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 7345: {
7346: int i, j=0, k1, l1, tj;
7347: int k2, l2, j1, z1;
7348: int k=0, l;
7349: int first=1, first1, first2;
1.326 brouard 7350: int nres=0; /* New */
1.222 brouard 7351: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
7352: double **dnewm,**doldm;
7353: double *xp;
7354: double *gp, *gm;
7355: double **gradg, **trgradg;
7356: double **mu;
7357: double age, cov[NCOVMAX+1];
7358: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
7359: int theta;
7360: char fileresprob[FILENAMELENGTH];
7361: char fileresprobcov[FILENAMELENGTH];
7362: char fileresprobcor[FILENAMELENGTH];
7363: double ***varpij;
7364:
7365: strcpy(fileresprob,"PROB_");
7366: strcat(fileresprob,fileres);
7367: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
7368: printf("Problem with resultfile: %s\n", fileresprob);
7369: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
7370: }
7371: strcpy(fileresprobcov,"PROBCOV_");
7372: strcat(fileresprobcov,fileresu);
7373: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
7374: printf("Problem with resultfile: %s\n", fileresprobcov);
7375: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
7376: }
7377: strcpy(fileresprobcor,"PROBCOR_");
7378: strcat(fileresprobcor,fileresu);
7379: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
7380: printf("Problem with resultfile: %s\n", fileresprobcor);
7381: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
7382: }
7383: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7384: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7385: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7386: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7387: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7388: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7389: pstamp(ficresprob);
7390: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
7391: fprintf(ficresprob,"# Age");
7392: pstamp(ficresprobcov);
7393: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
7394: fprintf(ficresprobcov,"# Age");
7395: pstamp(ficresprobcor);
7396: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
7397: fprintf(ficresprobcor,"# Age");
1.126 brouard 7398:
7399:
1.222 brouard 7400: for(i=1; i<=nlstate;i++)
7401: for(j=1; j<=(nlstate+ndeath);j++){
7402: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
7403: fprintf(ficresprobcov," p%1d-%1d ",i,j);
7404: fprintf(ficresprobcor," p%1d-%1d ",i,j);
7405: }
7406: /* fprintf(ficresprob,"\n");
7407: fprintf(ficresprobcov,"\n");
7408: fprintf(ficresprobcor,"\n");
7409: */
7410: xp=vector(1,npar);
7411: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7412: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7413: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
7414: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
7415: first=1;
7416: fprintf(ficgp,"\n# Routine varprob");
7417: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
7418: fprintf(fichtm,"\n");
7419:
1.288 brouard 7420: 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 7421: 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);
7422: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 7423: and drawn. It helps understanding how is the covariance between two incidences.\
7424: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 7425: 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 7426: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
7427: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
7428: standard deviations wide on each axis. <br>\
7429: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
7430: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
7431: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
7432:
1.222 brouard 7433: cov[1]=1;
7434: /* tj=cptcoveff; */
1.225 brouard 7435: tj = (int) pow(2,cptcoveff);
1.222 brouard 7436: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
7437: j1=0;
1.332 brouard 7438:
7439: for(nres=1;nres <=nresult; nres++){ /* For each resultline */
7440: for(j1=1; j1<=tj;j1++){ /* For any combination of dummy covariates, fixed and varying */
1.342 brouard 7441: /* 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 7442: if(tj != 1 && TKresult[nres]!= j1)
7443: continue;
7444:
7445: /* for(j1=1; j1<=tj;j1++){ /\* For each valid combination of covariates or only once*\/ */
7446: /* for(nres=1;nres <=1; nres++){ /\* For each resultline *\/ */
7447: /* /\* for(nres=1;nres <=nresult; nres++){ /\\* For each resultline *\\/ *\/ */
1.222 brouard 7448: if (cptcovn>0) {
1.334 brouard 7449: fprintf(ficresprob, "\n#********** Variable ");
7450: fprintf(ficresprobcov, "\n#********** Variable ");
7451: fprintf(ficgp, "\n#********** Variable ");
7452: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
7453: fprintf(ficresprobcor, "\n#********** Variable ");
7454:
7455: /* Including quantitative variables of the resultline to be done */
7456: for (z1=1; z1<=cptcovs; z1++){ /* Loop on each variable of this resultline */
1.343 ! brouard 7457: /* printf("Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model); */
1.338 brouard 7458: fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model);
7459: /* 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 7460: if(Dummy[modelresult[nres][z1]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to z1 in resultline */
7461: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed referenced to model equation */
7462: 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 */
7463: 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 */
7464: 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 */
7465: 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 */
7466: 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 */
7467: fprintf(ficresprob,"fixed ");
7468: fprintf(ficresprobcov,"fixed ");
7469: fprintf(ficgp,"fixed ");
7470: fprintf(fichtmcov,"fixed ");
7471: fprintf(ficresprobcor,"fixed ");
7472: }else{
7473: fprintf(ficresprob,"varyi ");
7474: fprintf(ficresprobcov,"varyi ");
7475: fprintf(ficgp,"varyi ");
7476: fprintf(fichtmcov,"varyi ");
7477: fprintf(ficresprobcor,"varyi ");
7478: }
7479: }else if(Dummy[modelresult[nres][z1]]==1){ /* Quanti variable */
7480: /* For each selected (single) quantitative value */
1.337 brouard 7481: fprintf(ficresprob," V%d=%lg ",Tvqresult[nres][z1],Tqresult[nres][z1]);
1.334 brouard 7482: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed */
7483: fprintf(ficresprob,"fixed ");
7484: fprintf(ficresprobcov,"fixed ");
7485: fprintf(ficgp,"fixed ");
7486: fprintf(fichtmcov,"fixed ");
7487: fprintf(ficresprobcor,"fixed ");
7488: }else{
7489: fprintf(ficresprob,"varyi ");
7490: fprintf(ficresprobcov,"varyi ");
7491: fprintf(ficgp,"varyi ");
7492: fprintf(fichtmcov,"varyi ");
7493: fprintf(ficresprobcor,"varyi ");
7494: }
7495: }else{
7496: 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 */
7497: 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 */
7498: exit(1);
7499: }
7500: } /* End loop on variable of this resultline */
7501: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]); */
1.222 brouard 7502: fprintf(ficresprob, "**********\n#\n");
7503: fprintf(ficresprobcov, "**********\n#\n");
7504: fprintf(ficgp, "**********\n#\n");
7505: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
7506: fprintf(ficresprobcor, "**********\n#");
7507: if(invalidvarcomb[j1]){
7508: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
7509: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
7510: continue;
7511: }
7512: }
7513: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
7514: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7515: gp=vector(1,(nlstate)*(nlstate+ndeath));
7516: gm=vector(1,(nlstate)*(nlstate+ndeath));
1.334 brouard 7517: for (age=bage; age<=fage; age ++){ /* Fo each age we feed the model equation with covariates, using precov as in hpxij() ? */
1.222 brouard 7518: cov[2]=age;
7519: if(nagesqr==1)
7520: cov[3]= age*age;
1.334 brouard 7521: /* New code end of combination but for each resultline */
7522: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
7523: if(Typevar[k1]==1){ /* A product with age */
7524: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.326 brouard 7525: }else{
1.334 brouard 7526: cov[2+nagesqr+k1]=precov[nres][k1];
1.326 brouard 7527: }
1.334 brouard 7528: }/* End of loop on model equation */
7529: /* Old code */
7530: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
7531: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; *\/ */
7532: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
7533: /* /\* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates *\/ */
7534: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TnsdVar[TvarsD[k]])]; */
7535: /* /\*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*\//\* j1 1 2 3 4 */
7536: /* * 1 1 1 1 1 */
7537: /* * 2 2 1 1 1 */
7538: /* * 3 1 2 1 1 */
7539: /* *\/ */
7540: /* /\* nbcode[1][1]=0 nbcode[1][2]=1;*\/ */
7541: /* } */
7542: /* /\* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
7543: /* /\* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] *\/ */
7544: /* /\*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; *\/ */
7545: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
7546: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
7547: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,TnsdVar[Tvar[Tage[k]]])]*cov[2]; */
7548: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7549: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
7550: /* 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]); */
7551: /* /\* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\\* Using the mean of quantitative variable Tvar[Tage[k]] /\\* Tqresult[nres][k]; *\\/ *\/ */
7552: /* /\* exit(1); *\/ */
7553: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
7554: /* } */
7555: /* /\* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7556: /* } */
7557: /* for (k=1; k<=cptcovprod;k++){/\* For product without age *\/ */
7558: /* if(Dummy[Tvard[k][1]]==0){ */
7559: /* if(Dummy[Tvard[k][2]]==0){ */
7560: /* 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]])]; */
7561: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7562: /* }else{ /\* Should we use the mean of the quantitative variables? *\/ */
7563: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * Tqresult[nres][resultmodel[nres][k]]; */
7564: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
7565: /* } */
7566: /* }else{ */
7567: /* if(Dummy[Tvard[k][2]]==0){ */
7568: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][TnsdVar[Tvard[k][1]]]; */
7569: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
7570: /* }else{ */
7571: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][TnsdVar[Tvard[k][1]]]* Tqinvresult[nres][TnsdVar[Tvard[k][2]]]; */
7572: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
7573: /* } */
7574: /* } */
7575: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7576: /* } */
1.326 brouard 7577: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 7578: for(theta=1; theta <=npar; theta++){
7579: for(i=1; i<=npar; i++)
7580: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7581:
1.222 brouard 7582: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7583:
1.222 brouard 7584: k=0;
7585: for(i=1; i<= (nlstate); i++){
7586: for(j=1; j<=(nlstate+ndeath);j++){
7587: k=k+1;
7588: gp[k]=pmmij[i][j];
7589: }
7590: }
1.220 brouard 7591:
1.222 brouard 7592: for(i=1; i<=npar; i++)
7593: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7594:
1.222 brouard 7595: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7596: k=0;
7597: for(i=1; i<=(nlstate); i++){
7598: for(j=1; j<=(nlstate+ndeath);j++){
7599: k=k+1;
7600: gm[k]=pmmij[i][j];
7601: }
7602: }
1.220 brouard 7603:
1.222 brouard 7604: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7605: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7606: }
1.126 brouard 7607:
1.222 brouard 7608: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7609: for(theta=1; theta <=npar; theta++)
7610: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7611:
1.222 brouard 7612: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7613: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7614:
1.222 brouard 7615: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7616:
1.222 brouard 7617: k=0;
7618: for(i=1; i<=(nlstate); i++){
7619: for(j=1; j<=(nlstate+ndeath);j++){
7620: k=k+1;
7621: mu[k][(int) age]=pmmij[i][j];
7622: }
7623: }
7624: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7625: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7626: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7627:
1.222 brouard 7628: /*printf("\n%d ",(int)age);
7629: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7630: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7631: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7632: }*/
1.220 brouard 7633:
1.222 brouard 7634: fprintf(ficresprob,"\n%d ",(int)age);
7635: fprintf(ficresprobcov,"\n%d ",(int)age);
7636: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7637:
1.222 brouard 7638: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7639: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7640: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7641: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7642: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7643: }
7644: i=0;
7645: for (k=1; k<=(nlstate);k++){
7646: for (l=1; l<=(nlstate+ndeath);l++){
7647: i++;
7648: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7649: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7650: for (j=1; j<=i;j++){
7651: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7652: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7653: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7654: }
7655: }
7656: }/* end of loop for state */
7657: } /* end of loop for age */
7658: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7659: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7660: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7661: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7662:
7663: /* Confidence intervalle of pij */
7664: /*
7665: fprintf(ficgp,"\nunset parametric;unset label");
7666: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7667: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7668: 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);
7669: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7670: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7671: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7672: */
7673:
7674: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7675: first1=1;first2=2;
7676: for (k2=1; k2<=(nlstate);k2++){
7677: for (l2=1; l2<=(nlstate+ndeath);l2++){
7678: if(l2==k2) continue;
7679: j=(k2-1)*(nlstate+ndeath)+l2;
7680: for (k1=1; k1<=(nlstate);k1++){
7681: for (l1=1; l1<=(nlstate+ndeath);l1++){
7682: if(l1==k1) continue;
7683: i=(k1-1)*(nlstate+ndeath)+l1;
7684: if(i<=j) continue;
7685: for (age=bage; age<=fage; age ++){
7686: if ((int)age %5==0){
7687: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7688: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7689: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7690: mu1=mu[i][(int) age]/stepm*YEARM ;
7691: mu2=mu[j][(int) age]/stepm*YEARM;
7692: c12=cv12/sqrt(v1*v2);
7693: /* Computing eigen value of matrix of covariance */
7694: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7695: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7696: if ((lc2 <0) || (lc1 <0) ){
7697: if(first2==1){
7698: first1=0;
7699: 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);
7700: }
7701: 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);
7702: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7703: /* lc2=fabs(lc2); */
7704: }
1.220 brouard 7705:
1.222 brouard 7706: /* Eigen vectors */
1.280 brouard 7707: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7708: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7709: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7710: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7711: }else
7712: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7713: /*v21=sqrt(1.-v11*v11); *//* error */
7714: v21=(lc1-v1)/cv12*v11;
7715: v12=-v21;
7716: v22=v11;
7717: tnalp=v21/v11;
7718: if(first1==1){
7719: first1=0;
7720: 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);
7721: }
7722: 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);
7723: /*printf(fignu*/
7724: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7725: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7726: if(first==1){
7727: first=0;
7728: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7729: fprintf(ficgp,"\nset parametric;unset label");
7730: 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);
7731: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7732: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7733: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7734: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7735: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7736: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7737: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7738: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7739: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7740: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7741: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7742: 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 7743: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7744: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7745: }else{
7746: first=0;
7747: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7748: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7749: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7750: 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 7751: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7752: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7753: }/* if first */
7754: } /* age mod 5 */
7755: } /* end loop age */
7756: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7757: first=1;
7758: } /*l12 */
7759: } /* k12 */
7760: } /*l1 */
7761: }/* k1 */
1.332 brouard 7762: } /* loop on combination of covariates j1 */
1.326 brouard 7763: } /* loop on nres */
1.222 brouard 7764: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7765: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7766: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7767: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7768: free_vector(xp,1,npar);
7769: fclose(ficresprob);
7770: fclose(ficresprobcov);
7771: fclose(ficresprobcor);
7772: fflush(ficgp);
7773: fflush(fichtmcov);
7774: }
1.126 brouard 7775:
7776:
7777: /******************* Printing html file ***********/
1.201 brouard 7778: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7779: int lastpass, int stepm, int weightopt, char model[],\
7780: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7781: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7782: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7783: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7784: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7785: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7786: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7787: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7788: </ul>");
1.319 brouard 7789: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7790: /* </ul>", model); */
1.214 brouard 7791: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7792: 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",
7793: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
1.332 brouard 7794: 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 7795: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7796: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7797: fprintf(fichtm,"\
7798: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7799: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7800: fprintf(fichtm,"\
1.217 brouard 7801: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7802: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7803: fprintf(fichtm,"\
1.288 brouard 7804: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7805: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7806: fprintf(fichtm,"\
1.288 brouard 7807: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7808: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7809: fprintf(fichtm,"\
1.211 brouard 7810: - (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 7811: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7812: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7813: if(prevfcast==1){
7814: fprintf(fichtm,"\
7815: - Prevalence projections by age and states: \
1.201 brouard 7816: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7817: }
1.126 brouard 7818:
7819:
1.225 brouard 7820: m=pow(2,cptcoveff);
1.222 brouard 7821: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7822:
1.317 brouard 7823: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7824:
7825: jj1=0;
7826:
7827: fprintf(fichtm," \n<ul>");
1.337 brouard 7828: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7829: /* k1=nres; */
1.338 brouard 7830: k1=TKresult[nres];
7831: if(TKresult[nres]==0)k1=1; /* To be checked for no result */
1.337 brouard 7832: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
7833: /* if(m != 1 && TKresult[nres]!= k1) */
7834: /* continue; */
1.264 brouard 7835: jj1++;
7836: if (cptcovn > 0) {
7837: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
1.337 brouard 7838: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
7839: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7840: }
1.337 brouard 7841: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
7842: /* fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
7843: /* } */
7844: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7845: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7846: /* } */
1.264 brouard 7847: fprintf(fichtm,"\">");
7848:
7849: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7850: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 7851: for (cpt=1; cpt<=cptcovs;cpt++){
7852: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7853: }
1.337 brouard 7854: /* fprintf(fichtm,"************ Results for covariates"); */
7855: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
7856: /* fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
7857: /* } */
7858: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7859: /* fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7860: /* } */
1.264 brouard 7861: if(invalidvarcomb[k1]){
7862: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7863: continue;
7864: }
7865: fprintf(fichtm,"</a></li>");
7866: } /* cptcovn >0 */
7867: }
1.317 brouard 7868: fprintf(fichtm," \n</ul>");
1.264 brouard 7869:
1.222 brouard 7870: jj1=0;
1.237 brouard 7871:
1.337 brouard 7872: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7873: /* k1=nres; */
1.338 brouard 7874: k1=TKresult[nres];
7875: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 7876: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
7877: /* if(m != 1 && TKresult[nres]!= k1) */
7878: /* continue; */
1.220 brouard 7879:
1.222 brouard 7880: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7881: jj1++;
7882: if (cptcovn > 0) {
1.264 brouard 7883: fprintf(fichtm,"\n<p><a name=\"rescov");
1.337 brouard 7884: for (cpt=1; cpt<=cptcovs;cpt++){
7885: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7886: }
1.337 brouard 7887: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7888: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7889: /* } */
1.264 brouard 7890: fprintf(fichtm,"\"</a>");
7891:
1.222 brouard 7892: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 7893: for (cpt=1; cpt<=cptcovs;cpt++){
7894: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
7895: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 7896: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7897: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7898: }
1.230 brouard 7899: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.338 brouard 7900: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7901: if(invalidvarcomb[k1]){
7902: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7903: printf("\nCombination (%d) ignored because no cases \n",k1);
7904: continue;
7905: }
7906: }
7907: /* aij, bij */
1.259 brouard 7908: 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 7909: <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 7910: /* Pij */
1.241 brouard 7911: 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> \
7912: <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 7913: /* Quasi-incidences */
7914: 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 7915: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7916: 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 7917: 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> \
7918: <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 7919: /* Survival functions (period) in state j */
7920: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7921: 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);
7922: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7923: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7924: }
7925: /* State specific survival functions (period) */
7926: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7927: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7928: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.329 brouard 7929: <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);
7930: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7931: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7932: }
1.288 brouard 7933: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7934: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7935: 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 7936: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
1.329 brouard 7937: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7938: }
1.296 brouard 7939: if(prevbcast==1){
1.288 brouard 7940: /* Backward prevalence in each health state */
1.222 brouard 7941: for(cpt=1; cpt<=nlstate;cpt++){
1.338 brouard 7942: 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);
7943: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJB_"),subdirf2(optionfilefiname,"PIJB_"));
7944: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.222 brouard 7945: }
1.217 brouard 7946: }
1.222 brouard 7947: if(prevfcast==1){
1.288 brouard 7948: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7949: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7950: 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);
7951: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7952: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7953: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7954: }
7955: }
1.296 brouard 7956: if(prevbcast==1){
1.268 brouard 7957: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7958: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7959: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7960: 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 \
7961: 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 7962: 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);
7963: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7964: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7965: }
7966: }
1.220 brouard 7967:
1.222 brouard 7968: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7969: 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);
7970: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7971: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7972: }
7973: /* } /\* end i1 *\/ */
1.337 brouard 7974: }/* End k1=nres */
1.222 brouard 7975: fprintf(fichtm,"</ul>");
1.126 brouard 7976:
1.222 brouard 7977: fprintf(fichtm,"\
1.126 brouard 7978: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7979: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7980: - 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 7981: But because parameters are usually highly correlated (a higher incidence of disability \
7982: and a higher incidence of recovery can give very close observed transition) it might \
7983: be very useful to look not only at linear confidence intervals estimated from the \
7984: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7985: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7986: covariance matrix of the one-step probabilities. \
7987: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7988:
1.222 brouard 7989: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7990: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7991: fprintf(fichtm,"\
1.126 brouard 7992: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7993: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7994:
1.222 brouard 7995: fprintf(fichtm,"\
1.126 brouard 7996: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7997: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7998: fprintf(fichtm,"\
1.126 brouard 7999: - 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): \
8000: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 8001: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 8002: fprintf(fichtm,"\
1.126 brouard 8003: - (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): \
8004: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 8005: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 8006: fprintf(fichtm,"\
1.288 brouard 8007: - 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 8008: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
8009: fprintf(fichtm,"\
1.128 brouard 8010: - 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 8011: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
8012: fprintf(fichtm,"\
1.288 brouard 8013: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 8014: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 8015:
8016: /* if(popforecast==1) fprintf(fichtm,"\n */
8017: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
8018: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
8019: /* <br>",fileres,fileres,fileres,fileres); */
8020: /* else */
1.338 brouard 8021: /* 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 8022: fflush(fichtm);
1.126 brouard 8023:
1.225 brouard 8024: m=pow(2,cptcoveff);
1.222 brouard 8025: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 8026:
1.317 brouard 8027: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
8028:
8029: jj1=0;
8030:
8031: fprintf(fichtm," \n<ul>");
1.337 brouard 8032: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8033: /* k1=nres; */
1.338 brouard 8034: k1=TKresult[nres];
1.337 brouard 8035: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
8036: /* if(m != 1 && TKresult[nres]!= k1) */
8037: /* continue; */
1.317 brouard 8038: jj1++;
8039: if (cptcovn > 0) {
8040: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
1.337 brouard 8041: for (cpt=1; cpt<=cptcovs;cpt++){
8042: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8043: }
8044: fprintf(fichtm,"\">");
8045:
8046: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
8047: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 8048: for (cpt=1; cpt<=cptcovs;cpt++){
8049: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8050: }
8051: if(invalidvarcomb[k1]){
8052: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
8053: continue;
8054: }
8055: fprintf(fichtm,"</a></li>");
8056: } /* cptcovn >0 */
1.337 brouard 8057: } /* End nres */
1.317 brouard 8058: fprintf(fichtm," \n</ul>");
8059:
1.222 brouard 8060: jj1=0;
1.237 brouard 8061:
1.241 brouard 8062: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8063: /* k1=nres; */
1.338 brouard 8064: k1=TKresult[nres];
8065: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8066: /* for(k1=1; k1<=m;k1++){ */
8067: /* if(m != 1 && TKresult[nres]!= k1) */
8068: /* continue; */
1.222 brouard 8069: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
8070: jj1++;
1.126 brouard 8071: if (cptcovn > 0) {
1.317 brouard 8072: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
1.337 brouard 8073: for (cpt=1; cpt<=cptcovs;cpt++){
8074: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8075: }
8076: fprintf(fichtm,"\"</a>");
8077:
1.126 brouard 8078: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 8079: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcoveff number of variables */
8080: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
8081: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 8082: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 8083: }
1.237 brouard 8084:
1.338 brouard 8085: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 8086:
1.222 brouard 8087: if(invalidvarcomb[k1]){
8088: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
8089: continue;
8090: }
1.337 brouard 8091: } /* If cptcovn >0 */
1.126 brouard 8092: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 8093: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 8094: 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);
8095: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
8096: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 8097: }
8098: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 8099: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 8100: true period expectancies (those weighted with period prevalences are also\
8101: drawn in addition to the population based expectancies computed using\
1.314 brouard 8102: 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);
8103: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
8104: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 8105: /* } /\* end i1 *\/ */
1.241 brouard 8106: }/* End nres */
1.222 brouard 8107: fprintf(fichtm,"</ul>");
8108: fflush(fichtm);
1.126 brouard 8109: }
8110:
8111: /******************* Gnuplot file **************/
1.296 brouard 8112: 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 8113:
8114: char dirfileres[132],optfileres[132];
1.264 brouard 8115: char gplotcondition[132], gplotlabel[132];
1.343 ! brouard 8116: 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 8117: int lv=0, vlv=0, kl=0;
1.130 brouard 8118: int ng=0;
1.201 brouard 8119: int vpopbased;
1.223 brouard 8120: int ioffset; /* variable offset for columns */
1.270 brouard 8121: int iyearc=1; /* variable column for year of projection */
8122: int iagec=1; /* variable column for age of projection */
1.235 brouard 8123: int nres=0; /* Index of resultline */
1.266 brouard 8124: int istart=1; /* For starting graphs in projections */
1.219 brouard 8125:
1.126 brouard 8126: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
8127: /* printf("Problem with file %s",optionfilegnuplot); */
8128: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
8129: /* } */
8130:
8131: /*#ifdef windows */
8132: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 8133: /*#endif */
1.225 brouard 8134: m=pow(2,cptcoveff);
1.126 brouard 8135:
1.274 brouard 8136: /* diagram of the model */
8137: fprintf(ficgp,"\n#Diagram of the model \n");
8138: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
8139: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
8140: 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);
8141:
1.343 ! brouard 8142: 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 8143: fprintf(ficgp,"\n#show arrow\nunset label\n");
8144: 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);
8145: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
8146: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
8147: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
8148: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
8149:
1.202 brouard 8150: /* Contribution to likelihood */
8151: /* Plot the probability implied in the likelihood */
1.223 brouard 8152: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
8153: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
8154: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
8155: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 8156: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 8157: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
8158: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 8159: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
8160: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
8161: 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));
8162: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
8163: 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));
8164: for (i=1; i<= nlstate ; i ++) {
8165: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
8166: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
8167: 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);
8168: for (j=2; j<= nlstate+ndeath ; j ++) {
8169: 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);
8170: }
8171: fprintf(ficgp,";\nset out; unset ylabel;\n");
8172: }
8173: /* 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 */
8174: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
8175: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
8176: fprintf(ficgp,"\nset out;unset log\n");
8177: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 8178:
1.343 ! brouard 8179: /* Plot the probability implied in the likelihood by covariate value */
! 8180: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
! 8181: /* if(debugILK==1){ */
! 8182: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
! 8183: kvar=Tvar[TvarFind[kf]]; /* variable */
! 8184: k=18+Tvar[TvarFind[kf]];/*offset because there are 18 columns in the ILK_ file */
! 8185: for (i=1; i<= nlstate ; i ++) {
! 8186: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
! 8187: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
! 8188: 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);
! 8189: for (j=2; j<= nlstate+ndeath ; j ++) {
! 8190: 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);
! 8191: }
! 8192: fprintf(ficgp,";\nset out; unset ylabel;\n");
! 8193: }
! 8194: } /* End of each covariate dummy */
! 8195: for(ncovv=1, iposold=0, kk=0; ncovv <= ncovvt ; ncovv++){
! 8196: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
! 8197: * kmodel = 1 2 3 4 5 6 7 8 9
! 8198: * varying 1 2 3 4 5
! 8199: * ncovv 1 2 3 4 5 6 7 8
! 8200: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
! 8201: * TvarVVind[ncovv]=kmodel 2 3 7 7 8 8 9 9
! 8202: * TvarFind[kmodel] 1 0 0 0 0 0 0 0 0
! 8203: * kdata ncovcol=[V1 V2] nqv=0 ntv=[V3 V4] nqtv=V5
! 8204: * Dummy[kmodel] 0 0 1 2 2 3 1 1 1
! 8205: */
! 8206: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
! 8207: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
! 8208: /* 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]); */
! 8209: if(ipos!=iposold){ /* Not a product or first of a product */
! 8210: /* printf(" %d",ipos); */
! 8211: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
! 8212: /* printf(" DebugILK ficgp suite ipos=%d != iposold=%d\n", ipos, iposold); */
! 8213: kk++; /* Position of the ncovv column in ILK_ */
! 8214: k=18+ncovf+kk; /*offset because there are 18 columns in the ILK_ file plus ncovf fixed covariate */
! 8215: 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) */
! 8216: for (i=1; i<= nlstate ; i ++) {
! 8217: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
! 8218: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
! 8219:
! 8220: if(gnuplotversion >=5.2){ /* Former gnuplot versions do not have variable pointsize!! */
! 8221: /* printf("DebugILK gnuplotversion=%g >=5.2\n",gnuplotversion); */
! 8222: 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);
! 8223: for (j=2; j<= nlstate+ndeath ; j ++) {
! 8224: 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);
! 8225: }
! 8226: }else{
! 8227: /* printf("DebugILK gnuplotversion=%g <5.2\n",gnuplotversion); */
! 8228: 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);
! 8229: for (j=2; j<= nlstate+ndeath ; j ++) {
! 8230: 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);
! 8231: }
! 8232: }
! 8233: fprintf(ficgp,";\nset out; unset ylabel;\n");
! 8234: }
! 8235: }/* End if dummy varying */
! 8236: }else{ /*Product */
! 8237: /* printf("*"); */
! 8238: /* fprintf(ficresilk,"*"); */
! 8239: }
! 8240: iposold=ipos;
! 8241: } /* For each time varying covariate */
! 8242: /* } /\* debugILK==1 *\/ */
! 8243: /* 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 */
! 8244: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
! 8245: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
! 8246: fprintf(ficgp,"\nset out;unset log\n");
! 8247: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
! 8248:
! 8249:
! 8250:
1.126 brouard 8251: strcpy(dirfileres,optionfilefiname);
8252: strcpy(optfileres,"vpl");
1.223 brouard 8253: /* 1eme*/
1.238 brouard 8254: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
1.337 brouard 8255: /* for (k1=1; k1<= m ; k1 ++){ /\* For each valid combination of covariate *\/ */
1.236 brouard 8256: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8257: k1=TKresult[nres];
1.338 brouard 8258: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.238 brouard 8259: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.337 brouard 8260: /* if(m != 1 && TKresult[nres]!= k1) */
8261: /* continue; */
1.238 brouard 8262: /* We are interested in selected combination by the resultline */
1.246 brouard 8263: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 8264: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 8265: strcpy(gplotlabel,"(");
1.337 brouard 8266: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8267: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8268: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8269:
8270: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate k get corresponding value lv for combination k1 *\/ */
8271: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the value of the covariate corresponding to k1 combination *\\/ *\/ */
8272: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8273: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8274: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8275: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8276: /* vlv= nbcode[Tvaraff[k]][lv]; /\* vlv is the value of the covariate lv, 0 or 1 *\/ */
8277: /* /\* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv *\/ */
8278: /* /\* printf(" V%d=%d ",Tvaraff[k],vlv); *\/ */
8279: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8280: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8281: /* } */
8282: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8283: /* /\* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); *\/ */
8284: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8285: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.264 brouard 8286: }
8287: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 8288: /* printf("\n#\n"); */
1.238 brouard 8289: fprintf(ficgp,"\n#\n");
8290: if(invalidvarcomb[k1]){
1.260 brouard 8291: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 8292: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8293: continue;
8294: }
1.235 brouard 8295:
1.241 brouard 8296: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
8297: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 8298: /* 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 8299: fprintf(ficgp,"set title \"Alive state %d %s model=1+age+%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 8300: 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);
8301: /* 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); */
8302: /* k1-1 error should be nres-1*/
1.238 brouard 8303: for (i=1; i<= nlstate ; i ++) {
8304: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8305: else fprintf(ficgp," %%*lf (%%*lf)");
8306: }
1.288 brouard 8307: 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 8308: for (i=1; i<= nlstate ; i ++) {
8309: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8310: else fprintf(ficgp," %%*lf (%%*lf)");
8311: }
1.260 brouard 8312: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 8313: for (i=1; i<= nlstate ; i ++) {
8314: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8315: else fprintf(ficgp," %%*lf (%%*lf)");
8316: }
1.265 brouard 8317: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
8318:
8319: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
8320: if(cptcoveff ==0){
1.271 brouard 8321: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 8322: }else{
8323: kl=0;
8324: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8325: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8326: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.265 brouard 8327: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8328: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8329: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8330: vlv= nbcode[Tvaraff[k]][lv];
8331: kl++;
8332: /* 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 *\/ */
8333: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8334: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8335: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8336: if(k==cptcoveff){
8337: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
8338: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
8339: }else{
8340: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
8341: kl++;
8342: }
8343: } /* end covariate */
8344: } /* end if no covariate */
8345:
1.296 brouard 8346: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 8347: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 8348: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 8349: if(cptcoveff ==0){
1.245 brouard 8350: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 8351: }else{
8352: kl=0;
8353: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8354: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8355: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 8356: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8357: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8358: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8359: /* vlv= nbcode[Tvaraff[k]][lv]; */
8360: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 8361: kl++;
1.238 brouard 8362: /* 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 *\/ */
8363: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8364: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8365: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8366: if(k==cptcoveff){
1.245 brouard 8367: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 8368: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 8369: }else{
1.332 brouard 8370: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]);
1.238 brouard 8371: kl++;
8372: }
8373: } /* end covariate */
8374: } /* end if no covariate */
1.296 brouard 8375: if(prevbcast == 1){
1.268 brouard 8376: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
8377: /* k1-1 error should be nres-1*/
8378: for (i=1; i<= nlstate ; i ++) {
8379: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8380: else fprintf(ficgp," %%*lf (%%*lf)");
8381: }
1.271 brouard 8382: 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 8383: for (i=1; i<= nlstate ; i ++) {
8384: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8385: else fprintf(ficgp," %%*lf (%%*lf)");
8386: }
1.276 brouard 8387: 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 8388: for (i=1; i<= nlstate ; i ++) {
8389: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8390: else fprintf(ficgp," %%*lf (%%*lf)");
8391: }
1.274 brouard 8392: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 8393: } /* end if backprojcast */
1.296 brouard 8394: } /* end if prevbcast */
1.276 brouard 8395: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
8396: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 8397: } /* nres */
1.337 brouard 8398: /* } /\* k1 *\/ */
1.201 brouard 8399: } /* cpt */
1.235 brouard 8400:
8401:
1.126 brouard 8402: /*2 eme*/
1.337 brouard 8403: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8404: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8405: k1=TKresult[nres];
1.338 brouard 8406: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8407: /* if(m != 1 && TKresult[nres]!= k1) */
8408: /* continue; */
1.238 brouard 8409: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 8410: strcpy(gplotlabel,"(");
1.337 brouard 8411: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8412: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8413: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8414: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8415: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8416: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8417: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8418: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8419: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8420: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8421: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8422: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8423: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8424: /* } */
8425: /* /\* for(k=1; k <= ncovds; k++){ *\/ */
8426: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8427: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8428: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8429: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 8430: }
1.264 brouard 8431: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8432: fprintf(ficgp,"\n#\n");
1.223 brouard 8433: if(invalidvarcomb[k1]){
8434: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8435: continue;
8436: }
1.219 brouard 8437:
1.241 brouard 8438: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 8439: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 8440: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
8441: if(vpopbased==0){
1.238 brouard 8442: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 8443: }else
1.238 brouard 8444: fprintf(ficgp,"\nreplot ");
8445: for (i=1; i<= nlstate+1 ; i ++) {
8446: k=2*i;
1.261 brouard 8447: 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 8448: for (j=1; j<= nlstate+1 ; j ++) {
8449: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8450: else fprintf(ficgp," %%*lf (%%*lf)");
8451: }
8452: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
8453: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 8454: 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 8455: for (j=1; j<= nlstate+1 ; j ++) {
8456: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8457: else fprintf(ficgp," %%*lf (%%*lf)");
8458: }
8459: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 8460: 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 8461: for (j=1; j<= nlstate+1 ; j ++) {
8462: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8463: else fprintf(ficgp," %%*lf (%%*lf)");
8464: }
8465: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
8466: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
8467: } /* state */
8468: } /* vpopbased */
1.264 brouard 8469: 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 8470: } /* end nres */
1.337 brouard 8471: /* } /\* k1 end 2 eme*\/ */
1.238 brouard 8472:
8473:
8474: /*3eme*/
1.337 brouard 8475: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8476: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8477: k1=TKresult[nres];
1.338 brouard 8478: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8479: /* if(m != 1 && TKresult[nres]!= k1) */
8480: /* continue; */
1.238 brouard 8481:
1.332 brouard 8482: for (cpt=1; cpt<= nlstate ; cpt ++) { /* Fragile no verification of covariate values */
1.261 brouard 8483: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 8484: strcpy(gplotlabel,"(");
1.337 brouard 8485: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8486: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8487: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8488: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8489: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8490: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8491: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8492: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8493: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8494: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8495: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8496: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8497: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8498: /* } */
8499: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8500: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
8501: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
8502: }
1.264 brouard 8503: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8504: fprintf(ficgp,"\n#\n");
8505: if(invalidvarcomb[k1]){
8506: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8507: continue;
8508: }
8509:
8510: /* k=2+nlstate*(2*cpt-2); */
8511: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 8512: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 8513: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 8514: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 8515: 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 8516: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8517: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8518: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
8519: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8520: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8521: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 8522:
1.238 brouard 8523: */
8524: for (i=1; i< nlstate ; i ++) {
1.261 brouard 8525: 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 8526: /* 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 8527:
1.238 brouard 8528: }
1.261 brouard 8529: 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 8530: }
1.264 brouard 8531: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 8532: } /* end nres */
1.337 brouard 8533: /* } /\* end kl 3eme *\/ */
1.126 brouard 8534:
1.223 brouard 8535: /* 4eme */
1.201 brouard 8536: /* Survival functions (period) from state i in state j by initial state i */
1.337 brouard 8537: /* for (k1=1; k1<=m; k1++){ /\* For each covariate and each value *\/ */
1.238 brouard 8538: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8539: k1=TKresult[nres];
1.338 brouard 8540: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8541: /* if(m != 1 && TKresult[nres]!= k1) */
8542: /* continue; */
1.238 brouard 8543: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 8544: strcpy(gplotlabel,"(");
1.337 brouard 8545: fprintf(ficgp,"\n#\n#\n# Survival functions in state %d : 'LIJ_' files, cov=%d state=%d", cpt, k1, cpt);
8546: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8547: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8548: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8549: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8550: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8551: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8552: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8553: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8554: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8555: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8556: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8557: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8558: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8559: /* } */
8560: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8561: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8562: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8563: }
1.264 brouard 8564: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8565: fprintf(ficgp,"\n#\n");
8566: if(invalidvarcomb[k1]){
8567: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8568: continue;
1.223 brouard 8569: }
1.238 brouard 8570:
1.241 brouard 8571: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 8572: 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 8573: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8574: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8575: k=3;
8576: for (i=1; i<= nlstate ; i ++){
8577: if(i==1){
8578: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8579: }else{
8580: fprintf(ficgp,", '' ");
8581: }
8582: l=(nlstate+ndeath)*(i-1)+1;
8583: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8584: for (j=2; j<= nlstate+ndeath ; j ++)
8585: fprintf(ficgp,"+$%d",k+l+j-1);
8586: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
8587: } /* nlstate */
1.264 brouard 8588: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8589: } /* end cpt state*/
8590: } /* end nres */
1.337 brouard 8591: /* } /\* end covariate k1 *\/ */
1.238 brouard 8592:
1.220 brouard 8593: /* 5eme */
1.201 brouard 8594: /* Survival functions (period) from state i in state j by final state j */
1.337 brouard 8595: /* for (k1=1; k1<= m ; k1++){ /\* For each covariate combination if any *\/ */
1.238 brouard 8596: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8597: k1=TKresult[nres];
1.338 brouard 8598: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8599: /* if(m != 1 && TKresult[nres]!= k1) */
8600: /* continue; */
1.238 brouard 8601: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 8602: strcpy(gplotlabel,"(");
1.238 brouard 8603: 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 8604: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8605: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8606: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8607: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8608: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8609: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8610: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8611: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8612: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8613: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8614: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8615: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8616: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8617: /* } */
8618: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8619: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8620: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8621: }
1.264 brouard 8622: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8623: fprintf(ficgp,"\n#\n");
8624: if(invalidvarcomb[k1]){
8625: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8626: continue;
8627: }
1.227 brouard 8628:
1.241 brouard 8629: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 8630: 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 8631: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8632: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8633: k=3;
8634: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8635: if(j==1)
8636: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8637: else
8638: fprintf(ficgp,", '' ");
8639: l=(nlstate+ndeath)*(cpt-1) +j;
8640: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
8641: /* for (i=2; i<= nlstate+ndeath ; i ++) */
8642: /* fprintf(ficgp,"+$%d",k+l+i-1); */
8643: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
8644: } /* nlstate */
8645: fprintf(ficgp,", '' ");
8646: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
8647: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8648: l=(nlstate+ndeath)*(cpt-1) +j;
8649: if(j < nlstate)
8650: fprintf(ficgp,"$%d +",k+l);
8651: else
8652: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
8653: }
1.264 brouard 8654: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8655: } /* end cpt state*/
1.337 brouard 8656: /* } /\* end covariate *\/ */
1.238 brouard 8657: } /* end nres */
1.227 brouard 8658:
1.220 brouard 8659: /* 6eme */
1.202 brouard 8660: /* CV preval stable (period) for each covariate */
1.337 brouard 8661: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8662: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8663: k1=TKresult[nres];
1.338 brouard 8664: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8665: /* if(m != 1 && TKresult[nres]!= k1) */
8666: /* continue; */
1.255 brouard 8667: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 8668: strcpy(gplotlabel,"(");
1.288 brouard 8669: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8670: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8671: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8672: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8673: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8674: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8675: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8676: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8677: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8678: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8679: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8680: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8681: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8682: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8683: /* } */
8684: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8685: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8686: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8687: }
1.264 brouard 8688: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8689: fprintf(ficgp,"\n#\n");
1.223 brouard 8690: if(invalidvarcomb[k1]){
1.227 brouard 8691: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8692: continue;
1.223 brouard 8693: }
1.227 brouard 8694:
1.241 brouard 8695: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 8696: 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 8697: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8698: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 8699: k=3; /* Offset */
1.255 brouard 8700: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8701: if(i==1)
8702: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8703: else
8704: fprintf(ficgp,", '' ");
1.255 brouard 8705: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8706: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8707: for (j=2; j<= nlstate ; j ++)
8708: fprintf(ficgp,"+$%d",k+l+j-1);
8709: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8710: } /* nlstate */
1.264 brouard 8711: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8712: } /* end cpt state*/
8713: } /* end covariate */
1.227 brouard 8714:
8715:
1.220 brouard 8716: /* 7eme */
1.296 brouard 8717: if(prevbcast == 1){
1.288 brouard 8718: /* CV backward prevalence for each covariate */
1.337 brouard 8719: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8720: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8721: k1=TKresult[nres];
1.338 brouard 8722: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8723: /* if(m != 1 && TKresult[nres]!= k1) */
8724: /* continue; */
1.268 brouard 8725: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8726: strcpy(gplotlabel,"(");
1.288 brouard 8727: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8728: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8729: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8730: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8731: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8732: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8733: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8734: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8735: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8736: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8737: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8738: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8739: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8740: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8741: /* } */
8742: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8743: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8744: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8745: }
1.264 brouard 8746: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8747: fprintf(ficgp,"\n#\n");
8748: if(invalidvarcomb[k1]){
8749: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8750: continue;
8751: }
8752:
1.241 brouard 8753: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8754: 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 8755: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8756: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8757: k=3; /* Offset */
1.268 brouard 8758: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8759: if(i==1)
8760: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8761: else
8762: fprintf(ficgp,", '' ");
8763: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8764: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8765: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8766: /* 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 8767: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8768: /* for (j=2; j<= nlstate ; j ++) */
8769: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8770: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8771: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8772: } /* nlstate */
1.264 brouard 8773: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8774: } /* end cpt state*/
8775: } /* end covariate */
1.296 brouard 8776: } /* End if prevbcast */
1.218 brouard 8777:
1.223 brouard 8778: /* 8eme */
1.218 brouard 8779: if(prevfcast==1){
1.288 brouard 8780: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8781:
1.337 brouard 8782: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8783: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8784: k1=TKresult[nres];
1.338 brouard 8785: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8786: /* if(m != 1 && TKresult[nres]!= k1) */
8787: /* continue; */
1.211 brouard 8788: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8789: strcpy(gplotlabel,"(");
1.288 brouard 8790: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8791: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8792: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8793: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8794: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
8795: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
8796: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8797: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8798: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8799: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8800: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8801: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8802: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8803: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8804: /* } */
8805: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8806: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8807: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8808: }
1.264 brouard 8809: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8810: fprintf(ficgp,"\n#\n");
8811: if(invalidvarcomb[k1]){
8812: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8813: continue;
8814: }
8815:
8816: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8817: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8818: 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 8819: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8820: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8821:
8822: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8823: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8824: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8825: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8826: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8827: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8828: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8829: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8830: if(i==istart){
1.227 brouard 8831: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8832: }else{
8833: fprintf(ficgp,",\\\n '' ");
8834: }
8835: if(cptcoveff ==0){ /* No covariate */
8836: ioffset=2; /* Age is in 2 */
8837: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8838: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8839: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8840: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8841: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8842: if(i==nlstate+1){
1.270 brouard 8843: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8844: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8845: fprintf(ficgp,",\\\n '' ");
8846: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8847: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8848: offyear, \
1.268 brouard 8849: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8850: }else
1.227 brouard 8851: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8852: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8853: }else{ /* more than 2 covariates */
1.270 brouard 8854: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8855: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8856: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8857: iyearc=ioffset-1;
8858: iagec=ioffset;
1.227 brouard 8859: fprintf(ficgp," u %d:(",ioffset);
8860: kl=0;
8861: strcpy(gplotcondition,"(");
8862: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8863: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8864: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8865: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8866: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8867: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8868: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8869: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8870: kl++;
8871: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8872: kl++;
8873: if(k <cptcoveff && cptcoveff>1)
8874: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8875: }
8876: strcpy(gplotcondition+strlen(gplotcondition),")");
8877: /* 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 *\/ */
8878: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8879: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8880: /* '' 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*/
8881: if(i==nlstate+1){
1.270 brouard 8882: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8883: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8884: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8885: fprintf(ficgp," u %d:(",iagec);
8886: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8887: iyearc, iagec, offyear, \
8888: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8889: /* '' 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 8890: }else{
8891: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8892: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8893: }
8894: } /* end if covariate */
8895: } /* nlstate */
1.264 brouard 8896: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8897: } /* end cpt state*/
8898: } /* end covariate */
8899: } /* End if prevfcast */
1.227 brouard 8900:
1.296 brouard 8901: if(prevbcast==1){
1.268 brouard 8902: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8903:
1.337 brouard 8904: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.268 brouard 8905: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8906: k1=TKresult[nres];
1.338 brouard 8907: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8908: /* if(m != 1 && TKresult[nres]!= k1) */
8909: /* continue; */
1.268 brouard 8910: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8911: strcpy(gplotlabel,"(");
8912: 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 8913: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8914: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8915: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8916: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
8917: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
8918: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8919: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8920: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8921: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8922: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8923: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8924: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8925: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8926: /* } */
8927: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8928: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8929: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.268 brouard 8930: }
8931: strcpy(gplotlabel+strlen(gplotlabel),")");
8932: fprintf(ficgp,"\n#\n");
8933: if(invalidvarcomb[k1]){
8934: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8935: continue;
8936: }
8937:
8938: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8939: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8940: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8941: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8942: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8943:
8944: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8945: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8946: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8947: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8948: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8949: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8950: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8951: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8952: if(i==istart){
8953: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8954: }else{
8955: fprintf(ficgp,",\\\n '' ");
8956: }
8957: if(cptcoveff ==0){ /* No covariate */
8958: ioffset=2; /* Age is in 2 */
8959: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8960: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8961: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8962: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8963: fprintf(ficgp," u %d:(", ioffset);
8964: if(i==nlstate+1){
1.270 brouard 8965: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8966: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8967: fprintf(ficgp,",\\\n '' ");
8968: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8969: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8970: offbyear, \
8971: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8972: }else
8973: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8974: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8975: }else{ /* more than 2 covariates */
1.270 brouard 8976: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8977: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8978: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8979: iyearc=ioffset-1;
8980: iagec=ioffset;
1.268 brouard 8981: fprintf(ficgp," u %d:(",ioffset);
8982: kl=0;
8983: strcpy(gplotcondition,"(");
1.337 brouard 8984: for (k=1; k<=cptcovs; k++){ /* For each covariate k of the resultline, get corresponding value lv for combination k1 */
1.338 brouard 8985: if(Dummy[modelresult[nres][k]]==0){ /* To be verified */
1.337 brouard 8986: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate writing the chain of conditions *\/ */
8987: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8988: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8989: lv=Tvresult[nres][k];
8990: vlv=TinvDoQresult[nres][Tvresult[nres][k]];
8991: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8992: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8993: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8994: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8995: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8996: kl++;
8997: /* sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]); */
8998: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%lg " ,kl,Tvresult[nres][k], kl+1,TinvDoQresult[nres][Tvresult[nres][k]]);
8999: kl++;
1.338 brouard 9000: if(k <cptcovs && cptcovs>1)
1.337 brouard 9001: sprintf(gplotcondition+strlen(gplotcondition)," && ");
9002: }
1.268 brouard 9003: }
9004: strcpy(gplotcondition+strlen(gplotcondition),")");
9005: /* 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 *\/ */
9006: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
9007: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
9008: /* '' 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*/
9009: if(i==nlstate+1){
1.270 brouard 9010: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
9011: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 9012: fprintf(ficgp,",\\\n '' ");
1.270 brouard 9013: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 9014: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 9015: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
9016: iyearc,iagec,offbyear, \
9017: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 9018: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
9019: }else{
9020: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
9021: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
9022: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
9023: }
9024: } /* end if covariate */
9025: } /* nlstate */
9026: fprintf(ficgp,"\nset out; unset label;\n");
9027: } /* end cpt state*/
9028: } /* end covariate */
1.296 brouard 9029: } /* End if prevbcast */
1.268 brouard 9030:
1.227 brouard 9031:
1.238 brouard 9032: /* 9eme writing MLE parameters */
9033: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 9034: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 9035: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 9036: for(k=1; k <=(nlstate+ndeath); k++){
9037: if (k != i) {
1.227 brouard 9038: fprintf(ficgp,"# current state %d\n",k);
9039: for(j=1; j <=ncovmodel; j++){
9040: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
9041: jk++;
9042: }
9043: fprintf(ficgp,"\n");
1.126 brouard 9044: }
9045: }
1.223 brouard 9046: }
1.187 brouard 9047: fprintf(ficgp,"##############\n#\n");
1.227 brouard 9048:
1.145 brouard 9049: /*goto avoid;*/
1.238 brouard 9050: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
9051: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 9052: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
9053: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
9054: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
9055: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
9056: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9057: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
9058: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9059: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
9060: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
9061: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9062: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
9063: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
9064: fprintf(ficgp,"#\n");
1.223 brouard 9065: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 9066: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.338 brouard 9067: fprintf(ficgp,"#model=1+age+%s \n",model);
1.238 brouard 9068: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 9069: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
1.337 brouard 9070: /* for(k1=1; k1 <=m; k1++) /\* For each combination of covariate *\/ */
1.237 brouard 9071: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 9072: /* k1=nres; */
1.338 brouard 9073: k1=TKresult[nres];
9074: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 9075: fprintf(ficgp,"\n\n# Resultline k1=%d ",k1);
1.264 brouard 9076: strcpy(gplotlabel,"(");
1.276 brouard 9077: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.337 brouard 9078: for (k=1; k<=cptcovs; k++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
9079: /* for each resultline nres, and position k, Tvresult[nres][k] gives the name of the variable and
9080: TinvDoQresult[nres][Tvresult[nres][k]] gives its value double or integer) */
9081: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
9082: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
9083: }
9084: /* if(m != 1 && TKresult[nres]!= k1) */
9085: /* continue; */
9086: /* fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1); */
9087: /* strcpy(gplotlabel,"("); */
9088: /* /\*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*\/ */
9089: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
9090: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
9091: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
9092: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
9093: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
9094: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
9095: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
9096: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
9097: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
9098: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
9099: /* } */
9100: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
9101: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
9102: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
9103: /* } */
1.264 brouard 9104: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 9105: fprintf(ficgp,"\n#\n");
1.264 brouard 9106: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 9107: fprintf(ficgp,"\nset key outside ");
9108: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
9109: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 9110: fprintf(ficgp,"\nset ter svg size 640, 480 ");
9111: if (ng==1){
9112: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
9113: fprintf(ficgp,"\nunset log y");
9114: }else if (ng==2){
9115: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
9116: fprintf(ficgp,"\nset log y");
9117: }else if (ng==3){
9118: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
9119: fprintf(ficgp,"\nset log y");
9120: }else
9121: fprintf(ficgp,"\nunset title ");
9122: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
9123: i=1;
9124: for(k2=1; k2<=nlstate; k2++) {
9125: k3=i;
9126: for(k=1; k<=(nlstate+ndeath); k++) {
9127: if (k != k2){
9128: switch( ng) {
9129: case 1:
9130: if(nagesqr==0)
9131: fprintf(ficgp," p%d+p%d*x",i,i+1);
9132: else /* nagesqr =1 */
9133: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
9134: break;
9135: case 2: /* ng=2 */
9136: if(nagesqr==0)
9137: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
9138: else /* nagesqr =1 */
9139: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
9140: break;
9141: case 3:
9142: if(nagesqr==0)
9143: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
9144: else /* nagesqr =1 */
9145: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
9146: break;
9147: }
9148: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 9149: ijp=1; /* product no age */
9150: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
9151: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 9152: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 9153: switch(Typevar[j]){
9154: case 1:
9155: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
9156: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
9157: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
9158: if(DummyV[j]==0){/* Bug valgrind */
9159: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
9160: }else{ /* quantitative */
9161: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
9162: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9163: }
9164: ij++;
1.268 brouard 9165: }
1.237 brouard 9166: }
1.329 brouard 9167: }
9168: break;
9169: case 2:
9170: if(cptcovprod >0){
9171: if(j==Tprod[ijp]) { /* */
9172: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
9173: if(ijp <=cptcovprod) { /* Product */
9174: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
9175: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
9176: /* 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)]); */
9177: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
9178: }else{ /* Vn is dummy and Vm is quanti */
9179: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
9180: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9181: }
9182: }else{ /* Vn*Vm Vn is quanti */
9183: if(DummyV[Tvard[ijp][2]]==0){
9184: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
9185: }else{ /* Both quanti */
9186: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9187: }
1.268 brouard 9188: }
1.329 brouard 9189: ijp++;
1.237 brouard 9190: }
1.329 brouard 9191: } /* end Tprod */
9192: }
9193: break;
9194: case 0:
9195: /* simple covariate */
1.264 brouard 9196: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 9197: if(Dummy[j]==0){
9198: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
9199: }else{ /* quantitative */
9200: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 9201: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 9202: }
1.329 brouard 9203: /* end simple */
9204: break;
9205: default:
9206: break;
9207: } /* end switch */
1.237 brouard 9208: } /* end j */
1.329 brouard 9209: }else{ /* k=k2 */
9210: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
9211: fprintf(ficgp," (1.");i=i-ncovmodel;
9212: }else
9213: i=i-ncovmodel;
1.223 brouard 9214: }
1.227 brouard 9215:
1.223 brouard 9216: if(ng != 1){
9217: fprintf(ficgp,")/(1");
1.227 brouard 9218:
1.264 brouard 9219: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 9220: if(nagesqr==0)
1.264 brouard 9221: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 9222: else /* nagesqr =1 */
1.264 brouard 9223: 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 9224:
1.223 brouard 9225: ij=1;
1.329 brouard 9226: ijp=1;
9227: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
9228: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
9229: switch(Typevar[j]){
9230: case 1:
9231: if(cptcovage >0){
9232: if(j==Tage[ij]) { /* Bug valgrind */
9233: if(ij <=cptcovage) { /* Bug valgrind */
9234: if(DummyV[j]==0){/* Bug valgrind */
9235: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
9236: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
9237: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
9238: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
9239: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9240: }else{ /* quantitative */
9241: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
9242: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
9243: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
9244: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9245: }
9246: ij++;
9247: }
9248: }
9249: }
9250: break;
9251: case 2:
9252: if(cptcovprod >0){
9253: if(j==Tprod[ijp]) { /* */
9254: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
9255: if(ijp <=cptcovprod) { /* Product */
9256: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
9257: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
9258: /* 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)]); */
9259: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
9260: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
9261: }else{ /* Vn is dummy and Vm is quanti */
9262: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
9263: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9264: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
9265: }
9266: }else{ /* Vn*Vm Vn is quanti */
9267: if(DummyV[Tvard[ijp][2]]==0){
9268: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
9269: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
9270: }else{ /* Both quanti */
9271: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9272: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
9273: }
9274: }
9275: ijp++;
9276: }
9277: } /* end Tprod */
9278: } /* end if */
9279: break;
9280: case 0:
9281: /* simple covariate */
9282: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
9283: if(Dummy[j]==0){
9284: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
9285: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
9286: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
9287: }else{ /* quantitative */
9288: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
9289: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
9290: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9291: }
9292: /* end simple */
9293: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
9294: break;
9295: default:
9296: break;
9297: } /* end switch */
1.223 brouard 9298: }
9299: fprintf(ficgp,")");
9300: }
9301: fprintf(ficgp,")");
9302: if(ng ==2)
1.276 brouard 9303: 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 9304: else /* ng= 3 */
1.276 brouard 9305: 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 9306: }else{ /* end ng <> 1 */
1.223 brouard 9307: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 9308: 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 9309: }
9310: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
9311: fprintf(ficgp,",");
9312: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
9313: fprintf(ficgp,",");
9314: i=i+ncovmodel;
9315: } /* end k */
9316: } /* end k2 */
1.276 brouard 9317: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
9318: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.337 brouard 9319: } /* end resultline */
1.223 brouard 9320: } /* end ng */
9321: /* avoid: */
9322: fflush(ficgp);
1.126 brouard 9323: } /* end gnuplot */
9324:
9325:
9326: /*************** Moving average **************/
1.219 brouard 9327: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 9328: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 9329:
1.222 brouard 9330: int i, cpt, cptcod;
9331: int modcovmax =1;
9332: int mobilavrange, mob;
9333: int iage=0;
1.288 brouard 9334: int firstA1=0, firstA2=0;
1.222 brouard 9335:
1.266 brouard 9336: double sum=0., sumr=0.;
1.222 brouard 9337: double age;
1.266 brouard 9338: double *sumnewp, *sumnewm, *sumnewmr;
9339: double *agemingood, *agemaxgood;
9340: double *agemingoodr, *agemaxgoodr;
1.222 brouard 9341:
9342:
1.278 brouard 9343: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
9344: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 9345:
9346: sumnewp = vector(1,ncovcombmax);
9347: sumnewm = vector(1,ncovcombmax);
1.266 brouard 9348: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 9349: agemingood = vector(1,ncovcombmax);
1.266 brouard 9350: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 9351: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 9352: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 9353:
9354: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 9355: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 9356: sumnewp[cptcod]=0.;
1.266 brouard 9357: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
9358: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 9359: }
9360: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
9361:
1.266 brouard 9362: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
9363: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 9364: else mobilavrange=mobilav;
9365: for (age=bage; age<=fage; age++)
9366: for (i=1; i<=nlstate;i++)
9367: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
9368: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9369: /* We keep the original values on the extreme ages bage, fage and for
9370: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
9371: we use a 5 terms etc. until the borders are no more concerned.
9372: */
9373: for (mob=3;mob <=mobilavrange;mob=mob+2){
9374: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 9375: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
9376: sumnewm[cptcod]=0.;
9377: for (i=1; i<=nlstate;i++){
1.222 brouard 9378: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
9379: for (cpt=1;cpt<=(mob-1)/2;cpt++){
9380: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
9381: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
9382: }
9383: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 9384: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9385: } /* end i */
9386: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
9387: } /* end cptcod */
1.222 brouard 9388: }/* end age */
9389: }/* end mob */
1.266 brouard 9390: }else{
9391: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 9392: return -1;
1.266 brouard 9393: }
9394:
9395: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 9396: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
9397: if(invalidvarcomb[cptcod]){
9398: printf("\nCombination (%d) ignored because no cases \n",cptcod);
9399: continue;
9400: }
1.219 brouard 9401:
1.266 brouard 9402: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
9403: sumnewm[cptcod]=0.;
9404: sumnewmr[cptcod]=0.;
9405: for (i=1; i<=nlstate;i++){
9406: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9407: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9408: }
9409: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9410: agemingoodr[cptcod]=age;
9411: }
9412: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9413: agemingood[cptcod]=age;
9414: }
9415: } /* age */
9416: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 9417: sumnewm[cptcod]=0.;
1.266 brouard 9418: sumnewmr[cptcod]=0.;
1.222 brouard 9419: for (i=1; i<=nlstate;i++){
9420: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9421: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9422: }
9423: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9424: agemaxgoodr[cptcod]=age;
1.222 brouard 9425: }
9426: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 9427: agemaxgood[cptcod]=age;
9428: }
9429: } /* age */
9430: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
9431: /* but they will change */
1.288 brouard 9432: firstA1=0;firstA2=0;
1.266 brouard 9433: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
9434: sumnewm[cptcod]=0.;
9435: sumnewmr[cptcod]=0.;
9436: for (i=1; i<=nlstate;i++){
9437: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9438: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9439: }
9440: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9441: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9442: agemaxgoodr[cptcod]=age; /* age min */
9443: for (i=1; i<=nlstate;i++)
9444: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9445: }else{ /* bad we change the value with the values of good ages */
9446: for (i=1; i<=nlstate;i++){
9447: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
9448: } /* i */
9449: } /* end bad */
9450: }else{
9451: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9452: agemaxgood[cptcod]=age;
9453: }else{ /* bad we change the value with the values of good ages */
9454: for (i=1; i<=nlstate;i++){
9455: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
9456: } /* i */
9457: } /* end bad */
9458: }/* end else */
9459: sum=0.;sumr=0.;
9460: for (i=1; i<=nlstate;i++){
9461: sum+=mobaverage[(int)age][i][cptcod];
9462: sumr+=probs[(int)age][i][cptcod];
9463: }
9464: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 9465: if(!firstA1){
9466: firstA1=1;
9467: 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);
9468: }
9469: 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 9470: } /* end bad */
9471: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9472: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 9473: if(!firstA2){
9474: firstA2=1;
9475: 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);
9476: }
9477: 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 9478: } /* end bad */
9479: }/* age */
1.266 brouard 9480:
9481: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 9482: sumnewm[cptcod]=0.;
1.266 brouard 9483: sumnewmr[cptcod]=0.;
1.222 brouard 9484: for (i=1; i<=nlstate;i++){
9485: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9486: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9487: }
9488: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9489: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
9490: agemingoodr[cptcod]=age;
9491: for (i=1; i<=nlstate;i++)
9492: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9493: }else{ /* bad we change the value with the values of good ages */
9494: for (i=1; i<=nlstate;i++){
9495: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
9496: } /* i */
9497: } /* end bad */
9498: }else{
9499: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9500: agemingood[cptcod]=age;
9501: }else{ /* bad */
9502: for (i=1; i<=nlstate;i++){
9503: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
9504: } /* i */
9505: } /* end bad */
9506: }/* end else */
9507: sum=0.;sumr=0.;
9508: for (i=1; i<=nlstate;i++){
9509: sum+=mobaverage[(int)age][i][cptcod];
9510: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 9511: }
1.266 brouard 9512: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 9513: 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 9514: } /* end bad */
9515: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9516: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 9517: 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 9518: } /* end bad */
9519: }/* age */
1.266 brouard 9520:
1.222 brouard 9521:
9522: for (age=bage; age<=fage; age++){
1.235 brouard 9523: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 9524: sumnewp[cptcod]=0.;
9525: sumnewm[cptcod]=0.;
9526: for (i=1; i<=nlstate;i++){
9527: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
9528: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9529: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
9530: }
9531: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
9532: }
9533: /* printf("\n"); */
9534: /* } */
1.266 brouard 9535:
1.222 brouard 9536: /* brutal averaging */
1.266 brouard 9537: /* for (i=1; i<=nlstate;i++){ */
9538: /* for (age=1; age<=bage; age++){ */
9539: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
9540: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9541: /* } */
9542: /* for (age=fage; age<=AGESUP; age++){ */
9543: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
9544: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9545: /* } */
9546: /* } /\* end i status *\/ */
9547: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
9548: /* for (age=1; age<=AGESUP; age++){ */
9549: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
9550: /* mobaverage[(int)age][i][cptcod]=0.; */
9551: /* } */
9552: /* } */
1.222 brouard 9553: }/* end cptcod */
1.266 brouard 9554: free_vector(agemaxgoodr,1, ncovcombmax);
9555: free_vector(agemaxgood,1, ncovcombmax);
9556: free_vector(agemingood,1, ncovcombmax);
9557: free_vector(agemingoodr,1, ncovcombmax);
9558: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 9559: free_vector(sumnewm,1, ncovcombmax);
9560: free_vector(sumnewp,1, ncovcombmax);
9561: return 0;
9562: }/* End movingaverage */
1.218 brouard 9563:
1.126 brouard 9564:
1.296 brouard 9565:
1.126 brouard 9566: /************** Forecasting ******************/
1.296 brouard 9567: /* 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)*/
9568: 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){
9569: /* dateintemean, mean date of interviews
9570: dateprojd, year, month, day of starting projection
9571: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 9572: agemin, agemax range of age
9573: dateprev1 dateprev2 range of dates during which prevalence is computed
9574: */
1.296 brouard 9575: /* double anprojd, mprojd, jprojd; */
9576: /* double anprojf, mprojf, jprojf; */
1.267 brouard 9577: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 9578: double agec; /* generic age */
1.296 brouard 9579: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 9580: double *popeffectif,*popcount;
9581: double ***p3mat;
1.218 brouard 9582: /* double ***mobaverage; */
1.126 brouard 9583: char fileresf[FILENAMELENGTH];
9584:
9585: agelim=AGESUP;
1.211 brouard 9586: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9587: in each health status at the date of interview (if between dateprev1 and dateprev2).
9588: We still use firstpass and lastpass as another selection.
9589: */
1.214 brouard 9590: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9591: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 9592:
1.201 brouard 9593: strcpy(fileresf,"F_");
9594: strcat(fileresf,fileresu);
1.126 brouard 9595: if((ficresf=fopen(fileresf,"w"))==NULL) {
9596: printf("Problem with forecast resultfile: %s\n", fileresf);
9597: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
9598: }
1.235 brouard 9599: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
9600: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 9601:
1.225 brouard 9602: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 9603:
9604:
9605: stepsize=(int) (stepm+YEARM-1)/YEARM;
9606: if (stepm<=12) stepsize=1;
9607: if(estepm < stepm){
9608: printf ("Problem %d lower than %d\n",estepm, stepm);
9609: }
1.270 brouard 9610: else{
9611: hstepm=estepm;
9612: }
9613: if(estepm > stepm){ /* Yes every two year */
9614: stepsize=2;
9615: }
1.296 brouard 9616: hstepm=hstepm/stepm;
1.126 brouard 9617:
1.296 brouard 9618:
9619: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9620: /* fractional in yp1 *\/ */
9621: /* aintmean=yp; */
9622: /* yp2=modf((yp1*12),&yp); */
9623: /* mintmean=yp; */
9624: /* yp1=modf((yp2*30.5),&yp); */
9625: /* jintmean=yp; */
9626: /* if(jintmean==0) jintmean=1; */
9627: /* if(mintmean==0) mintmean=1; */
1.126 brouard 9628:
1.296 brouard 9629:
9630: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
9631: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
9632: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 9633: i1=pow(2,cptcoveff);
1.126 brouard 9634: if (cptcovn < 1){i1=1;}
9635:
1.296 brouard 9636: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 9637:
9638: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 9639:
1.126 brouard 9640: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 9641: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9642: 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 9643: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9644: continue;
1.227 brouard 9645: if(invalidvarcomb[k]){
9646: printf("\nCombination (%d) projection ignored because no cases \n",k);
9647: continue;
9648: }
9649: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
9650: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9651: /* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); */
9652: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.227 brouard 9653: }
1.235 brouard 9654: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 9655: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 9656: }
1.227 brouard 9657: fprintf(ficresf," yearproj age");
9658: for(j=1; j<=nlstate+ndeath;j++){
9659: for(i=1; i<=nlstate;i++)
9660: fprintf(ficresf," p%d%d",i,j);
9661: fprintf(ficresf," wp.%d",j);
9662: }
1.296 brouard 9663: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 9664: fprintf(ficresf,"\n");
1.296 brouard 9665: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 9666: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
9667: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 9668: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
9669: nhstepm = nhstepm/hstepm;
9670: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9671: oldm=oldms;savm=savms;
1.268 brouard 9672: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 9673: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 9674: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 9675: for (h=0; h<=nhstepm; h++){
9676: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 9677: break;
9678: }
9679: }
9680: fprintf(ficresf,"\n");
9681: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9682: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Tvaraff not correct *\/ */
9683: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* TnsdVar[Tvaraff] correct */
1.296 brouard 9684: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 9685:
9686: for(j=1; j<=nlstate+ndeath;j++) {
9687: ppij=0.;
9688: for(i=1; i<=nlstate;i++) {
1.278 brouard 9689: if (mobilav>=1)
9690: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
9691: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
9692: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
9693: }
1.268 brouard 9694: fprintf(ficresf," %.3f", p3mat[i][j][h]);
9695: } /* end i */
9696: fprintf(ficresf," %.3f", ppij);
9697: }/* end j */
1.227 brouard 9698: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9699: } /* end agec */
1.266 brouard 9700: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
9701: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 9702: } /* end yearp */
9703: } /* end k */
1.219 brouard 9704:
1.126 brouard 9705: fclose(ficresf);
1.215 brouard 9706: printf("End of Computing forecasting \n");
9707: fprintf(ficlog,"End of Computing forecasting\n");
9708:
1.126 brouard 9709: }
9710:
1.269 brouard 9711: /************** Back Forecasting ******************/
1.296 brouard 9712: /* 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){ */
9713: 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){
9714: /* back1, year, month, day of starting backprojection
1.267 brouard 9715: agemin, agemax range of age
9716: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 9717: anback2 year of end of backprojection (same day and month as back1).
9718: prevacurrent and prev are prevalences.
1.267 brouard 9719: */
9720: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
9721: double agec; /* generic age */
1.302 brouard 9722: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 9723: double *popeffectif,*popcount;
9724: double ***p3mat;
9725: /* double ***mobaverage; */
9726: char fileresfb[FILENAMELENGTH];
9727:
1.268 brouard 9728: agelim=AGEINF;
1.267 brouard 9729: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9730: in each health status at the date of interview (if between dateprev1 and dateprev2).
9731: We still use firstpass and lastpass as another selection.
9732: */
9733: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9734: /* firstpass, lastpass, stepm, weightopt, model); */
9735:
9736: /*Do we need to compute prevalence again?*/
9737:
9738: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
9739:
9740: strcpy(fileresfb,"FB_");
9741: strcat(fileresfb,fileresu);
9742: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
9743: printf("Problem with back forecast resultfile: %s\n", fileresfb);
9744: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
9745: }
9746: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9747: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9748:
9749: if (cptcoveff==0) ncodemax[cptcoveff]=1;
9750:
9751:
9752: stepsize=(int) (stepm+YEARM-1)/YEARM;
9753: if (stepm<=12) stepsize=1;
9754: if(estepm < stepm){
9755: printf ("Problem %d lower than %d\n",estepm, stepm);
9756: }
1.270 brouard 9757: else{
9758: hstepm=estepm;
9759: }
9760: if(estepm >= stepm){ /* Yes every two year */
9761: stepsize=2;
9762: }
1.267 brouard 9763:
9764: hstepm=hstepm/stepm;
1.296 brouard 9765: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9766: /* fractional in yp1 *\/ */
9767: /* aintmean=yp; */
9768: /* yp2=modf((yp1*12),&yp); */
9769: /* mintmean=yp; */
9770: /* yp1=modf((yp2*30.5),&yp); */
9771: /* jintmean=yp; */
9772: /* if(jintmean==0) jintmean=1; */
9773: /* if(mintmean==0) jintmean=1; */
1.267 brouard 9774:
9775: i1=pow(2,cptcoveff);
9776: if (cptcovn < 1){i1=1;}
9777:
1.296 brouard 9778: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
9779: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 9780:
9781: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
9782:
9783: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9784: for(k=1; k<=i1;k++){
9785: if(i1 != 1 && TKresult[nres]!= k)
9786: continue;
9787: if(invalidvarcomb[k]){
9788: printf("\nCombination (%d) projection ignored because no cases \n",k);
9789: continue;
9790: }
1.268 brouard 9791: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 9792: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9793: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.267 brouard 9794: }
9795: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9796: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9797: }
9798: fprintf(ficresfb," yearbproj age");
9799: for(j=1; j<=nlstate+ndeath;j++){
9800: for(i=1; i<=nlstate;i++)
1.268 brouard 9801: fprintf(ficresfb," b%d%d",i,j);
9802: fprintf(ficresfb," b.%d",j);
1.267 brouard 9803: }
1.296 brouard 9804: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 9805: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
9806: fprintf(ficresfb,"\n");
1.296 brouard 9807: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 9808: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 9809: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
9810: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 9811: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 9812: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 9813: nhstepm = nhstepm/hstepm;
9814: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9815: oldm=oldms;savm=savms;
1.268 brouard 9816: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 9817: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 9818: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 9819: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
9820: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
9821: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9822: for (h=0; h<=nhstepm; h++){
1.268 brouard 9823: if (h*hstepm/YEARM*stepm ==-yearp) {
9824: break;
9825: }
9826: }
9827: fprintf(ficresfb,"\n");
9828: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9829: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.296 brouard 9830: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9831: for(i=1; i<=nlstate+ndeath;i++) {
9832: ppij=0.;ppi=0.;
9833: for(j=1; j<=nlstate;j++) {
9834: /* if (mobilav==1) */
1.269 brouard 9835: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9836: ppi=ppi+prevacurrent[(int)agec][j][k];
9837: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9838: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9839: /* else { */
9840: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9841: /* } */
1.268 brouard 9842: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9843: } /* end j */
9844: if(ppi <0.99){
9845: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9846: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9847: }
9848: fprintf(ficresfb," %.3f", ppij);
9849: }/* end j */
1.267 brouard 9850: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9851: } /* end agec */
9852: } /* end yearp */
9853: } /* end k */
1.217 brouard 9854:
1.267 brouard 9855: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9856:
1.267 brouard 9857: fclose(ficresfb);
9858: printf("End of Computing Back forecasting \n");
9859: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9860:
1.267 brouard 9861: }
1.217 brouard 9862:
1.269 brouard 9863: /* Variance of prevalence limit: varprlim */
9864: 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 9865: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9866:
9867: char fileresvpl[FILENAMELENGTH];
9868: FILE *ficresvpl;
9869: double **oldm, **savm;
9870: double **varpl; /* Variances of prevalence limits by age */
9871: int i1, k, nres, j ;
9872:
9873: strcpy(fileresvpl,"VPL_");
9874: strcat(fileresvpl,fileresu);
9875: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9876: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9877: exit(0);
9878: }
1.288 brouard 9879: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9880: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9881:
9882: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9883: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9884:
9885: i1=pow(2,cptcoveff);
9886: if (cptcovn < 1){i1=1;}
9887:
1.337 brouard 9888: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9889: k=TKresult[nres];
1.338 brouard 9890: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 9891: /* for(k=1; k<=i1;k++){ /\* We find the combination equivalent to result line values of dummies *\/ */
1.269 brouard 9892: if(i1 != 1 && TKresult[nres]!= k)
9893: continue;
9894: fprintf(ficresvpl,"\n#****** ");
9895: printf("\n#****** ");
9896: fprintf(ficlog,"\n#****** ");
1.337 brouard 9897: for(j=1;j<=cptcovs;j++) {
9898: fprintf(ficresvpl,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9899: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9900: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9901: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9902: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.269 brouard 9903: }
1.337 brouard 9904: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
9905: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9906: /* fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9907: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9908: /* } */
1.269 brouard 9909: fprintf(ficresvpl,"******\n");
9910: printf("******\n");
9911: fprintf(ficlog,"******\n");
9912:
9913: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9914: oldm=oldms;savm=savms;
9915: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9916: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9917: /*}*/
9918: }
9919:
9920: fclose(ficresvpl);
1.288 brouard 9921: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9922: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9923:
9924: }
9925: /* Variance of back prevalence: varbprlim */
9926: 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){
9927: /*------- Variance of back (stable) prevalence------*/
9928:
9929: char fileresvbl[FILENAMELENGTH];
9930: FILE *ficresvbl;
9931:
9932: double **oldm, **savm;
9933: double **varbpl; /* Variances of back prevalence limits by age */
9934: int i1, k, nres, j ;
9935:
9936: strcpy(fileresvbl,"VBL_");
9937: strcat(fileresvbl,fileresu);
9938: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9939: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9940: exit(0);
9941: }
9942: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9943: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9944:
9945:
9946: i1=pow(2,cptcoveff);
9947: if (cptcovn < 1){i1=1;}
9948:
1.337 brouard 9949: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9950: k=TKresult[nres];
1.338 brouard 9951: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 9952: /* for(k=1; k<=i1;k++){ */
9953: /* if(i1 != 1 && TKresult[nres]!= k) */
9954: /* continue; */
1.269 brouard 9955: fprintf(ficresvbl,"\n#****** ");
9956: printf("\n#****** ");
9957: fprintf(ficlog,"\n#****** ");
1.337 brouard 9958: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */
1.338 brouard 9959: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
9960: fprintf(ficresvbl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
9961: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
1.337 brouard 9962: /* for(j=1;j<=cptcoveff;j++) { */
9963: /* fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9964: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9965: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9966: /* } */
9967: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
9968: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9969: /* fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9970: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
1.269 brouard 9971: }
9972: fprintf(ficresvbl,"******\n");
9973: printf("******\n");
9974: fprintf(ficlog,"******\n");
9975:
9976: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9977: oldm=oldms;savm=savms;
9978:
9979: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9980: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9981: /*}*/
9982: }
9983:
9984: fclose(ficresvbl);
9985: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9986: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9987:
9988: } /* End of varbprlim */
9989:
1.126 brouard 9990: /************** Forecasting *****not tested NB*************/
1.227 brouard 9991: /* 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 9992:
1.227 brouard 9993: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
9994: /* int *popage; */
9995: /* double calagedatem, agelim, kk1, kk2; */
9996: /* double *popeffectif,*popcount; */
9997: /* double ***p3mat,***tabpop,***tabpopprev; */
9998: /* /\* double ***mobaverage; *\/ */
9999: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 10000:
1.227 brouard 10001: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10002: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10003: /* agelim=AGESUP; */
10004: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 10005:
1.227 brouard 10006: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 10007:
10008:
1.227 brouard 10009: /* strcpy(filerespop,"POP_"); */
10010: /* strcat(filerespop,fileresu); */
10011: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
10012: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
10013: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
10014: /* } */
10015: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
10016: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 10017:
1.227 brouard 10018: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 10019:
1.227 brouard 10020: /* /\* if (mobilav!=0) { *\/ */
10021: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
10022: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
10023: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
10024: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
10025: /* /\* } *\/ */
10026: /* /\* } *\/ */
1.126 brouard 10027:
1.227 brouard 10028: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
10029: /* if (stepm<=12) stepsize=1; */
1.126 brouard 10030:
1.227 brouard 10031: /* agelim=AGESUP; */
1.126 brouard 10032:
1.227 brouard 10033: /* hstepm=1; */
10034: /* hstepm=hstepm/stepm; */
1.218 brouard 10035:
1.227 brouard 10036: /* if (popforecast==1) { */
10037: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
10038: /* printf("Problem with population file : %s\n",popfile);exit(0); */
10039: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
10040: /* } */
10041: /* popage=ivector(0,AGESUP); */
10042: /* popeffectif=vector(0,AGESUP); */
10043: /* popcount=vector(0,AGESUP); */
1.126 brouard 10044:
1.227 brouard 10045: /* i=1; */
10046: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 10047:
1.227 brouard 10048: /* imx=i; */
10049: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
10050: /* } */
1.218 brouard 10051:
1.227 brouard 10052: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
10053: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
10054: /* k=k+1; */
10055: /* fprintf(ficrespop,"\n#******"); */
10056: /* for(j=1;j<=cptcoveff;j++) { */
10057: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
10058: /* } */
10059: /* fprintf(ficrespop,"******\n"); */
10060: /* fprintf(ficrespop,"# Age"); */
10061: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
10062: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 10063:
1.227 brouard 10064: /* for (cpt=0; cpt<=0;cpt++) { */
10065: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 10066:
1.227 brouard 10067: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
10068: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
10069: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 10070:
1.227 brouard 10071: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10072: /* oldm=oldms;savm=savms; */
10073: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 10074:
1.227 brouard 10075: /* for (h=0; h<=nhstepm; h++){ */
10076: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
10077: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
10078: /* } */
10079: /* for(j=1; j<=nlstate+ndeath;j++) { */
10080: /* kk1=0.;kk2=0; */
10081: /* for(i=1; i<=nlstate;i++) { */
10082: /* if (mobilav==1) */
10083: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
10084: /* else { */
10085: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
10086: /* } */
10087: /* } */
10088: /* if (h==(int)(calagedatem+12*cpt)){ */
10089: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
10090: /* /\*fprintf(ficrespop," %.3f", kk1); */
10091: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
10092: /* } */
10093: /* } */
10094: /* for(i=1; i<=nlstate;i++){ */
10095: /* kk1=0.; */
10096: /* for(j=1; j<=nlstate;j++){ */
10097: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
10098: /* } */
10099: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
10100: /* } */
1.218 brouard 10101:
1.227 brouard 10102: /* if (h==(int)(calagedatem+12*cpt)) */
10103: /* for(j=1; j<=nlstate;j++) */
10104: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
10105: /* } */
10106: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10107: /* } */
10108: /* } */
1.218 brouard 10109:
1.227 brouard 10110: /* /\******\/ */
1.218 brouard 10111:
1.227 brouard 10112: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
10113: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
10114: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
10115: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
10116: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 10117:
1.227 brouard 10118: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10119: /* oldm=oldms;savm=savms; */
10120: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
10121: /* for (h=0; h<=nhstepm; h++){ */
10122: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
10123: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
10124: /* } */
10125: /* for(j=1; j<=nlstate+ndeath;j++) { */
10126: /* kk1=0.;kk2=0; */
10127: /* for(i=1; i<=nlstate;i++) { */
10128: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
10129: /* } */
10130: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
10131: /* } */
10132: /* } */
10133: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10134: /* } */
10135: /* } */
10136: /* } */
10137: /* } */
1.218 brouard 10138:
1.227 brouard 10139: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 10140:
1.227 brouard 10141: /* if (popforecast==1) { */
10142: /* free_ivector(popage,0,AGESUP); */
10143: /* free_vector(popeffectif,0,AGESUP); */
10144: /* free_vector(popcount,0,AGESUP); */
10145: /* } */
10146: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10147: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10148: /* fclose(ficrespop); */
10149: /* } /\* End of popforecast *\/ */
1.218 brouard 10150:
1.126 brouard 10151: int fileappend(FILE *fichier, char *optionfich)
10152: {
10153: if((fichier=fopen(optionfich,"a"))==NULL) {
10154: printf("Problem with file: %s\n", optionfich);
10155: fprintf(ficlog,"Problem with file: %s\n", optionfich);
10156: return (0);
10157: }
10158: fflush(fichier);
10159: return (1);
10160: }
10161:
10162:
10163: /**************** function prwizard **********************/
10164: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
10165: {
10166:
10167: /* Wizard to print covariance matrix template */
10168:
1.164 brouard 10169: char ca[32], cb[32];
10170: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 10171: int numlinepar;
10172:
10173: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10174: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10175: for(i=1; i <=nlstate; i++){
10176: jj=0;
10177: for(j=1; j <=nlstate+ndeath; j++){
10178: if(j==i) continue;
10179: jj++;
10180: /*ca[0]= k+'a'-1;ca[1]='\0';*/
10181: printf("%1d%1d",i,j);
10182: fprintf(ficparo,"%1d%1d",i,j);
10183: for(k=1; k<=ncovmodel;k++){
10184: /* printf(" %lf",param[i][j][k]); */
10185: /* fprintf(ficparo," %lf",param[i][j][k]); */
10186: printf(" 0.");
10187: fprintf(ficparo," 0.");
10188: }
10189: printf("\n");
10190: fprintf(ficparo,"\n");
10191: }
10192: }
10193: printf("# Scales (for hessian or gradient estimation)\n");
10194: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
10195: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
10196: for(i=1; i <=nlstate; i++){
10197: jj=0;
10198: for(j=1; j <=nlstate+ndeath; j++){
10199: if(j==i) continue;
10200: jj++;
10201: fprintf(ficparo,"%1d%1d",i,j);
10202: printf("%1d%1d",i,j);
10203: fflush(stdout);
10204: for(k=1; k<=ncovmodel;k++){
10205: /* printf(" %le",delti3[i][j][k]); */
10206: /* fprintf(ficparo," %le",delti3[i][j][k]); */
10207: printf(" 0.");
10208: fprintf(ficparo," 0.");
10209: }
10210: numlinepar++;
10211: printf("\n");
10212: fprintf(ficparo,"\n");
10213: }
10214: }
10215: printf("# Covariance matrix\n");
10216: /* # 121 Var(a12)\n\ */
10217: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10218: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10219: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10220: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10221: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10222: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10223: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10224: fflush(stdout);
10225: fprintf(ficparo,"# Covariance matrix\n");
10226: /* # 121 Var(a12)\n\ */
10227: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10228: /* # ...\n\ */
10229: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10230:
10231: for(itimes=1;itimes<=2;itimes++){
10232: jj=0;
10233: for(i=1; i <=nlstate; i++){
10234: for(j=1; j <=nlstate+ndeath; j++){
10235: if(j==i) continue;
10236: for(k=1; k<=ncovmodel;k++){
10237: jj++;
10238: ca[0]= k+'a'-1;ca[1]='\0';
10239: if(itimes==1){
10240: printf("#%1d%1d%d",i,j,k);
10241: fprintf(ficparo,"#%1d%1d%d",i,j,k);
10242: }else{
10243: printf("%1d%1d%d",i,j,k);
10244: fprintf(ficparo,"%1d%1d%d",i,j,k);
10245: /* printf(" %.5le",matcov[i][j]); */
10246: }
10247: ll=0;
10248: for(li=1;li <=nlstate; li++){
10249: for(lj=1;lj <=nlstate+ndeath; lj++){
10250: if(lj==li) continue;
10251: for(lk=1;lk<=ncovmodel;lk++){
10252: ll++;
10253: if(ll<=jj){
10254: cb[0]= lk +'a'-1;cb[1]='\0';
10255: if(ll<jj){
10256: if(itimes==1){
10257: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10258: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10259: }else{
10260: printf(" 0.");
10261: fprintf(ficparo," 0.");
10262: }
10263: }else{
10264: if(itimes==1){
10265: printf(" Var(%s%1d%1d)",ca,i,j);
10266: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
10267: }else{
10268: printf(" 0.");
10269: fprintf(ficparo," 0.");
10270: }
10271: }
10272: }
10273: } /* end lk */
10274: } /* end lj */
10275: } /* end li */
10276: printf("\n");
10277: fprintf(ficparo,"\n");
10278: numlinepar++;
10279: } /* end k*/
10280: } /*end j */
10281: } /* end i */
10282: } /* end itimes */
10283:
10284: } /* end of prwizard */
10285: /******************* Gompertz Likelihood ******************************/
10286: double gompertz(double x[])
10287: {
1.302 brouard 10288: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 10289: int i,n=0; /* n is the size of the sample */
10290:
1.220 brouard 10291: for (i=1;i<=imx ; i++) {
1.126 brouard 10292: sump=sump+weight[i];
10293: /* sump=sump+1;*/
10294: num=num+1;
10295: }
1.302 brouard 10296: L=0.0;
10297: /* agegomp=AGEGOMP; */
1.126 brouard 10298: /* for (i=0; i<=imx; i++)
10299: 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]);*/
10300:
1.302 brouard 10301: for (i=1;i<=imx ; i++) {
10302: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
10303: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
10304: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
10305: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
10306: * +
10307: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
10308: */
10309: if (wav[i] > 1 || agedc[i] < AGESUP) {
10310: if (cens[i] == 1){
10311: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
10312: } else if (cens[i] == 0){
1.126 brouard 10313: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 10314: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
10315: } else
10316: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 10317: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 10318: L=L+A*weight[i];
1.126 brouard 10319: /* 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 10320: }
10321: }
1.126 brouard 10322:
1.302 brouard 10323: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 10324:
10325: return -2*L*num/sump;
10326: }
10327:
1.136 brouard 10328: #ifdef GSL
10329: /******************* Gompertz_f Likelihood ******************************/
10330: double gompertz_f(const gsl_vector *v, void *params)
10331: {
1.302 brouard 10332: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 10333: double *x= (double *) v->data;
10334: int i,n=0; /* n is the size of the sample */
10335:
10336: for (i=0;i<=imx-1 ; i++) {
10337: sump=sump+weight[i];
10338: /* sump=sump+1;*/
10339: num=num+1;
10340: }
10341:
10342:
10343: /* for (i=0; i<=imx; i++)
10344: 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]);*/
10345: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
10346: for (i=1;i<=imx ; i++)
10347: {
10348: if (cens[i] == 1 && wav[i]>1)
10349: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
10350:
10351: if (cens[i] == 0 && wav[i]>1)
10352: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
10353: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
10354:
10355: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
10356: if (wav[i] > 1 ) { /* ??? */
10357: LL=LL+A*weight[i];
10358: /* 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]);*/
10359: }
10360: }
10361:
10362: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
10363: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
10364:
10365: return -2*LL*num/sump;
10366: }
10367: #endif
10368:
1.126 brouard 10369: /******************* Printing html file ***********/
1.201 brouard 10370: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 10371: int lastpass, int stepm, int weightopt, char model[],\
10372: int imx, double p[],double **matcov,double agemortsup){
10373: int i,k;
10374:
10375: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
10376: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
10377: for (i=1;i<=2;i++)
10378: 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 10379: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 10380: fprintf(fichtm,"</ul>");
10381:
10382: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
10383:
10384: 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>");
10385:
10386: for (k=agegomp;k<(agemortsup-2);k++)
10387: 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]);
10388:
10389:
10390: fflush(fichtm);
10391: }
10392:
10393: /******************* Gnuplot file **************/
1.201 brouard 10394: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 10395:
10396: char dirfileres[132],optfileres[132];
1.164 brouard 10397:
1.126 brouard 10398: int ng;
10399:
10400:
10401: /*#ifdef windows */
10402: fprintf(ficgp,"cd \"%s\" \n",pathc);
10403: /*#endif */
10404:
10405:
10406: strcpy(dirfileres,optionfilefiname);
10407: strcpy(optfileres,"vpl");
1.199 brouard 10408: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 10409: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 10410: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 10411: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 10412: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
10413:
10414: }
10415:
1.136 brouard 10416: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
10417: {
1.126 brouard 10418:
1.136 brouard 10419: /*-------- data file ----------*/
10420: FILE *fic;
10421: char dummy[]=" ";
1.240 brouard 10422: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 10423: int lstra;
1.136 brouard 10424: int linei, month, year,iout;
1.302 brouard 10425: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 10426: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 10427: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 10428: char *stratrunc;
1.223 brouard 10429:
1.240 brouard 10430: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
10431: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.328 brouard 10432: for(v=1;v<NCOVMAX;v++){
10433: DummyV[v]=0;
10434: FixedV[v]=0;
10435: }
1.126 brouard 10436:
1.240 brouard 10437: for(v=1; v <=ncovcol;v++){
10438: DummyV[v]=0;
10439: FixedV[v]=0;
10440: }
10441: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
10442: DummyV[v]=1;
10443: FixedV[v]=0;
10444: }
10445: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
10446: DummyV[v]=0;
10447: FixedV[v]=1;
10448: }
10449: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
10450: DummyV[v]=1;
10451: FixedV[v]=1;
10452: }
10453: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
10454: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
10455: 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]);
10456: }
1.339 brouard 10457:
10458: ncovcolt=ncovcol+nqv+ntv+nqtv; /* total of covariates in the data, not in the model equation */
10459:
1.136 brouard 10460: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 10461: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10462: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 10463: }
1.126 brouard 10464:
1.302 brouard 10465: /* Is it a BOM UTF-8 Windows file? */
10466: /* First data line */
10467: linei=0;
10468: while(fgets(line, MAXLINE, fic)) {
10469: noffset=0;
10470: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10471: {
10472: noffset=noffset+3;
10473: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
10474: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
10475: fflush(ficlog); return 1;
10476: }
10477: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
10478: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
10479: {
10480: noffset=noffset+2;
1.304 brouard 10481: 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);
10482: 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 10483: fflush(ficlog); return 1;
10484: }
10485: else if( line[0] == 0 && line[1] == 0)
10486: {
10487: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10488: noffset=noffset+4;
1.304 brouard 10489: 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);
10490: 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 10491: fflush(ficlog); return 1;
10492: }
10493: } else{
10494: ;/*printf(" Not a BOM file\n");*/
10495: }
10496: /* If line starts with a # it is a comment */
10497: if (line[noffset] == '#') {
10498: linei=linei+1;
10499: break;
10500: }else{
10501: break;
10502: }
10503: }
10504: fclose(fic);
10505: if((fic=fopen(datafile,"r"))==NULL) {
10506: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10507: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
10508: }
10509: /* Not a Bom file */
10510:
1.136 brouard 10511: i=1;
10512: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
10513: linei=linei+1;
10514: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
10515: if(line[j] == '\t')
10516: line[j] = ' ';
10517: }
10518: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
10519: ;
10520: };
10521: line[j+1]=0; /* Trims blanks at end of line */
10522: if(line[0]=='#'){
10523: fprintf(ficlog,"Comment line\n%s\n",line);
10524: printf("Comment line\n%s\n",line);
10525: continue;
10526: }
10527: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 10528: strcpy(line, linetmp);
1.223 brouard 10529:
10530: /* Loops on waves */
10531: for (j=maxwav;j>=1;j--){
10532: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 10533: cutv(stra, strb, line, ' ');
10534: if(strb[0]=='.') { /* Missing value */
10535: lval=-1;
10536: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
1.341 brouard 10537: cotvar[j][ncovcol+nqv+ntv+iv][i]=-1; /* For performance reasons */
1.238 brouard 10538: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
10539: 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);
10540: 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);
10541: return 1;
10542: }
10543: }else{
10544: errno=0;
10545: /* what_kind_of_number(strb); */
10546: dval=strtod(strb,&endptr);
10547: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
10548: /* if(strb != endptr && *endptr == '\0') */
10549: /* dval=dlval; */
10550: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10551: if( strb[0]=='\0' || (*endptr != '\0')){
10552: 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);
10553: 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);
10554: return 1;
10555: }
10556: cotqvar[j][iv][i]=dval;
1.341 brouard 10557: cotvar[j][ncovcol+nqv+ntv+iv][i]=dval; /* because cotvar starts now at first ntv */
1.238 brouard 10558: }
10559: strcpy(line,stra);
1.223 brouard 10560: }/* end loop ntqv */
1.225 brouard 10561:
1.223 brouard 10562: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 10563: cutv(stra, strb, line, ' ');
10564: if(strb[0]=='.') { /* Missing value */
10565: lval=-1;
10566: }else{
10567: errno=0;
10568: lval=strtol(strb,&endptr,10);
10569: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10570: if( strb[0]=='\0' || (*endptr != '\0')){
10571: 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);
10572: 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);
10573: return 1;
10574: }
10575: }
10576: if(lval <-1 || lval >1){
10577: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10578: 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 10579: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10580: For example, for multinomial values like 1, 2 and 3,\n \
10581: build V1=0 V2=0 for the reference value (1),\n \
10582: V1=1 V2=0 for (2) \n \
1.223 brouard 10583: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10584: output of IMaCh is often meaningless.\n \
1.319 brouard 10585: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 10586: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10587: 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 10588: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10589: For example, for multinomial values like 1, 2 and 3,\n \
10590: build V1=0 V2=0 for the reference value (1),\n \
10591: V1=1 V2=0 for (2) \n \
1.223 brouard 10592: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10593: output of IMaCh is often meaningless.\n \
1.319 brouard 10594: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 10595: return 1;
10596: }
1.341 brouard 10597: cotvar[j][ncovcol+nqv+iv][i]=(double)(lval);
1.238 brouard 10598: strcpy(line,stra);
1.223 brouard 10599: }/* end loop ntv */
1.225 brouard 10600:
1.223 brouard 10601: /* Statuses at wave */
1.137 brouard 10602: cutv(stra, strb, line, ' ');
1.223 brouard 10603: if(strb[0]=='.') { /* Missing value */
1.238 brouard 10604: lval=-1;
1.136 brouard 10605: }else{
1.238 brouard 10606: errno=0;
10607: lval=strtol(strb,&endptr,10);
10608: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10609: if( strb[0]=='\0' || (*endptr != '\0')){
10610: 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);
10611: 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);
10612: return 1;
10613: }
1.136 brouard 10614: }
1.225 brouard 10615:
1.136 brouard 10616: s[j][i]=lval;
1.225 brouard 10617:
1.223 brouard 10618: /* Date of Interview */
1.136 brouard 10619: strcpy(line,stra);
10620: cutv(stra, strb,line,' ');
1.169 brouard 10621: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10622: }
1.169 brouard 10623: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 10624: month=99;
10625: year=9999;
1.136 brouard 10626: }else{
1.225 brouard 10627: 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);
10628: 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);
10629: return 1;
1.136 brouard 10630: }
10631: anint[j][i]= (double) year;
1.302 brouard 10632: mint[j][i]= (double)month;
10633: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
10634: /* 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]); */
10635: /* 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]); */
10636: /* } */
1.136 brouard 10637: strcpy(line,stra);
1.223 brouard 10638: } /* End loop on waves */
1.225 brouard 10639:
1.223 brouard 10640: /* Date of death */
1.136 brouard 10641: cutv(stra, strb,line,' ');
1.169 brouard 10642: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10643: }
1.169 brouard 10644: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 10645: month=99;
10646: year=9999;
10647: }else{
1.141 brouard 10648: 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 10649: 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);
10650: return 1;
1.136 brouard 10651: }
10652: andc[i]=(double) year;
10653: moisdc[i]=(double) month;
10654: strcpy(line,stra);
10655:
1.223 brouard 10656: /* Date of birth */
1.136 brouard 10657: cutv(stra, strb,line,' ');
1.169 brouard 10658: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10659: }
1.169 brouard 10660: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 10661: month=99;
10662: year=9999;
10663: }else{
1.141 brouard 10664: 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);
10665: 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 10666: return 1;
1.136 brouard 10667: }
10668: if (year==9999) {
1.141 brouard 10669: 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);
10670: 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 10671: return 1;
10672:
1.136 brouard 10673: }
10674: annais[i]=(double)(year);
1.302 brouard 10675: moisnais[i]=(double)(month);
10676: for (j=1;j<=maxwav;j++){
10677: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
10678: 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]);
10679: 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]);
10680: }
10681: }
10682:
1.136 brouard 10683: strcpy(line,stra);
1.225 brouard 10684:
1.223 brouard 10685: /* Sample weight */
1.136 brouard 10686: cutv(stra, strb,line,' ');
10687: errno=0;
10688: dval=strtod(strb,&endptr);
10689: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 10690: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
10691: 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 10692: fflush(ficlog);
10693: return 1;
10694: }
10695: weight[i]=dval;
10696: strcpy(line,stra);
1.225 brouard 10697:
1.223 brouard 10698: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
10699: cutv(stra, strb, line, ' ');
10700: if(strb[0]=='.') { /* Missing value */
1.225 brouard 10701: lval=-1;
1.311 brouard 10702: coqvar[iv][i]=NAN;
10703: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 10704: }else{
1.225 brouard 10705: errno=0;
10706: /* what_kind_of_number(strb); */
10707: dval=strtod(strb,&endptr);
10708: /* if(strb != endptr && *endptr == '\0') */
10709: /* dval=dlval; */
10710: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10711: if( strb[0]=='\0' || (*endptr != '\0')){
10712: 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);
10713: 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);
10714: return 1;
10715: }
10716: coqvar[iv][i]=dval;
1.226 brouard 10717: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 10718: }
10719: strcpy(line,stra);
10720: }/* end loop nqv */
1.136 brouard 10721:
1.223 brouard 10722: /* Covariate values */
1.136 brouard 10723: for (j=ncovcol;j>=1;j--){
10724: cutv(stra, strb,line,' ');
1.223 brouard 10725: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 10726: lval=-1;
1.136 brouard 10727: }else{
1.225 brouard 10728: errno=0;
10729: lval=strtol(strb,&endptr,10);
10730: if( strb[0]=='\0' || (*endptr != '\0')){
10731: 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);
10732: 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);
10733: return 1;
10734: }
1.136 brouard 10735: }
10736: if(lval <-1 || lval >1){
1.225 brouard 10737: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10738: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10739: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10740: For example, for multinomial values like 1, 2 and 3,\n \
10741: build V1=0 V2=0 for the reference value (1),\n \
10742: V1=1 V2=0 for (2) \n \
1.136 brouard 10743: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10744: output of IMaCh is often meaningless.\n \
1.136 brouard 10745: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 10746: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10747: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10748: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10749: For example, for multinomial values like 1, 2 and 3,\n \
10750: build V1=0 V2=0 for the reference value (1),\n \
10751: V1=1 V2=0 for (2) \n \
1.136 brouard 10752: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10753: output of IMaCh is often meaningless.\n \
1.136 brouard 10754: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 10755: return 1;
1.136 brouard 10756: }
10757: covar[j][i]=(double)(lval);
10758: strcpy(line,stra);
10759: }
10760: lstra=strlen(stra);
1.225 brouard 10761:
1.136 brouard 10762: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
10763: stratrunc = &(stra[lstra-9]);
10764: num[i]=atol(stratrunc);
10765: }
10766: else
10767: num[i]=atol(stra);
10768: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
10769: 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;}*/
10770:
10771: i=i+1;
10772: } /* End loop reading data */
1.225 brouard 10773:
1.136 brouard 10774: *imax=i-1; /* Number of individuals */
10775: fclose(fic);
1.225 brouard 10776:
1.136 brouard 10777: return (0);
1.164 brouard 10778: /* endread: */
1.225 brouard 10779: printf("Exiting readdata: ");
10780: fclose(fic);
10781: return (1);
1.223 brouard 10782: }
1.126 brouard 10783:
1.234 brouard 10784: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 10785: char *p1 = *stri, *p2 = *stri;
1.235 brouard 10786: while (*p2 == ' ')
1.234 brouard 10787: p2++;
10788: /* while ((*p1++ = *p2++) !=0) */
10789: /* ; */
10790: /* do */
10791: /* while (*p2 == ' ') */
10792: /* p2++; */
10793: /* while (*p1++ == *p2++); */
10794: *stri=p2;
1.145 brouard 10795: }
10796:
1.330 brouard 10797: int decoderesult( char resultline[], int nres)
1.230 brouard 10798: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
10799: {
1.235 brouard 10800: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 10801: char resultsav[MAXLINE];
1.330 brouard 10802: /* int resultmodel[MAXLINE]; */
1.334 brouard 10803: /* int modelresult[MAXLINE]; */
1.230 brouard 10804: char stra[80], strb[80], strc[80], strd[80],stre[80];
10805:
1.234 brouard 10806: removefirstspace(&resultline);
1.332 brouard 10807: printf("decoderesult:%s\n",resultline);
1.230 brouard 10808:
1.332 brouard 10809: strcpy(resultsav,resultline);
1.342 brouard 10810: /* printf("Decoderesult resultsav=\"%s\" resultline=\"%s\"\n", resultsav, resultline); */
1.230 brouard 10811: if (strlen(resultsav) >1){
1.334 brouard 10812: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' in this resultline */
1.230 brouard 10813: }
1.253 brouard 10814: if(j == 0){ /* Resultline but no = */
10815: TKresult[nres]=0; /* Combination for the nresult and the model */
10816: return (0);
10817: }
1.234 brouard 10818: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.334 brouard 10819: 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);
10820: 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 10821: /* return 1;*/
1.234 brouard 10822: }
1.334 brouard 10823: for(k=1; k<=j;k++){ /* Loop on any covariate of the RESULT LINE */
1.234 brouard 10824: if(nbocc(resultsav,'=') >1){
1.318 brouard 10825: 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 10826: /* 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 10827: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.332 brouard 10828: /* If a blank, then strc="V4=" and strd='\0' */
10829: if(strc[0]=='\0'){
10830: printf("Error in resultline, probably a blank after the \"%s\", \"result:%s\", stra=\"%s\" resultsav=\"%s\"\n",strb,resultline, stra, resultsav);
10831: fprintf(ficlog,"Error in resultline, probably a blank after the \"V%s=\", resultline=%s\n",strb,resultline);
10832: return 1;
10833: }
1.234 brouard 10834: }else
10835: cutl(strc,strd,resultsav,'=');
1.318 brouard 10836: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 10837:
1.230 brouard 10838: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 10839: 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 10840: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
10841: /* cptcovsel++; */
10842: if (nbocc(stra,'=') >0)
10843: strcpy(resultsav,stra); /* and analyzes it */
10844: }
1.235 brouard 10845: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10846: /* 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 10847: 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 10848: if(Typevar[k1]==0){ /* Single covariate in model */
10849: /* 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 10850: match=0;
1.318 brouard 10851: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10852: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10853: modelresult[nres][k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10854: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10855: break;
10856: }
10857: }
10858: if(match == 0){
1.338 brouard 10859: 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]);
10860: 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 10861: return 1;
1.234 brouard 10862: }
1.332 brouard 10863: }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*/
10864: /* We feed resultmodel[k1]=k2; */
10865: match=0;
10866: 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 */
10867: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10868: 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 10869: resultmodel[nres][k1]=k2; /* Added here */
1.342 brouard 10870: /* printf("Decoderesult first modelresult[k2=%d]=%d (k1) V%d*AGE\n",k2,k1,Tvar[k1]); */
1.332 brouard 10871: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10872: break;
10873: }
10874: }
10875: if(match == 0){
1.338 brouard 10876: 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]);
10877: 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 10878: return 1;
10879: }
10880: }else if(Typevar[k1]==2){ /* Product No age We want to get the position in the resultline of the product in the model line*/
10881: /* resultmodel[nres][of such a Vn * Vm product k1] is not unique, so can't exist, we feed Tvard[k1][1] and [2] */
10882: match=0;
1.342 brouard 10883: /* 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 10884: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10885: if(Tvardk[k1][1]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10886: /* modelresult[k2]=k1; */
1.342 brouard 10887: /* printf("Decoderesult first Product modelresult[k2=%d]=%d (k1) V%d * \n",k2,k1,Tvarsel[k2]); */
1.332 brouard 10888: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10889: }
10890: }
10891: if(match == 0){
1.338 brouard 10892: 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);
10893: 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 10894: return 1;
10895: }
10896: match=0;
10897: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10898: if(Tvardk[k1][2]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10899: /* modelresult[k2]=k1;*/
1.342 brouard 10900: /* printf("Decoderesult second Product modelresult[k2=%d]=%d (k1) * V%d \n ",k2,k1,Tvarsel[k2]); */
1.332 brouard 10901: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10902: break;
10903: }
10904: }
10905: if(match == 0){
1.338 brouard 10906: 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);
10907: 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 10908: return 1;
10909: }
10910: }/* End of testing */
1.333 brouard 10911: }/* End loop cptcovt */
1.235 brouard 10912: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10913: /* Feeds resultmodel[nres][k1]=k2 for single covariate (k1) in the model equation */
1.334 brouard 10914: 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)
10915: * Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 10916: match=0;
1.318 brouard 10917: 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 10918: if(Typevar[k1]==0){ /* Single only */
1.237 brouard 10919: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.330 brouard 10920: 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 10921: 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 10922: ++match;
10923: }
10924: }
10925: }
10926: if(match == 0){
1.338 brouard 10927: printf("Error in result line: variable V%d is missing in model; result: %s, model=1+age+%s\n",Tvarsel[k2], resultline, model);
10928: 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 10929: return 1;
1.234 brouard 10930: }else if(match > 1){
1.338 brouard 10931: printf("Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
10932: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
1.310 brouard 10933: return 1;
1.234 brouard 10934: }
10935: }
1.334 brouard 10936: /* cptcovres=j /\* Number of variables in the resultline is equal to cptcovs and thus useless *\/ */
1.234 brouard 10937: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10938: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 brouard 10939: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
10940: /* 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*/
10941: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 10942: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
10943: /* 1 0 0 0 */
10944: /* 2 1 0 0 */
10945: /* 3 0 1 0 */
1.330 brouard 10946: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 10947: /* 5 0 0 1 */
1.330 brouard 10948: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 10949: /* 7 0 1 1 */
10950: /* 8 1 1 1 */
1.237 brouard 10951: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
10952: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
10953: /* V5*age V5 known which value for nres? */
10954: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.334 brouard 10955: 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.
10956: * loop on position k1 in the MODEL LINE */
1.331 brouard 10957: /* k counting number of combination of single dummies in the equation model */
10958: /* k4 counting single dummies in the equation model */
10959: /* k4q counting single quantitatives in the equation model */
1.334 brouard 10960: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Dummy and Single, k1 is sorting according to MODEL, but k3 to resultline */
10961: /* 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 10962: /* 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 10963: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
1.332 brouard 10964: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
10965: /* k3 is the position in the nres result line of the k1th variable of the model equation */
10966: /* Tvarsel[k3]: Name of the variable at the k3th position in the result line. */
10967: /* Tvalsel[k3]: Value of the variable at the k3th position in the result line. */
10968: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.334 brouard 10969: /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 10970: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1.330 brouard 10971: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.332 brouard 10972: k3= resultmodel[nres][k1]; /* From position k1 in model get position k3 in result line */
10973: /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
10974: 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 10975: 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 10976: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][Name]=Value; stores the value into the name of the variable. */
1.332 brouard 10977: /* Tinvresult[nres][4]=1 */
1.334 brouard 10978: /* Tresult[nres][k4+1]=Tvalsel[k3];/\* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) *\/ */
10979: Tresult[nres][k3]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
10980: /* Tvresult[nres][k4+1]=(int)Tvarsel[k3];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
10981: Tvresult[nres][k3]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
1.237 brouard 10982: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.334 brouard 10983: precov[nres][k1]=Tvalsel[k3]; /* Value from resultline of the variable at the k1 position in the model */
1.342 brouard 10984: /* 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 10985: k4++;;
1.331 brouard 10986: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Quantitative and single */
1.330 brouard 10987: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.332 brouard 10988: /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.330 brouard 10989: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
1.332 brouard 10990: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 5 =k3q */
10991: k2q=(int)Tvarsel[k3q]; /* Name of variable at k3q th position in the resultline */
10992: /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 10993: /* Tqresult[nres][k4q+1]=Tvalsel[k3q]; /\* Tqresult[nres][1]=25.1 *\/ */
10994: /* Tvresult[nres][k4q+1]=(int)Tvarsel[k3q];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
10995: /* Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /\* Tvqresult[nres][1]=5 *\/ */
10996: Tqresult[nres][k3q]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
10997: Tvresult[nres][k3q]=(int)Tvarsel[k3q];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
10998: Tvqresult[nres][k3q]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
1.237 brouard 10999: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 brouard 11000: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 11001: precov[nres][k1]=Tvalsel[k3q];
1.342 brouard 11002: /* 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 11003: k4q++;;
1.331 brouard 11004: }else if( Dummy[k1]==2 ){ /* For dummy with age product */
11005: /* Tvar[k1]; */ /* Age variable */
1.332 brouard 11006: /* Wrong we want the value of variable name Tvar[k1] */
11007:
11008: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
1.331 brouard 11009: 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 11010: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][4]=1 */
1.332 brouard 11011: precov[nres][k1]=Tvalsel[k3];
1.342 brouard 11012: /* 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 11013: }else if( Dummy[k1]==3 ){ /* For quant with age product */
1.332 brouard 11014: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
1.331 brouard 11015: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 11016: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* TinvDoQresult[nres][5]=25.1 */
1.332 brouard 11017: precov[nres][k1]=Tvalsel[k3q];
1.342 brouard 11018: /* 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 11019: }else if(Typevar[k1]==2 ){ /* For product quant or dummy (not with age) */
1.332 brouard 11020: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
1.342 brouard 11021: /* 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 11022: }else{
1.332 brouard 11023: printf("Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
11024: fprintf(ficlog,"Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 11025: }
11026: }
1.234 brouard 11027:
1.334 brouard 11028: TKresult[nres]=++k; /* Number of combinations of dummies for the nresult and the model =Tvalsel[k3]*pow(2,k4) + 1*/
1.230 brouard 11029: return (0);
11030: }
1.235 brouard 11031:
1.230 brouard 11032: int decodemodel( char model[], int lastobs)
11033: /**< This routine decodes the model and returns:
1.224 brouard 11034: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
11035: * - nagesqr = 1 if age*age in the model, otherwise 0.
11036: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
11037: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
11038: * - cptcovage number of covariates with age*products =2
11039: * - cptcovs number of simple covariates
1.339 brouard 11040: * ncovcolt=ncovcol+nqv+ntv+nqtv total of covariates in the data, not in the model equation
1.224 brouard 11041: * - 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 11042: * which is a new column after the 9 (ncovcol+nqv+ntv+nqtv) variables.
1.319 brouard 11043: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 11044: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
11045: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
11046: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
11047: */
1.319 brouard 11048: /* 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 11049: {
1.238 brouard 11050: int i, j, k, ks, v;
1.227 brouard 11051: int j1, k1, k2, k3, k4;
1.136 brouard 11052: char modelsav[80];
1.145 brouard 11053: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 11054: char *strpt;
1.136 brouard 11055:
1.145 brouard 11056: /*removespace(model);*/
1.136 brouard 11057: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 11058: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 11059: if (strstr(model,"AGE") !=0){
1.192 brouard 11060: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
11061: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 11062: return 1;
11063: }
1.141 brouard 11064: if (strstr(model,"v") !=0){
1.338 brouard 11065: printf("Error. 'v' must be in upper case 'V' model=1+age+%s ",model);
11066: fprintf(ficlog,"Error. 'v' must be in upper case model=1+age+%s ",model);fflush(ficlog);
1.141 brouard 11067: return 1;
11068: }
1.187 brouard 11069: strcpy(modelsav,model);
11070: if ((strpt=strstr(model,"age*age")) !=0){
1.338 brouard 11071: printf(" strpt=%s, model=1+age+%s\n",strpt, model);
1.187 brouard 11072: if(strpt != model){
1.338 brouard 11073: printf("Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 11074: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 11075: corresponding column of parameters.\n",model);
1.338 brouard 11076: fprintf(ficlog,"Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 11077: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 11078: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 11079: return 1;
1.225 brouard 11080: }
1.187 brouard 11081: nagesqr=1;
11082: if (strstr(model,"+age*age") !=0)
1.234 brouard 11083: substrchaine(modelsav, model, "+age*age");
1.187 brouard 11084: else if (strstr(model,"age*age+") !=0)
1.234 brouard 11085: substrchaine(modelsav, model, "age*age+");
1.187 brouard 11086: else
1.234 brouard 11087: substrchaine(modelsav, model, "age*age");
1.187 brouard 11088: }else
11089: nagesqr=0;
11090: if (strlen(modelsav) >1){
11091: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
11092: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 11093: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 11094: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 11095: * cst, age and age*age
11096: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
11097: /* including age products which are counted in cptcovage.
11098: * but the covariates which are products must be treated
11099: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 11100: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
11101: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 11102:
11103:
1.187 brouard 11104: /* Design
11105: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
11106: * < ncovcol=8 >
11107: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
11108: * k= 1 2 3 4 5 6 7 8
11109: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
11110: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 11111: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
11112: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 11113: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
11114: * Tage[++cptcovage]=k
11115: * if products, new covar are created after ncovcol with k1
11116: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
11117: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
11118: * 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
11119: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
11120: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
11121: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
11122: * < ncovcol=8 >
11123: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
11124: * k= 1 2 3 4 5 6 7 8 9 10 11 12
11125: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 11126: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 11127: * p Tprod[1]@2={ 6, 5}
11128: *p Tvard[1][1]@4= {7, 8, 5, 6}
11129: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
11130: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 11131: *How to reorganize? Tvars(orted)
1.187 brouard 11132: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
11133: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
11134: * {2, 1, 4, 8, 5, 6, 3, 7}
11135: * Struct []
11136: */
1.225 brouard 11137:
1.187 brouard 11138: /* This loop fills the array Tvar from the string 'model'.*/
11139: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
11140: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
11141: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
11142: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
11143: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
11144: /* k=1 Tvar[1]=2 (from V2) */
11145: /* k=5 Tvar[5] */
11146: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 11147: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 11148: /* } */
1.198 brouard 11149: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 11150: /*
11151: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 11152: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
11153: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
11154: }
1.187 brouard 11155: cptcovage=0;
1.319 brouard 11156: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
11157: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
11158: 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" */
11159: if (nbocc(modelsav,'+')==0)
11160: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 11161: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
11162: /*scanf("%d",i);*/
1.319 brouard 11163: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
11164: 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 11165: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
11166: /* covar is not filled and then is empty */
11167: cptcovprod--;
11168: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 11169: 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 11170: Typevar[k]=1; /* 1 for age product */
1.319 brouard 11171: cptcovage++; /* Counts the number of covariates which include age as a product */
11172: 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 11173: /*printf("stre=%s ", stre);*/
11174: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
11175: cptcovprod--;
11176: cutl(stre,strb,strc,'V');
11177: Tvar[k]=atoi(stre);
11178: Typevar[k]=1; /* 1 for age product */
11179: cptcovage++;
11180: Tage[cptcovage]=k;
11181: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
11182: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
11183: cptcovn++;
11184: cptcovprodnoage++;k1++;
11185: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
1.339 brouard 11186: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* ncovcolt+k1; For model-covariate k tells which data-covariate to use but
1.234 brouard 11187: because this model-covariate is a construction we invent a new column
11188: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.335 brouard 11189: If already ncovcol=4 and model= V2 + V1 + V1*V4 + age*V3 + V3*V2
1.319 brouard 11190: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
1.339 brouard 11191: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=3 etc */
1.335 brouard 11192: /* Please remark that the new variables are model dependent */
11193: /* If we have 4 variable but the model uses only 3, like in
11194: * model= V1 + age*V1 + V2 + V3 + age*V2 + age*V3 + V1*V2 + V1*V3
11195: * k= 1 2 3 4 5 6 7 8
11196: * Tvar[k]=1 1 2 3 2 3 (5 6) (and not 4 5 because of V4 missing)
11197: * Tage[kk] [1]= 2 [2]=5 [3]=6 kk=1 to cptcovage=3
11198: * Tvar[Tage[kk]][1]=2 [2]=2 [3]=3
11199: */
1.339 brouard 11200: Typevar[k]=2; /* 2 for product */
1.234 brouard 11201: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
11202: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 11203: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 11204: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
1.330 brouard 11205: Tvardk[k][1] =atoi(strc); /* m 1 for V1*/
1.234 brouard 11206: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
1.330 brouard 11207: Tvardk[k][2] =atoi(stre); /* n 4 for V4*/
1.234 brouard 11208: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
11209: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
11210: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 11211: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 11212: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
1.339 brouard 11213: 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 */
11214: for (i=1; i<=lastobs;i++){/* For fixed product */
1.234 brouard 11215: /* Computes the new covariate which is a product of
11216: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
1.339 brouard 11217: covar[ncovcolt+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
11218: }
11219: } /*End of FixedV */
1.234 brouard 11220: } /* End age is not in the model */
11221: } /* End if model includes a product */
1.319 brouard 11222: else { /* not a product */
1.234 brouard 11223: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
11224: /* scanf("%d",i);*/
11225: cutl(strd,strc,strb,'V');
11226: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
11227: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
11228: Tvar[k]=atoi(strd);
11229: Typevar[k]=0; /* 0 for simple covariates */
11230: }
11231: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 11232: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 11233: scanf("%d",i);*/
1.187 brouard 11234: } /* end of loop + on total covariates */
11235: } /* end if strlen(modelsave == 0) age*age might exist */
11236: } /* end if strlen(model == 0) */
1.136 brouard 11237:
11238: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
11239: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 11240:
1.136 brouard 11241: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 11242: printf("cptcovprod=%d ", cptcovprod);
11243: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
11244: scanf("%d ",i);*/
11245:
11246:
1.230 brouard 11247: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
11248: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 11249: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
11250: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
11251: k = 1 2 3 4 5 6 7 8 9
11252: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 11253: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 11254: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
11255: Dummy[k] 1 0 0 0 3 1 1 2 3
11256: Tmodelind[combination of covar]=k;
1.225 brouard 11257: */
11258: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 11259: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 11260: /* 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 11261: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 11262: printf("Model=1+age+%s\n\
1.227 brouard 11263: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
11264: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
11265: 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 11266: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 11267: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
11268: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
11269: 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 11270: for(k=-1;k<=NCOVMAX; k++){ Fixed[k]=0; Dummy[k]=0;}
11271: for(k=1;k<=NCOVMAX; k++){TvarFind[k]=0; TvarVind[k]=0;}
1.343 ! brouard 11272: 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 11273: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 11274: Fixed[k]= 0;
11275: Dummy[k]= 0;
1.225 brouard 11276: ncoveff++;
1.232 brouard 11277: ncovf++;
1.234 brouard 11278: nsd++;
11279: modell[k].maintype= FTYPE;
11280: TvarsD[nsd]=Tvar[k];
11281: TvarsDind[nsd]=k;
1.330 brouard 11282: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 11283: TvarF[ncovf]=Tvar[k];
11284: TvarFind[ncovf]=k;
11285: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
11286: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.339 brouard 11287: /* }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /\* Product of fixed dummy (<=ncovcol) covariates For a fixed product k is higher than ncovcol *\/ */
11288: }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 11289: Fixed[k]= 0;
11290: Dummy[k]= 0;
11291: ncoveff++;
11292: ncovf++;
11293: modell[k].maintype= FTYPE;
11294: TvarF[ncovf]=Tvar[k];
1.330 brouard 11295: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
1.234 brouard 11296: TvarFind[ncovf]=k;
1.230 brouard 11297: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 11298: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 11299: }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 11300: Fixed[k]= 0;
11301: Dummy[k]= 1;
1.230 brouard 11302: nqfveff++;
1.234 brouard 11303: modell[k].maintype= FTYPE;
11304: modell[k].subtype= FQ;
11305: nsq++;
1.334 brouard 11306: TvarsQ[nsq]=Tvar[k]; /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary see below */
11307: TvarsQind[nsq]=k; /* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.232 brouard 11308: ncovf++;
1.234 brouard 11309: TvarF[ncovf]=Tvar[k];
11310: TvarFind[ncovf]=k;
1.231 brouard 11311: 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 11312: 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 11313: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.339 brouard 11314: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11315: /* model V1+V3+age*V1+age*V3+V1*V3 */
11316: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11317: ncovvt++;
11318: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
11319: TvarVVind[ncovvt]=k; /* TvarVVind[1]=2 (second position in the model equation */
11320:
1.227 brouard 11321: Fixed[k]= 1;
11322: Dummy[k]= 0;
1.225 brouard 11323: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 11324: modell[k].maintype= VTYPE;
11325: modell[k].subtype= VD;
11326: nsd++;
11327: TvarsD[nsd]=Tvar[k];
11328: TvarsDind[nsd]=k;
1.330 brouard 11329: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 11330: ncovv++; /* Only simple time varying variables */
11331: TvarV[ncovv]=Tvar[k];
1.242 brouard 11332: 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 11333: 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 */
11334: 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 11335: 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);
11336: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 11337: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.339 brouard 11338: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11339: /* model V1+V3+age*V1+age*V3+V1*V3 */
11340: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11341: ncovvt++;
11342: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
11343: TvarVVind[ncovvt]=k; /* TvarVV[1]=V3 (first time varying in the model equation */
11344:
1.234 brouard 11345: Fixed[k]= 1;
11346: Dummy[k]= 1;
11347: nqtveff++;
11348: modell[k].maintype= VTYPE;
11349: modell[k].subtype= VQ;
11350: ncovv++; /* Only simple time varying variables */
11351: nsq++;
1.334 brouard 11352: 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) */
11353: 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 11354: TvarV[ncovv]=Tvar[k];
1.242 brouard 11355: 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 11356: 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 */
11357: 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 11358: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
11359: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
1.342 brouard 11360: /* 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); */
11361: /* printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv); */
1.227 brouard 11362: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 11363: ncova++;
11364: TvarA[ncova]=Tvar[k];
11365: TvarAind[ncova]=k;
1.231 brouard 11366: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 11367: Fixed[k]= 2;
11368: Dummy[k]= 2;
11369: modell[k].maintype= ATYPE;
11370: modell[k].subtype= APFD;
11371: /* ncoveff++; */
1.227 brouard 11372: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 11373: Fixed[k]= 2;
11374: Dummy[k]= 3;
11375: modell[k].maintype= ATYPE;
11376: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
11377: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 11378: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 11379: Fixed[k]= 3;
11380: Dummy[k]= 2;
11381: modell[k].maintype= ATYPE;
11382: modell[k].subtype= APVD; /* Product age * varying dummy */
11383: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 11384: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 11385: Fixed[k]= 3;
11386: Dummy[k]= 3;
11387: modell[k].maintype= ATYPE;
11388: modell[k].subtype= APVQ; /* Product age * varying quantitative */
11389: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 11390: }
1.339 brouard 11391: }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 */
11392: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11393: /* model V1+V3+age*V1+age*V3+V1*V3 */
11394: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11395: 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 */
11396: ncovvt++;
11397: TvarVV[ncovvt]=Tvard[k1][1]; /* TvarVV[2]=V1 (because TvarVV[1] was V3, first time varying covariates */
11398: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
11399: ncovvt++;
11400: TvarVV[ncovvt]=Tvard[k1][2]; /* TvarVV[3]=V3 */
11401: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
11402:
11403:
11404: if(Tvard[k1][1] <=ncovcol){ /* Vn is dummy fixed, (Tvard[1][1]=V1), (Tvard[1][1]=V3 time varying) */
11405: if(Tvard[k1][2] <=ncovcol){ /* Vm is dummy fixed */
1.240 brouard 11406: Fixed[k]= 1;
11407: Dummy[k]= 0;
11408: modell[k].maintype= FTYPE;
11409: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
11410: ncovf++; /* Fixed variables without age */
11411: TvarF[ncovf]=Tvar[k];
11412: TvarFind[ncovf]=k;
1.339 brouard 11413: }else if(Tvard[k1][2] <=ncovcol+nqv){ /* Vm is quanti fixed */
11414: Fixed[k]= 0; /* Fixed product */
1.240 brouard 11415: Dummy[k]= 1;
11416: modell[k].maintype= FTYPE;
11417: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
11418: ncovf++; /* Varying variables without age */
11419: TvarF[ncovf]=Tvar[k];
11420: TvarFind[ncovf]=k;
1.339 brouard 11421: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is a time varying dummy covariate */
1.240 brouard 11422: Fixed[k]= 1;
11423: Dummy[k]= 0;
11424: modell[k].maintype= VTYPE;
11425: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
11426: ncovv++; /* Varying variables without age */
1.339 brouard 11427: TvarV[ncovv]=Tvar[k]; /* TvarV[1]=Tvar[5]=5 because there is a V4 */
11428: TvarVind[ncovv]=k;/* TvarVind[1]=5 */
11429: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is a time varying quantitative covariate */
1.240 brouard 11430: Fixed[k]= 1;
11431: Dummy[k]= 1;
11432: modell[k].maintype= VTYPE;
11433: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
11434: ncovv++; /* Varying variables without age */
11435: TvarV[ncovv]=Tvar[k];
11436: TvarVind[ncovv]=k;
11437: }
1.339 brouard 11438: }else if(Tvard[k1][1] <=ncovcol+nqv){ /* Vn is fixed quanti */
11439: if(Tvard[k1][2] <=ncovcol){ /* Vm is fixed dummy */
11440: Fixed[k]= 0; /* Fixed product */
1.240 brouard 11441: Dummy[k]= 1;
11442: modell[k].maintype= FTYPE;
11443: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
11444: ncovf++; /* Fixed variables without age */
11445: TvarF[ncovf]=Tvar[k];
11446: TvarFind[ncovf]=k;
1.339 brouard 11447: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is time varying */
1.240 brouard 11448: Fixed[k]= 1;
11449: Dummy[k]= 1;
11450: modell[k].maintype= VTYPE;
11451: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
11452: ncovv++; /* Varying variables without age */
11453: TvarV[ncovv]=Tvar[k];
11454: TvarVind[ncovv]=k;
1.339 brouard 11455: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is time varying quanti */
1.240 brouard 11456: Fixed[k]= 1;
11457: Dummy[k]= 1;
11458: modell[k].maintype= VTYPE;
11459: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
11460: ncovv++; /* Varying variables without age */
11461: TvarV[ncovv]=Tvar[k];
11462: TvarVind[ncovv]=k;
11463: ncovv++; /* Varying variables without age */
11464: TvarV[ncovv]=Tvar[k];
11465: TvarVind[ncovv]=k;
11466: }
1.339 brouard 11467: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){ /* Vn is time varying dummy */
1.240 brouard 11468: if(Tvard[k1][2] <=ncovcol){
11469: Fixed[k]= 1;
11470: Dummy[k]= 1;
11471: modell[k].maintype= VTYPE;
11472: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
11473: ncovv++; /* Varying variables without age */
11474: TvarV[ncovv]=Tvar[k];
11475: TvarVind[ncovv]=k;
11476: }else if(Tvard[k1][2] <=ncovcol+nqv){
11477: Fixed[k]= 1;
11478: Dummy[k]= 1;
11479: modell[k].maintype= VTYPE;
11480: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
11481: ncovv++; /* Varying variables without age */
11482: TvarV[ncovv]=Tvar[k];
11483: TvarVind[ncovv]=k;
11484: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11485: Fixed[k]= 1;
11486: Dummy[k]= 0;
11487: modell[k].maintype= VTYPE;
11488: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
11489: ncovv++; /* Varying variables without age */
11490: TvarV[ncovv]=Tvar[k];
11491: TvarVind[ncovv]=k;
11492: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11493: Fixed[k]= 1;
11494: Dummy[k]= 1;
11495: modell[k].maintype= VTYPE;
11496: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
11497: ncovv++; /* Varying variables without age */
11498: TvarV[ncovv]=Tvar[k];
11499: TvarVind[ncovv]=k;
11500: }
1.339 brouard 11501: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){ /* Vn is time varying quanti */
1.240 brouard 11502: if(Tvard[k1][2] <=ncovcol){
11503: Fixed[k]= 1;
11504: Dummy[k]= 1;
11505: modell[k].maintype= VTYPE;
11506: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
11507: ncovv++; /* Varying variables without age */
11508: TvarV[ncovv]=Tvar[k];
11509: TvarVind[ncovv]=k;
11510: }else if(Tvard[k1][2] <=ncovcol+nqv){
11511: Fixed[k]= 1;
11512: Dummy[k]= 1;
11513: modell[k].maintype= VTYPE;
11514: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
11515: ncovv++; /* Varying variables without age */
11516: TvarV[ncovv]=Tvar[k];
11517: TvarVind[ncovv]=k;
11518: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11519: Fixed[k]= 1;
11520: Dummy[k]= 1;
11521: modell[k].maintype= VTYPE;
11522: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
11523: ncovv++; /* Varying variables without age */
11524: TvarV[ncovv]=Tvar[k];
11525: TvarVind[ncovv]=k;
11526: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11527: Fixed[k]= 1;
11528: Dummy[k]= 1;
11529: modell[k].maintype= VTYPE;
11530: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
11531: ncovv++; /* Varying variables without age */
11532: TvarV[ncovv]=Tvar[k];
11533: TvarVind[ncovv]=k;
11534: }
1.227 brouard 11535: }else{
1.240 brouard 11536: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11537: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11538: } /*end k1*/
1.225 brouard 11539: }else{
1.226 brouard 11540: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
11541: 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 11542: }
1.342 brouard 11543: /* 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]); */
11544: /* printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype); */
1.227 brouard 11545: 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]);
11546: }
11547: /* Searching for doublons in the model */
11548: for(k1=1; k1<= cptcovt;k1++){
11549: for(k2=1; k2 <k1;k2++){
1.285 brouard 11550: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
11551: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 11552: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
11553: if(Tvar[k1]==Tvar[k2]){
1.338 brouard 11554: 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]);
11555: 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 11556: return(1);
11557: }
11558: }else if (Typevar[k1] ==2){
11559: k3=Tposprod[k1];
11560: k4=Tposprod[k2];
11561: 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 11562: 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]]);
11563: 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 11564: return(1);
11565: }
11566: }
1.227 brouard 11567: }
11568: }
1.225 brouard 11569: }
11570: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
11571: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 11572: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
11573: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 11574: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 11575: /*endread:*/
1.225 brouard 11576: printf("Exiting decodemodel: ");
11577: return (1);
1.136 brouard 11578: }
11579:
1.169 brouard 11580: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 11581: {/* Check ages at death */
1.136 brouard 11582: int i, m;
1.218 brouard 11583: int firstone=0;
11584:
1.136 brouard 11585: for (i=1; i<=imx; i++) {
11586: for(m=2; (m<= maxwav); m++) {
11587: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
11588: anint[m][i]=9999;
1.216 brouard 11589: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
11590: s[m][i]=-1;
1.136 brouard 11591: }
11592: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 11593: *nberr = *nberr + 1;
1.218 brouard 11594: if(firstone == 0){
11595: firstone=1;
1.260 brouard 11596: 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 11597: }
1.262 brouard 11598: 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 11599: s[m][i]=-1; /* Droping the death status */
1.136 brouard 11600: }
11601: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 11602: (*nberr)++;
1.259 brouard 11603: 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 11604: 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 11605: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 11606: }
11607: }
11608: }
11609:
11610: for (i=1; i<=imx; i++) {
11611: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
11612: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 11613: 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 11614: if (s[m][i] >= nlstate+1) {
1.169 brouard 11615: if(agedc[i]>0){
11616: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 11617: agev[m][i]=agedc[i];
1.214 brouard 11618: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 11619: }else {
1.136 brouard 11620: if ((int)andc[i]!=9999){
11621: nbwarn++;
11622: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
11623: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
11624: agev[m][i]=-1;
11625: }
11626: }
1.169 brouard 11627: } /* agedc > 0 */
1.214 brouard 11628: } /* end if */
1.136 brouard 11629: else if(s[m][i] !=9){ /* Standard case, age in fractional
11630: years but with the precision of a month */
11631: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
11632: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
11633: agev[m][i]=1;
11634: else if(agev[m][i] < *agemin){
11635: *agemin=agev[m][i];
11636: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
11637: }
11638: else if(agev[m][i] >*agemax){
11639: *agemax=agev[m][i];
1.156 brouard 11640: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 11641: }
11642: /*agev[m][i]=anint[m][i]-annais[i];*/
11643: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 11644: } /* en if 9*/
1.136 brouard 11645: else { /* =9 */
1.214 brouard 11646: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 11647: agev[m][i]=1;
11648: s[m][i]=-1;
11649: }
11650: }
1.214 brouard 11651: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 11652: agev[m][i]=1;
1.214 brouard 11653: else{
11654: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11655: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11656: agev[m][i]=0;
11657: }
11658: } /* End for lastpass */
11659: }
1.136 brouard 11660:
11661: for (i=1; i<=imx; i++) {
11662: for(m=firstpass; (m<=lastpass); m++){
11663: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 11664: (*nberr)++;
1.136 brouard 11665: 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);
11666: 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);
11667: return 1;
11668: }
11669: }
11670: }
11671:
11672: /*for (i=1; i<=imx; i++){
11673: for (m=firstpass; (m<lastpass); m++){
11674: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
11675: }
11676:
11677: }*/
11678:
11679:
1.139 brouard 11680: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
11681: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 11682:
11683: return (0);
1.164 brouard 11684: /* endread:*/
1.136 brouard 11685: printf("Exiting calandcheckages: ");
11686: return (1);
11687: }
11688:
1.172 brouard 11689: #if defined(_MSC_VER)
11690: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11691: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11692: //#include "stdafx.h"
11693: //#include <stdio.h>
11694: //#include <tchar.h>
11695: //#include <windows.h>
11696: //#include <iostream>
11697: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
11698:
11699: LPFN_ISWOW64PROCESS fnIsWow64Process;
11700:
11701: BOOL IsWow64()
11702: {
11703: BOOL bIsWow64 = FALSE;
11704:
11705: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
11706: // (HANDLE, PBOOL);
11707:
11708: //LPFN_ISWOW64PROCESS fnIsWow64Process;
11709:
11710: HMODULE module = GetModuleHandle(_T("kernel32"));
11711: const char funcName[] = "IsWow64Process";
11712: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
11713: GetProcAddress(module, funcName);
11714:
11715: if (NULL != fnIsWow64Process)
11716: {
11717: if (!fnIsWow64Process(GetCurrentProcess(),
11718: &bIsWow64))
11719: //throw std::exception("Unknown error");
11720: printf("Unknown error\n");
11721: }
11722: return bIsWow64 != FALSE;
11723: }
11724: #endif
1.177 brouard 11725:
1.191 brouard 11726: void syscompilerinfo(int logged)
1.292 brouard 11727: {
11728: #include <stdint.h>
11729:
11730: /* #include "syscompilerinfo.h"*/
1.185 brouard 11731: /* command line Intel compiler 32bit windows, XP compatible:*/
11732: /* /GS /W3 /Gy
11733: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
11734: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
11735: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 11736: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
11737: */
11738: /* 64 bits */
1.185 brouard 11739: /*
11740: /GS /W3 /Gy
11741: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
11742: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
11743: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
11744: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
11745: /* Optimization are useless and O3 is slower than O2 */
11746: /*
11747: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
11748: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
11749: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
11750: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
11751: */
1.186 brouard 11752: /* Link is */ /* /OUT:"visual studio
1.185 brouard 11753: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
11754: /PDB:"visual studio
11755: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
11756: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
11757: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
11758: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
11759: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
11760: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
11761: uiAccess='false'"
11762: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
11763: /NOLOGO /TLBID:1
11764: */
1.292 brouard 11765:
11766:
1.177 brouard 11767: #if defined __INTEL_COMPILER
1.178 brouard 11768: #if defined(__GNUC__)
11769: struct utsname sysInfo; /* For Intel on Linux and OS/X */
11770: #endif
1.177 brouard 11771: #elif defined(__GNUC__)
1.179 brouard 11772: #ifndef __APPLE__
1.174 brouard 11773: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 11774: #endif
1.177 brouard 11775: struct utsname sysInfo;
1.178 brouard 11776: int cross = CROSS;
11777: if (cross){
11778: printf("Cross-");
1.191 brouard 11779: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 11780: }
1.174 brouard 11781: #endif
11782:
1.191 brouard 11783: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 11784: #if defined(__clang__)
1.191 brouard 11785: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 11786: #endif
11787: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 11788: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 11789: #endif
11790: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 11791: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 11792: #endif
11793: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 11794: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 11795: #endif
11796: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 11797: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 11798: #endif
11799: #if defined(_MSC_VER)
1.191 brouard 11800: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 11801: #endif
11802: #if defined(__PGI)
1.191 brouard 11803: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 11804: #endif
11805: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 11806: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 11807: #endif
1.191 brouard 11808: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 11809:
1.167 brouard 11810: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
11811: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
11812: // Windows (x64 and x86)
1.191 brouard 11813: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 11814: #elif __unix__ // all unices, not all compilers
11815: // Unix
1.191 brouard 11816: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 11817: #elif __linux__
11818: // linux
1.191 brouard 11819: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 11820: #elif __APPLE__
1.174 brouard 11821: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 11822: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 11823: #endif
11824:
11825: /* __MINGW32__ */
11826: /* __CYGWIN__ */
11827: /* __MINGW64__ */
11828: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
11829: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
11830: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
11831: /* _WIN64 // Defined for applications for Win64. */
11832: /* _M_X64 // Defined for compilations that target x64 processors. */
11833: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 11834:
1.167 brouard 11835: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 11836: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 11837: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 11838: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 11839: #else
1.191 brouard 11840: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 11841: #endif
11842:
1.169 brouard 11843: #if defined(__GNUC__)
11844: # if defined(__GNUC_PATCHLEVEL__)
11845: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11846: + __GNUC_MINOR__ * 100 \
11847: + __GNUC_PATCHLEVEL__)
11848: # else
11849: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11850: + __GNUC_MINOR__ * 100)
11851: # endif
1.174 brouard 11852: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 11853: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 11854:
11855: if (uname(&sysInfo) != -1) {
11856: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 11857: 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 11858: }
11859: else
11860: perror("uname() error");
1.179 brouard 11861: //#ifndef __INTEL_COMPILER
11862: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 11863: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 11864: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 11865: #endif
1.169 brouard 11866: #endif
1.172 brouard 11867:
1.286 brouard 11868: // void main ()
1.172 brouard 11869: // {
1.169 brouard 11870: #if defined(_MSC_VER)
1.174 brouard 11871: if (IsWow64()){
1.191 brouard 11872: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
11873: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 11874: }
11875: else{
1.191 brouard 11876: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
11877: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 11878: }
1.172 brouard 11879: // printf("\nPress Enter to continue...");
11880: // getchar();
11881: // }
11882:
1.169 brouard 11883: #endif
11884:
1.167 brouard 11885:
1.219 brouard 11886: }
1.136 brouard 11887:
1.219 brouard 11888: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 11889: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.332 brouard 11890: /* Computes the prevalence limit for each combination of the dummy covariates */
1.235 brouard 11891: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 11892: /* double ftolpl = 1.e-10; */
1.180 brouard 11893: double age, agebase, agelim;
1.203 brouard 11894: double tot;
1.180 brouard 11895:
1.202 brouard 11896: strcpy(filerespl,"PL_");
11897: strcat(filerespl,fileresu);
11898: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 11899: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
11900: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 11901: }
1.288 brouard 11902: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
11903: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 11904: pstamp(ficrespl);
1.288 brouard 11905: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 11906: fprintf(ficrespl,"#Age ");
11907: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
11908: fprintf(ficrespl,"\n");
1.180 brouard 11909:
1.219 brouard 11910: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 11911:
1.219 brouard 11912: agebase=ageminpar;
11913: agelim=agemaxpar;
1.180 brouard 11914:
1.227 brouard 11915: /* i1=pow(2,ncoveff); */
1.234 brouard 11916: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 11917: if (cptcovn < 1){i1=1;}
1.180 brouard 11918:
1.337 brouard 11919: /* for(k=1; k<=i1;k++){ /\* For each combination k of dummy covariates in the model *\/ */
1.238 brouard 11920: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 11921: k=TKresult[nres];
1.338 brouard 11922: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 11923: /* if(i1 != 1 && TKresult[nres]!= k) /\* We found the combination k corresponding to the resultline value of dummies *\/ */
11924: /* continue; */
1.235 brouard 11925:
1.238 brouard 11926: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11927: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
11928: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
11929: /* k=k+1; */
11930: /* to clean */
1.332 brouard 11931: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11932: fprintf(ficrespl,"#******");
11933: printf("#******");
11934: fprintf(ficlog,"#******");
1.337 brouard 11935: 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 11936: /* fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Here problem for varying dummy*\/ */
1.337 brouard 11937: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11938: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11939: fprintf(ficrespl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11940: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11941: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11942: }
11943: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
11944: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11945: /* fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11946: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11947: /* } */
1.238 brouard 11948: fprintf(ficrespl,"******\n");
11949: printf("******\n");
11950: fprintf(ficlog,"******\n");
11951: if(invalidvarcomb[k]){
11952: printf("\nCombination (%d) ignored because no case \n",k);
11953: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
11954: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
11955: continue;
11956: }
1.219 brouard 11957:
1.238 brouard 11958: fprintf(ficrespl,"#Age ");
1.337 brouard 11959: /* for(j=1;j<=cptcoveff;j++) { */
11960: /* fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11961: /* } */
11962: for(j=1;j<=cptcovs;j++) { /* New the quanti variable is added */
11963: fprintf(ficrespl,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 11964: }
11965: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
11966: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 11967:
1.238 brouard 11968: for (age=agebase; age<=agelim; age++){
11969: /* for (age=agebase; age<=agebase; age++){ */
1.337 brouard 11970: /**< Computes the prevalence limit in each live state at age x and for covariate combination (k and) nres */
11971: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres); /* Nicely done */
1.238 brouard 11972: fprintf(ficrespl,"%.0f ",age );
1.337 brouard 11973: /* for(j=1;j<=cptcoveff;j++) */
11974: /* fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11975: for(j=1;j<=cptcovs;j++)
11976: fprintf(ficrespl,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 11977: tot=0.;
11978: for(i=1; i<=nlstate;i++){
11979: tot += prlim[i][i];
11980: fprintf(ficrespl," %.5f", prlim[i][i]);
11981: }
11982: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
11983: } /* Age */
11984: /* was end of cptcod */
1.337 brouard 11985: } /* nres */
11986: /* } /\* for each combination *\/ */
1.219 brouard 11987: return 0;
1.180 brouard 11988: }
11989:
1.218 brouard 11990: 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 11991: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 11992:
11993: /* Computes the back prevalence limit for any combination of covariate values
11994: * at any age between ageminpar and agemaxpar
11995: */
1.235 brouard 11996: int i, j, k, i1, nres=0 ;
1.217 brouard 11997: /* double ftolpl = 1.e-10; */
11998: double age, agebase, agelim;
11999: double tot;
1.218 brouard 12000: /* double ***mobaverage; */
12001: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 12002:
12003: strcpy(fileresplb,"PLB_");
12004: strcat(fileresplb,fileresu);
12005: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 12006: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
12007: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 12008: }
1.288 brouard 12009: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
12010: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 12011: pstamp(ficresplb);
1.288 brouard 12012: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 12013: fprintf(ficresplb,"#Age ");
12014: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
12015: fprintf(ficresplb,"\n");
12016:
1.218 brouard 12017:
12018: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
12019:
12020: agebase=ageminpar;
12021: agelim=agemaxpar;
12022:
12023:
1.227 brouard 12024: i1=pow(2,cptcoveff);
1.218 brouard 12025: if (cptcovn < 1){i1=1;}
1.227 brouard 12026:
1.238 brouard 12027: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.338 brouard 12028: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
12029: k=TKresult[nres];
12030: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
12031: /* if(i1 != 1 && TKresult[nres]!= k) */
12032: /* continue; */
12033: /* /\*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*\/ */
1.238 brouard 12034: fprintf(ficresplb,"#******");
12035: printf("#******");
12036: fprintf(ficlog,"#******");
1.338 brouard 12037: 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) */
12038: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12039: fprintf(ficresplb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12040: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12041: }
1.338 brouard 12042: /* for(j=1;j<=cptcoveff ;j++) {/\* all covariates *\/ */
12043: /* fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12044: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12045: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12046: /* } */
12047: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
12048: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12049: /* fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12050: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12051: /* } */
1.238 brouard 12052: fprintf(ficresplb,"******\n");
12053: printf("******\n");
12054: fprintf(ficlog,"******\n");
12055: if(invalidvarcomb[k]){
12056: printf("\nCombination (%d) ignored because no cases \n",k);
12057: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
12058: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
12059: continue;
12060: }
1.218 brouard 12061:
1.238 brouard 12062: fprintf(ficresplb,"#Age ");
1.338 brouard 12063: for(j=1;j<=cptcovs;j++) {
12064: fprintf(ficresplb,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12065: }
12066: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
12067: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 12068:
12069:
1.238 brouard 12070: for (age=agebase; age<=agelim; age++){
12071: /* for (age=agebase; age<=agebase; age++){ */
12072: if(mobilavproj > 0){
12073: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
12074: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 12075: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 12076: }else if (mobilavproj == 0){
12077: 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);
12078: 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);
12079: exit(1);
12080: }else{
12081: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 12082: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 12083: /* printf("TOTOT\n"); */
12084: /* exit(1); */
1.238 brouard 12085: }
12086: fprintf(ficresplb,"%.0f ",age );
1.338 brouard 12087: for(j=1;j<=cptcovs;j++)
12088: fprintf(ficresplb,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12089: tot=0.;
12090: for(i=1; i<=nlstate;i++){
12091: tot += bprlim[i][i];
12092: fprintf(ficresplb," %.5f", bprlim[i][i]);
12093: }
12094: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
12095: } /* Age */
12096: /* was end of cptcod */
1.255 brouard 12097: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.338 brouard 12098: /* } /\* end of any combination *\/ */
1.238 brouard 12099: } /* end of nres */
1.218 brouard 12100: /* hBijx(p, bage, fage); */
12101: /* fclose(ficrespijb); */
12102:
12103: return 0;
1.217 brouard 12104: }
1.218 brouard 12105:
1.180 brouard 12106: int hPijx(double *p, int bage, int fage){
12107: /*------------- h Pij x at various ages ------------*/
1.336 brouard 12108: /* to be optimized with precov */
1.180 brouard 12109: int stepsize;
12110: int agelim;
12111: int hstepm;
12112: int nhstepm;
1.235 brouard 12113: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 12114:
12115: double agedeb;
12116: double ***p3mat;
12117:
1.337 brouard 12118: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
12119: if((ficrespij=fopen(filerespij,"w"))==NULL) {
12120: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
12121: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
12122: }
12123: printf("Computing pij: result on file '%s' \n", filerespij);
12124: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
12125:
12126: stepsize=(int) (stepm+YEARM-1)/YEARM;
12127: /*if (stepm<=24) stepsize=2;*/
12128:
12129: agelim=AGESUP;
12130: hstepm=stepsize*YEARM; /* Every year of age */
12131: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
12132:
12133: /* hstepm=1; aff par mois*/
12134: pstamp(ficrespij);
12135: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
12136: i1= pow(2,cptcoveff);
12137: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
12138: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
12139: /* k=k+1; */
12140: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
12141: k=TKresult[nres];
1.338 brouard 12142: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 12143: /* for(k=1; k<=i1;k++){ */
12144: /* if(i1 != 1 && TKresult[nres]!= k) */
12145: /* continue; */
12146: fprintf(ficrespij,"\n#****** ");
12147: for(j=1;j<=cptcovs;j++){
12148: fprintf(ficrespij," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12149: /* fprintf(ficrespij,"@wV%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12150: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
12151: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12152: /* fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12153: }
12154: fprintf(ficrespij,"******\n");
12155:
12156: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
12157: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
12158: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
12159:
12160: /* nhstepm=nhstepm*YEARM; aff par mois*/
12161:
12162: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12163: oldm=oldms;savm=savms;
12164: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
12165: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
12166: for(i=1; i<=nlstate;i++)
12167: for(j=1; j<=nlstate+ndeath;j++)
12168: fprintf(ficrespij," %1d-%1d",i,j);
12169: fprintf(ficrespij,"\n");
12170: for (h=0; h<=nhstepm; h++){
12171: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
12172: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.183 brouard 12173: for(i=1; i<=nlstate;i++)
12174: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 12175: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.183 brouard 12176: fprintf(ficrespij,"\n");
12177: }
1.337 brouard 12178: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12179: fprintf(ficrespij,"\n");
1.180 brouard 12180: }
1.337 brouard 12181: }
12182: /*}*/
12183: return 0;
1.180 brouard 12184: }
1.218 brouard 12185:
12186: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 12187: /*------------- h Bij x at various ages ------------*/
1.336 brouard 12188: /* To be optimized with precov */
1.217 brouard 12189: int stepsize;
1.218 brouard 12190: /* int agelim; */
12191: int ageminl;
1.217 brouard 12192: int hstepm;
12193: int nhstepm;
1.238 brouard 12194: int h, i, i1, j, k, nres;
1.218 brouard 12195:
1.217 brouard 12196: double agedeb;
12197: double ***p3mat;
1.218 brouard 12198:
12199: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
12200: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
12201: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
12202: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
12203: }
12204: printf("Computing pij back: result on file '%s' \n", filerespijb);
12205: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
12206:
12207: stepsize=(int) (stepm+YEARM-1)/YEARM;
12208: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 12209:
1.218 brouard 12210: /* agelim=AGESUP; */
1.289 brouard 12211: ageminl=AGEINF; /* was 30 */
1.218 brouard 12212: hstepm=stepsize*YEARM; /* Every year of age */
12213: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
12214:
12215: /* hstepm=1; aff par mois*/
12216: pstamp(ficrespijb);
1.255 brouard 12217: 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 12218: i1= pow(2,cptcoveff);
1.218 brouard 12219: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
12220: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
12221: /* k=k+1; */
1.238 brouard 12222: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 12223: k=TKresult[nres];
1.338 brouard 12224: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 12225: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
12226: /* if(i1 != 1 && TKresult[nres]!= k) */
12227: /* continue; */
12228: fprintf(ficrespijb,"\n#****** ");
12229: for(j=1;j<=cptcovs;j++){
1.338 brouard 12230: fprintf(ficrespijb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.337 brouard 12231: /* for(j=1;j<=cptcoveff;j++) */
12232: /* fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12233: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
12234: /* fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12235: }
12236: fprintf(ficrespijb,"******\n");
12237: if(invalidvarcomb[k]){ /* Is it necessary here? */
12238: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
12239: continue;
12240: }
12241:
12242: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
12243: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
12244: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
12245: 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 */
12246: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
12247:
12248: /* nhstepm=nhstepm*YEARM; aff par mois*/
12249:
12250: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
12251: /* and memory limitations if stepm is small */
12252:
12253: /* oldm=oldms;savm=savms; */
12254: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
12255: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
12256: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
12257: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
12258: for(i=1; i<=nlstate;i++)
12259: for(j=1; j<=nlstate+ndeath;j++)
12260: fprintf(ficrespijb," %1d-%1d",i,j);
12261: fprintf(ficrespijb,"\n");
12262: for (h=0; h<=nhstepm; h++){
12263: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
12264: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
12265: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
1.217 brouard 12266: for(i=1; i<=nlstate;i++)
12267: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 12268: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.217 brouard 12269: fprintf(ficrespijb,"\n");
1.337 brouard 12270: }
12271: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12272: fprintf(ficrespijb,"\n");
12273: } /* end age deb */
12274: /* } /\* end combination *\/ */
1.238 brouard 12275: } /* end nres */
1.218 brouard 12276: return 0;
12277: } /* hBijx */
1.217 brouard 12278:
1.180 brouard 12279:
1.136 brouard 12280: /***********************************************/
12281: /**************** Main Program *****************/
12282: /***********************************************/
12283:
12284: int main(int argc, char *argv[])
12285: {
12286: #ifdef GSL
12287: const gsl_multimin_fminimizer_type *T;
12288: size_t iteri = 0, it;
12289: int rval = GSL_CONTINUE;
12290: int status = GSL_SUCCESS;
12291: double ssval;
12292: #endif
12293: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 12294: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
12295: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 12296: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 12297: int jj, ll, li, lj, lk;
1.136 brouard 12298: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 12299: int num_filled;
1.136 brouard 12300: int itimes;
12301: int NDIM=2;
12302: int vpopbased=0;
1.235 brouard 12303: int nres=0;
1.258 brouard 12304: int endishere=0;
1.277 brouard 12305: int noffset=0;
1.274 brouard 12306: int ncurrv=0; /* Temporary variable */
12307:
1.164 brouard 12308: char ca[32], cb[32];
1.136 brouard 12309: /* FILE *fichtm; *//* Html File */
12310: /* FILE *ficgp;*/ /*Gnuplot File */
12311: struct stat info;
1.191 brouard 12312: double agedeb=0.;
1.194 brouard 12313:
12314: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 12315: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 12316:
1.165 brouard 12317: double fret;
1.191 brouard 12318: double dum=0.; /* Dummy variable */
1.136 brouard 12319: double ***p3mat;
1.218 brouard 12320: /* double ***mobaverage; */
1.319 brouard 12321: double wald;
1.164 brouard 12322:
12323: char line[MAXLINE];
1.197 brouard 12324: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
12325:
1.234 brouard 12326: char modeltemp[MAXLINE];
1.332 brouard 12327: char resultline[MAXLINE], resultlineori[MAXLINE];
1.230 brouard 12328:
1.136 brouard 12329: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 12330: char *tok, *val; /* pathtot */
1.334 brouard 12331: /* int firstobs=1, lastobs=10; /\* nobs = lastobs-firstobs declared globally ;*\/ */
1.195 brouard 12332: int c, h , cpt, c2;
1.191 brouard 12333: int jl=0;
12334: int i1, j1, jk, stepsize=0;
1.194 brouard 12335: int count=0;
12336:
1.164 brouard 12337: int *tab;
1.136 brouard 12338: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 12339: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
12340: /* double anprojf, mprojf, jprojf; */
12341: /* double jintmean,mintmean,aintmean; */
12342: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
12343: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
12344: double yrfproj= 10.0; /* Number of years of forward projections */
12345: double yrbproj= 10.0; /* Number of years of backward projections */
12346: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 12347: int mobilav=0,popforecast=0;
1.191 brouard 12348: int hstepm=0, nhstepm=0;
1.136 brouard 12349: int agemortsup;
12350: float sumlpop=0.;
12351: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
12352: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
12353:
1.191 brouard 12354: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 12355: double ftolpl=FTOL;
12356: double **prlim;
1.217 brouard 12357: double **bprlim;
1.317 brouard 12358: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
12359: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 12360: double ***paramstart; /* Matrix of starting parameter values */
12361: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 12362: double **matcov; /* Matrix of covariance */
1.203 brouard 12363: double **hess; /* Hessian matrix */
1.136 brouard 12364: double ***delti3; /* Scale */
12365: double *delti; /* Scale */
12366: double ***eij, ***vareij;
12367: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 12368:
1.136 brouard 12369: double *epj, vepp;
1.164 brouard 12370:
1.273 brouard 12371: double dateprev1, dateprev2;
1.296 brouard 12372: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
12373: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
12374:
1.217 brouard 12375:
1.136 brouard 12376: double **ximort;
1.145 brouard 12377: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 12378: int *dcwave;
12379:
1.164 brouard 12380: char z[1]="c";
1.136 brouard 12381:
12382: /*char *strt;*/
12383: char strtend[80];
1.126 brouard 12384:
1.164 brouard 12385:
1.126 brouard 12386: /* setlocale (LC_ALL, ""); */
12387: /* bindtextdomain (PACKAGE, LOCALEDIR); */
12388: /* textdomain (PACKAGE); */
12389: /* setlocale (LC_CTYPE, ""); */
12390: /* setlocale (LC_MESSAGES, ""); */
12391:
12392: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 12393: rstart_time = time(NULL);
12394: /* (void) gettimeofday(&start_time,&tzp);*/
12395: start_time = *localtime(&rstart_time);
1.126 brouard 12396: curr_time=start_time;
1.157 brouard 12397: /*tml = *localtime(&start_time.tm_sec);*/
12398: /* strcpy(strstart,asctime(&tml)); */
12399: strcpy(strstart,asctime(&start_time));
1.126 brouard 12400:
12401: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 12402: /* tp.tm_sec = tp.tm_sec +86400; */
12403: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 12404: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
12405: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
12406: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 12407: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 12408: /* strt=asctime(&tmg); */
12409: /* printf("Time(after) =%s",strstart); */
12410: /* (void) time (&time_value);
12411: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
12412: * tm = *localtime(&time_value);
12413: * strstart=asctime(&tm);
12414: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
12415: */
12416:
12417: nberr=0; /* Number of errors and warnings */
12418: nbwarn=0;
1.184 brouard 12419: #ifdef WIN32
12420: _getcwd(pathcd, size);
12421: #else
1.126 brouard 12422: getcwd(pathcd, size);
1.184 brouard 12423: #endif
1.191 brouard 12424: syscompilerinfo(0);
1.196 brouard 12425: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 12426: if(argc <=1){
12427: printf("\nEnter the parameter file name: ");
1.205 brouard 12428: if(!fgets(pathr,FILENAMELENGTH,stdin)){
12429: printf("ERROR Empty parameter file name\n");
12430: goto end;
12431: }
1.126 brouard 12432: i=strlen(pathr);
12433: if(pathr[i-1]=='\n')
12434: pathr[i-1]='\0';
1.156 brouard 12435: i=strlen(pathr);
1.205 brouard 12436: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 12437: pathr[i-1]='\0';
1.205 brouard 12438: }
12439: i=strlen(pathr);
12440: if( i==0 ){
12441: printf("ERROR Empty parameter file name\n");
12442: goto end;
12443: }
12444: for (tok = pathr; tok != NULL; ){
1.126 brouard 12445: printf("Pathr |%s|\n",pathr);
12446: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
12447: printf("val= |%s| pathr=%s\n",val,pathr);
12448: strcpy (pathtot, val);
12449: if(pathr[0] == '\0') break; /* Dirty */
12450: }
12451: }
1.281 brouard 12452: else if (argc<=2){
12453: strcpy(pathtot,argv[1]);
12454: }
1.126 brouard 12455: else{
12456: strcpy(pathtot,argv[1]);
1.281 brouard 12457: strcpy(z,argv[2]);
12458: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 12459: }
12460: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
12461: /*cygwin_split_path(pathtot,path,optionfile);
12462: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
12463: /* cutv(path,optionfile,pathtot,'\\');*/
12464:
12465: /* Split argv[0], imach program to get pathimach */
12466: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
12467: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12468: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12469: /* strcpy(pathimach,argv[0]); */
12470: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
12471: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
12472: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 12473: #ifdef WIN32
12474: _chdir(path); /* Can be a relative path */
12475: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
12476: #else
1.126 brouard 12477: chdir(path); /* Can be a relative path */
1.184 brouard 12478: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
12479: #endif
12480: printf("Current directory %s!\n",pathcd);
1.126 brouard 12481: strcpy(command,"mkdir ");
12482: strcat(command,optionfilefiname);
12483: if((outcmd=system(command)) != 0){
1.169 brouard 12484: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 12485: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
12486: /* fclose(ficlog); */
12487: /* exit(1); */
12488: }
12489: /* if((imk=mkdir(optionfilefiname))<0){ */
12490: /* perror("mkdir"); */
12491: /* } */
12492:
12493: /*-------- arguments in the command line --------*/
12494:
1.186 brouard 12495: /* Main Log file */
1.126 brouard 12496: strcat(filelog, optionfilefiname);
12497: strcat(filelog,".log"); /* */
12498: if((ficlog=fopen(filelog,"w"))==NULL) {
12499: printf("Problem with logfile %s\n",filelog);
12500: goto end;
12501: }
12502: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 12503: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 12504: fprintf(ficlog,"\nEnter the parameter file name: \n");
12505: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
12506: path=%s \n\
12507: optionfile=%s\n\
12508: optionfilext=%s\n\
1.156 brouard 12509: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 12510:
1.197 brouard 12511: syscompilerinfo(1);
1.167 brouard 12512:
1.126 brouard 12513: printf("Local time (at start):%s",strstart);
12514: fprintf(ficlog,"Local time (at start): %s",strstart);
12515: fflush(ficlog);
12516: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 12517: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 12518:
12519: /* */
12520: strcpy(fileres,"r");
12521: strcat(fileres, optionfilefiname);
1.201 brouard 12522: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 12523: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 12524: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 12525:
1.186 brouard 12526: /* Main ---------arguments file --------*/
1.126 brouard 12527:
12528: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 12529: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
12530: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 12531: fflush(ficlog);
1.149 brouard 12532: /* goto end; */
12533: exit(70);
1.126 brouard 12534: }
12535:
12536: strcpy(filereso,"o");
1.201 brouard 12537: strcat(filereso,fileresu);
1.126 brouard 12538: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
12539: printf("Problem with Output resultfile: %s\n", filereso);
12540: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
12541: fflush(ficlog);
12542: goto end;
12543: }
1.278 brouard 12544: /*-------- Rewriting parameter file ----------*/
12545: strcpy(rfileres,"r"); /* "Rparameterfile */
12546: strcat(rfileres,optionfilefiname); /* Parameter file first name */
12547: strcat(rfileres,"."); /* */
12548: strcat(rfileres,optionfilext); /* Other files have txt extension */
12549: if((ficres =fopen(rfileres,"w"))==NULL) {
12550: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
12551: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
12552: fflush(ficlog);
12553: goto end;
12554: }
12555: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 12556:
1.278 brouard 12557:
1.126 brouard 12558: /* Reads comments: lines beginning with '#' */
12559: numlinepar=0;
1.277 brouard 12560: /* Is it a BOM UTF-8 Windows file? */
12561: /* First parameter line */
1.197 brouard 12562: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 12563: noffset=0;
12564: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
12565: {
12566: noffset=noffset+3;
12567: printf("# File is an UTF8 Bom.\n"); // 0xBF
12568: }
1.302 brouard 12569: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
12570: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 12571: {
12572: noffset=noffset+2;
12573: printf("# File is an UTF16BE BOM file\n");
12574: }
12575: else if( line[0] == 0 && line[1] == 0)
12576: {
12577: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
12578: noffset=noffset+4;
12579: printf("# File is an UTF16BE BOM file\n");
12580: }
12581: } else{
12582: ;/*printf(" Not a BOM file\n");*/
12583: }
12584:
1.197 brouard 12585: /* If line starts with a # it is a comment */
1.277 brouard 12586: if (line[noffset] == '#') {
1.197 brouard 12587: numlinepar++;
12588: fputs(line,stdout);
12589: fputs(line,ficparo);
1.278 brouard 12590: fputs(line,ficres);
1.197 brouard 12591: fputs(line,ficlog);
12592: continue;
12593: }else
12594: break;
12595: }
12596: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
12597: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
12598: if (num_filled != 5) {
12599: printf("Should be 5 parameters\n");
1.283 brouard 12600: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 12601: }
1.126 brouard 12602: numlinepar++;
1.197 brouard 12603: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 12604: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12605: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12606: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 12607: }
12608: /* Second parameter line */
12609: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 12610: /* while(fscanf(ficpar,"%[^\n]", line)) { */
12611: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 12612: if (line[0] == '#') {
12613: numlinepar++;
1.283 brouard 12614: printf("%s",line);
12615: fprintf(ficres,"%s",line);
12616: fprintf(ficparo,"%s",line);
12617: fprintf(ficlog,"%s",line);
1.197 brouard 12618: continue;
12619: }else
12620: break;
12621: }
1.223 brouard 12622: 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", \
12623: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
12624: if (num_filled != 11) {
12625: 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 12626: printf("but line=%s\n",line);
1.283 brouard 12627: 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");
12628: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 12629: }
1.286 brouard 12630: if( lastpass > maxwav){
12631: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12632: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12633: fflush(ficlog);
12634: goto end;
12635: }
12636: 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 12637: 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 12638: 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 12639: 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 12640: }
1.203 brouard 12641: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 12642: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 12643: /* Third parameter line */
12644: while(fgets(line, MAXLINE, ficpar)) {
12645: /* If line starts with a # it is a comment */
12646: if (line[0] == '#') {
12647: numlinepar++;
1.283 brouard 12648: printf("%s",line);
12649: fprintf(ficres,"%s",line);
12650: fprintf(ficparo,"%s",line);
12651: fprintf(ficlog,"%s",line);
1.197 brouard 12652: continue;
12653: }else
12654: break;
12655: }
1.201 brouard 12656: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 12657: if (num_filled != 1){
1.302 brouard 12658: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
12659: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 12660: model[0]='\0';
12661: goto end;
12662: }
12663: else{
12664: if (model[0]=='+'){
12665: for(i=1; i<=strlen(model);i++)
12666: modeltemp[i-1]=model[i];
1.201 brouard 12667: strcpy(model,modeltemp);
1.197 brouard 12668: }
12669: }
1.338 brouard 12670: /* printf(" model=1+age%s modeltemp= %s, model=1+age+%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 12671: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 12672: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
12673: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
12674: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 12675: }
12676: /* 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); */
12677: /* numlinepar=numlinepar+3; /\* In general *\/ */
12678: /* 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 12679: /* 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); */
12680: /* 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 12681: fflush(ficlog);
1.190 brouard 12682: /* if(model[0]=='#'|| model[0]== '\0'){ */
12683: if(model[0]=='#'){
1.279 brouard 12684: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
12685: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
12686: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 12687: if(mle != -1){
1.279 brouard 12688: 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 12689: exit(1);
12690: }
12691: }
1.126 brouard 12692: while((c=getc(ficpar))=='#' && c!= EOF){
12693: ungetc(c,ficpar);
12694: fgets(line, MAXLINE, ficpar);
12695: numlinepar++;
1.195 brouard 12696: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
12697: z[0]=line[1];
1.342 brouard 12698: }else if(line[1]=='d'){ /* For debugging individual values of covariates in ficresilk */
1.343 ! brouard 12699: debugILK=1;printf("DebugILK\n");
1.195 brouard 12700: }
12701: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 12702: fputs(line, stdout);
12703: //puts(line);
1.126 brouard 12704: fputs(line,ficparo);
12705: fputs(line,ficlog);
12706: }
12707: ungetc(c,ficpar);
12708:
12709:
1.290 brouard 12710: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
12711: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
12712: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
1.341 brouard 12713: /* if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /\**< Time varying covariate (dummy and quantitative)*\/ */
12714: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs); /**< Might be better */
1.136 brouard 12715: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
12716: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
12717: v1+v2*age+v2*v3 makes cptcovn = 3
12718: */
12719: if (strlen(model)>1)
1.187 brouard 12720: 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 12721: else
1.187 brouard 12722: ncovmodel=2; /* Constant and age */
1.133 brouard 12723: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
12724: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 12725: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
12726: 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);
12727: 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);
12728: fflush(stdout);
12729: fclose (ficlog);
12730: goto end;
12731: }
1.126 brouard 12732: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12733: delti=delti3[1][1];
12734: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
12735: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 12736: /* We could also provide initial parameters values giving by simple logistic regression
12737: * only one way, that is without matrix product. We will have nlstate maximizations */
12738: /* for(i=1;i<nlstate;i++){ */
12739: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12740: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12741: /* } */
1.126 brouard 12742: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 12743: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
12744: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12745: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12746: fclose (ficparo);
12747: fclose (ficlog);
12748: goto end;
12749: exit(0);
1.220 brouard 12750: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 12751: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 12752: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
12753: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12754: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12755: matcov=matrix(1,npar,1,npar);
1.203 brouard 12756: hess=matrix(1,npar,1,npar);
1.220 brouard 12757: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 12758: /* Read guessed parameters */
1.126 brouard 12759: /* Reads comments: lines beginning with '#' */
12760: while((c=getc(ficpar))=='#' && c!= EOF){
12761: ungetc(c,ficpar);
12762: fgets(line, MAXLINE, ficpar);
12763: numlinepar++;
1.141 brouard 12764: fputs(line,stdout);
1.126 brouard 12765: fputs(line,ficparo);
12766: fputs(line,ficlog);
12767: }
12768: ungetc(c,ficpar);
12769:
12770: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 12771: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 12772: for(i=1; i <=nlstate; i++){
1.234 brouard 12773: j=0;
1.126 brouard 12774: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 12775: if(jj==i) continue;
12776: j++;
1.292 brouard 12777: while((c=getc(ficpar))=='#' && c!= EOF){
12778: ungetc(c,ficpar);
12779: fgets(line, MAXLINE, ficpar);
12780: numlinepar++;
12781: fputs(line,stdout);
12782: fputs(line,ficparo);
12783: fputs(line,ficlog);
12784: }
12785: ungetc(c,ficpar);
1.234 brouard 12786: fscanf(ficpar,"%1d%1d",&i1,&j1);
12787: if ((i1 != i) || (j1 != jj)){
12788: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 12789: It might be a problem of design; if ncovcol and the model are correct\n \
12790: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 12791: exit(1);
12792: }
12793: fprintf(ficparo,"%1d%1d",i1,j1);
12794: if(mle==1)
12795: printf("%1d%1d",i,jj);
12796: fprintf(ficlog,"%1d%1d",i,jj);
12797: for(k=1; k<=ncovmodel;k++){
12798: fscanf(ficpar," %lf",¶m[i][j][k]);
12799: if(mle==1){
12800: printf(" %lf",param[i][j][k]);
12801: fprintf(ficlog," %lf",param[i][j][k]);
12802: }
12803: else
12804: fprintf(ficlog," %lf",param[i][j][k]);
12805: fprintf(ficparo," %lf",param[i][j][k]);
12806: }
12807: fscanf(ficpar,"\n");
12808: numlinepar++;
12809: if(mle==1)
12810: printf("\n");
12811: fprintf(ficlog,"\n");
12812: fprintf(ficparo,"\n");
1.126 brouard 12813: }
12814: }
12815: fflush(ficlog);
1.234 brouard 12816:
1.251 brouard 12817: /* Reads parameters values */
1.126 brouard 12818: p=param[1][1];
1.251 brouard 12819: pstart=paramstart[1][1];
1.126 brouard 12820:
12821: /* Reads comments: lines beginning with '#' */
12822: while((c=getc(ficpar))=='#' && c!= EOF){
12823: ungetc(c,ficpar);
12824: fgets(line, MAXLINE, ficpar);
12825: numlinepar++;
1.141 brouard 12826: fputs(line,stdout);
1.126 brouard 12827: fputs(line,ficparo);
12828: fputs(line,ficlog);
12829: }
12830: ungetc(c,ficpar);
12831:
12832: for(i=1; i <=nlstate; i++){
12833: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 12834: fscanf(ficpar,"%1d%1d",&i1,&j1);
12835: if ( (i1-i) * (j1-j) != 0){
12836: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
12837: exit(1);
12838: }
12839: printf("%1d%1d",i,j);
12840: fprintf(ficparo,"%1d%1d",i1,j1);
12841: fprintf(ficlog,"%1d%1d",i1,j1);
12842: for(k=1; k<=ncovmodel;k++){
12843: fscanf(ficpar,"%le",&delti3[i][j][k]);
12844: printf(" %le",delti3[i][j][k]);
12845: fprintf(ficparo," %le",delti3[i][j][k]);
12846: fprintf(ficlog," %le",delti3[i][j][k]);
12847: }
12848: fscanf(ficpar,"\n");
12849: numlinepar++;
12850: printf("\n");
12851: fprintf(ficparo,"\n");
12852: fprintf(ficlog,"\n");
1.126 brouard 12853: }
12854: }
12855: fflush(ficlog);
1.234 brouard 12856:
1.145 brouard 12857: /* Reads covariance matrix */
1.126 brouard 12858: delti=delti3[1][1];
1.220 brouard 12859:
12860:
1.126 brouard 12861: /* 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 12862:
1.126 brouard 12863: /* Reads comments: lines beginning with '#' */
12864: while((c=getc(ficpar))=='#' && c!= EOF){
12865: ungetc(c,ficpar);
12866: fgets(line, MAXLINE, ficpar);
12867: numlinepar++;
1.141 brouard 12868: fputs(line,stdout);
1.126 brouard 12869: fputs(line,ficparo);
12870: fputs(line,ficlog);
12871: }
12872: ungetc(c,ficpar);
1.220 brouard 12873:
1.126 brouard 12874: matcov=matrix(1,npar,1,npar);
1.203 brouard 12875: hess=matrix(1,npar,1,npar);
1.131 brouard 12876: for(i=1; i <=npar; i++)
12877: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 12878:
1.194 brouard 12879: /* Scans npar lines */
1.126 brouard 12880: for(i=1; i <=npar; i++){
1.226 brouard 12881: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 12882: if(count != 3){
1.226 brouard 12883: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12884: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12885: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12886: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12887: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12888: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12889: exit(1);
1.220 brouard 12890: }else{
1.226 brouard 12891: if(mle==1)
12892: printf("%1d%1d%d",i1,j1,jk);
12893: }
12894: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
12895: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 12896: for(j=1; j <=i; j++){
1.226 brouard 12897: fscanf(ficpar," %le",&matcov[i][j]);
12898: if(mle==1){
12899: printf(" %.5le",matcov[i][j]);
12900: }
12901: fprintf(ficlog," %.5le",matcov[i][j]);
12902: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 12903: }
12904: fscanf(ficpar,"\n");
12905: numlinepar++;
12906: if(mle==1)
1.220 brouard 12907: printf("\n");
1.126 brouard 12908: fprintf(ficlog,"\n");
12909: fprintf(ficparo,"\n");
12910: }
1.194 brouard 12911: /* End of read covariance matrix npar lines */
1.126 brouard 12912: for(i=1; i <=npar; i++)
12913: for(j=i+1;j<=npar;j++)
1.226 brouard 12914: matcov[i][j]=matcov[j][i];
1.126 brouard 12915:
12916: if(mle==1)
12917: printf("\n");
12918: fprintf(ficlog,"\n");
12919:
12920: fflush(ficlog);
12921:
12922: } /* End of mle != -3 */
1.218 brouard 12923:
1.186 brouard 12924: /* Main data
12925: */
1.290 brouard 12926: nobs=lastobs-firstobs+1; /* was = lastobs;*/
12927: /* num=lvector(1,n); */
12928: /* moisnais=vector(1,n); */
12929: /* annais=vector(1,n); */
12930: /* moisdc=vector(1,n); */
12931: /* andc=vector(1,n); */
12932: /* weight=vector(1,n); */
12933: /* agedc=vector(1,n); */
12934: /* cod=ivector(1,n); */
12935: /* for(i=1;i<=n;i++){ */
12936: num=lvector(firstobs,lastobs);
12937: moisnais=vector(firstobs,lastobs);
12938: annais=vector(firstobs,lastobs);
12939: moisdc=vector(firstobs,lastobs);
12940: andc=vector(firstobs,lastobs);
12941: weight=vector(firstobs,lastobs);
12942: agedc=vector(firstobs,lastobs);
12943: cod=ivector(firstobs,lastobs);
12944: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 12945: num[i]=0;
12946: moisnais[i]=0;
12947: annais[i]=0;
12948: moisdc[i]=0;
12949: andc[i]=0;
12950: agedc[i]=0;
12951: cod[i]=0;
12952: weight[i]=1.0; /* Equal weights, 1 by default */
12953: }
1.290 brouard 12954: mint=matrix(1,maxwav,firstobs,lastobs);
12955: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 12956: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.336 brouard 12957: /* printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel)); */
1.126 brouard 12958: tab=ivector(1,NCOVMAX);
1.144 brouard 12959: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 12960: 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 12961:
1.136 brouard 12962: /* Reads data from file datafile */
12963: if (readdata(datafile, firstobs, lastobs, &imx)==1)
12964: goto end;
12965:
12966: /* Calculation of the number of parameters from char model */
1.234 brouard 12967: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 12968: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
12969: k=3 V4 Tvar[k=3]= 4 (from V4)
12970: k=2 V1 Tvar[k=2]= 1 (from V1)
12971: k=1 Tvar[1]=2 (from V2)
1.234 brouard 12972: */
12973:
12974: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
12975: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 brouard 12976: TnsdVar=ivector(1,NCOVMAX); /* */
1.335 brouard 12977: /* for(i=1; i<=NCOVMAX;i++) TnsdVar[i]=3; */
1.234 brouard 12978: TvarsD=ivector(1,NCOVMAX); /* */
12979: TvarsQind=ivector(1,NCOVMAX); /* */
12980: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 12981: TvarF=ivector(1,NCOVMAX); /* */
12982: TvarFind=ivector(1,NCOVMAX); /* */
12983: TvarV=ivector(1,NCOVMAX); /* */
12984: TvarVind=ivector(1,NCOVMAX); /* */
12985: TvarA=ivector(1,NCOVMAX); /* */
12986: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 12987: TvarFD=ivector(1,NCOVMAX); /* */
12988: TvarFDind=ivector(1,NCOVMAX); /* */
12989: TvarFQ=ivector(1,NCOVMAX); /* */
12990: TvarFQind=ivector(1,NCOVMAX); /* */
12991: TvarVD=ivector(1,NCOVMAX); /* */
12992: TvarVDind=ivector(1,NCOVMAX); /* */
12993: TvarVQ=ivector(1,NCOVMAX); /* */
12994: TvarVQind=ivector(1,NCOVMAX); /* */
1.339 brouard 12995: TvarVV=ivector(1,NCOVMAX); /* */
12996: TvarVVind=ivector(1,NCOVMAX); /* */
1.231 brouard 12997:
1.230 brouard 12998: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 12999: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 13000: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
13001: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
13002: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 13003: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
13004: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
13005: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
13006: */
13007: /* For model-covariate k tells which data-covariate to use but
13008: because this model-covariate is a construction we invent a new column
13009: ncovcol + k1
13010: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
13011: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 13012: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
13013: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 13014: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
13015: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 13016: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 13017: */
1.145 brouard 13018: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
13019: 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 13020: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
13021: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.330 brouard 13022: Tvardk=imatrix(1,NCOVMAX,1,2);
1.145 brouard 13023: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 13024: 4 covariates (3 plus signs)
13025: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 13026: */
13027: for(i=1;i<NCOVMAX;i++)
13028: Tage[i]=0;
1.230 brouard 13029: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 13030: * individual dummy, fixed or varying:
13031: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
13032: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 13033: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
13034: * V1 df, V2 qf, V3 & V4 dv, V5 qv
13035: * Tmodelind[1]@9={9,0,3,2,}*/
13036: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
13037: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 13038: * individual quantitative, fixed or varying:
13039: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
13040: * 3, 1, 0, 0, 0, 0, 0, 0},
13041: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 13042: /* Main decodemodel */
13043:
1.187 brouard 13044:
1.223 brouard 13045: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 13046: goto end;
13047:
1.137 brouard 13048: if((double)(lastobs-imx)/(double)imx > 1.10){
13049: nbwarn++;
13050: 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);
13051: 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);
13052: }
1.136 brouard 13053: /* if(mle==1){*/
1.137 brouard 13054: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
13055: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 13056: }
13057:
13058: /*-calculation of age at interview from date of interview and age at death -*/
13059: agev=matrix(1,maxwav,1,imx);
13060:
13061: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
13062: goto end;
13063:
1.126 brouard 13064:
1.136 brouard 13065: agegomp=(int)agemin;
1.290 brouard 13066: free_vector(moisnais,firstobs,lastobs);
13067: free_vector(annais,firstobs,lastobs);
1.126 brouard 13068: /* free_matrix(mint,1,maxwav,1,n);
13069: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 13070: /* free_vector(moisdc,1,n); */
13071: /* free_vector(andc,1,n); */
1.145 brouard 13072: /* */
13073:
1.126 brouard 13074: wav=ivector(1,imx);
1.214 brouard 13075: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
13076: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
13077: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
13078: 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.*/
13079: bh=imatrix(1,lastpass-firstpass+2,1,imx);
13080: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 13081:
13082: /* Concatenates waves */
1.214 brouard 13083: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
13084: Death is a valid wave (if date is known).
13085: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
13086: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
13087: and mw[mi+1][i]. dh depends on stepm.
13088: */
13089:
1.126 brouard 13090: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 13091: /* Concatenates waves */
1.145 brouard 13092:
1.290 brouard 13093: free_vector(moisdc,firstobs,lastobs);
13094: free_vector(andc,firstobs,lastobs);
1.215 brouard 13095:
1.126 brouard 13096: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
13097: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
13098: ncodemax[1]=1;
1.145 brouard 13099: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 13100: cptcoveff=0;
1.220 brouard 13101: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
1.335 brouard 13102: 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 13103: }
13104:
13105: ncovcombmax=pow(2,cptcoveff);
1.338 brouard 13106: invalidvarcomb=ivector(0, ncovcombmax);
13107: for(i=0;i<ncovcombmax;i++)
1.227 brouard 13108: invalidvarcomb[i]=0;
13109:
1.211 brouard 13110: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 13111: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 13112: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 13113:
1.200 brouard 13114: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 13115: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 13116: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 13117: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
13118: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
13119: * (currently 0 or 1) in the data.
13120: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
13121: * corresponding modality (h,j).
13122: */
13123:
1.145 brouard 13124: h=0;
13125: /*if (cptcovn > 0) */
1.126 brouard 13126: m=pow(2,cptcoveff);
13127:
1.144 brouard 13128: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 13129: * For k=4 covariates, h goes from 1 to m=2**k
13130: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
13131: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 13132: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
13133: *______________________________ *______________________
13134: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
13135: * 2 2 1 1 1 * 1 0 0 0 1
13136: * 3 i=2 1 2 1 1 * 2 0 0 1 0
13137: * 4 2 2 1 1 * 3 0 0 1 1
13138: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
13139: * 6 2 1 2 1 * 5 0 1 0 1
13140: * 7 i=4 1 2 2 1 * 6 0 1 1 0
13141: * 8 2 2 2 1 * 7 0 1 1 1
13142: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
13143: * 10 2 1 1 2 * 9 1 0 0 1
13144: * 11 i=6 1 2 1 2 * 10 1 0 1 0
13145: * 12 2 2 1 2 * 11 1 0 1 1
13146: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
13147: * 14 2 1 2 2 * 13 1 1 0 1
13148: * 15 i=8 1 2 2 2 * 14 1 1 1 0
13149: * 16 2 2 2 2 * 15 1 1 1 1
13150: */
1.212 brouard 13151: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 13152: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
13153: * and the value of each covariate?
13154: * V1=1, V2=1, V3=2, V4=1 ?
13155: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
13156: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
13157: * In order to get the real value in the data, we use nbcode
13158: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
13159: * We are keeping this crazy system in order to be able (in the future?)
13160: * to have more than 2 values (0 or 1) for a covariate.
13161: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
13162: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
13163: * bbbbbbbb
13164: * 76543210
13165: * h-1 00000101 (6-1=5)
1.219 brouard 13166: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 13167: * &
13168: * 1 00000001 (1)
1.219 brouard 13169: * 00000000 = 1 & ((h-1) >> (k-1))
13170: * +1= 00000001 =1
1.211 brouard 13171: *
13172: * h=14, k=3 => h'=h-1=13, k'=k-1=2
13173: * h' 1101 =2^3+2^2+0x2^1+2^0
13174: * >>k' 11
13175: * & 00000001
13176: * = 00000001
13177: * +1 = 00000010=2 = codtabm(14,3)
13178: * Reverse h=6 and m=16?
13179: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
13180: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
13181: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
13182: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
13183: * V3=decodtabm(14,3,2**4)=2
13184: * h'=13 1101 =2^3+2^2+0x2^1+2^0
13185: *(h-1) >> (j-1) 0011 =13 >> 2
13186: * &1 000000001
13187: * = 000000001
13188: * +1= 000000010 =2
13189: * 2211
13190: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
13191: * V3=2
1.220 brouard 13192: * codtabm and decodtabm are identical
1.211 brouard 13193: */
13194:
1.145 brouard 13195:
13196: free_ivector(Ndum,-1,NCOVMAX);
13197:
13198:
1.126 brouard 13199:
1.186 brouard 13200: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 13201: strcpy(optionfilegnuplot,optionfilefiname);
13202: if(mle==-3)
1.201 brouard 13203: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 13204: strcat(optionfilegnuplot,".gp");
13205:
13206: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
13207: printf("Problem with file %s",optionfilegnuplot);
13208: }
13209: else{
1.204 brouard 13210: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 13211: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 13212: //fprintf(ficgp,"set missing 'NaNq'\n");
13213: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 13214: }
13215: /* fclose(ficgp);*/
1.186 brouard 13216:
13217:
13218: /* Initialisation of --------- index.htm --------*/
1.126 brouard 13219:
13220: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
13221: if(mle==-3)
1.201 brouard 13222: strcat(optionfilehtm,"-MORT_");
1.126 brouard 13223: strcat(optionfilehtm,".htm");
13224: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 13225: printf("Problem with %s \n",optionfilehtm);
13226: exit(0);
1.126 brouard 13227: }
13228:
13229: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
13230: strcat(optionfilehtmcov,"-cov.htm");
13231: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
13232: printf("Problem with %s \n",optionfilehtmcov), exit(0);
13233: }
13234: else{
13235: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
13236: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 13237: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 13238: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
13239: }
13240:
1.335 brouard 13241: fprintf(fichtm,"<html><head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n\
13242: <title>IMaCh %s</title></head>\n\
13243: <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n\
13244: <font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>\
13245: -EUROREVES-Institut de longévité-2013-2022-Japan Society for the Promotion of Sciences 日本学術振興会 \
13246: (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - \
13247: <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \n", optionfilehtm);
13248:
13249: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 13250: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 13251: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.337 brouard 13252: 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 13253: \n\
13254: <hr size=\"2\" color=\"#EC5E5E\">\
13255: <ul><li><h4>Parameter files</h4>\n\
13256: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
13257: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
13258: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
13259: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
13260: - Date and time at start: %s</ul>\n",\
1.335 brouard 13261: version,fullversion,optionfilehtm,optionfilehtm,title,datafile,datafile,firstpass,lastpass,stepm, weightopt, model, \
1.126 brouard 13262: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
13263: fileres,fileres,\
13264: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
13265: fflush(fichtm);
13266:
13267: strcpy(pathr,path);
13268: strcat(pathr,optionfilefiname);
1.184 brouard 13269: #ifdef WIN32
13270: _chdir(optionfilefiname); /* Move to directory named optionfile */
13271: #else
1.126 brouard 13272: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 13273: #endif
13274:
1.126 brouard 13275:
1.220 brouard 13276: /* Calculates basic frequencies. Computes observed prevalence at single age
13277: and for any valid combination of covariates
1.126 brouard 13278: and prints on file fileres'p'. */
1.251 brouard 13279: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 13280: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 13281:
13282: fprintf(fichtm,"\n");
1.286 brouard 13283: 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 13284: ftol, stepm);
13285: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
13286: ncurrv=1;
13287: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
13288: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
13289: ncurrv=i;
13290: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 13291: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 13292: ncurrv=i;
13293: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 13294: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 13295: ncurrv=i;
13296: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
13297: 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", \
13298: nlstate, ndeath, maxwav, mle, weightopt);
13299:
13300: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
13301: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
13302:
13303:
1.317 brouard 13304: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 13305: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
13306: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 13307: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 13308: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 13309: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13310: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13311: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13312: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 13313:
1.126 brouard 13314: /* For Powell, parameters are in a vector p[] starting at p[1]
13315: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
13316: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
13317:
13318: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 13319: /* For mortality only */
1.126 brouard 13320: if (mle==-3){
1.136 brouard 13321: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 13322: for(i=1;i<=NDIM;i++)
13323: for(j=1;j<=NDIM;j++)
13324: ximort[i][j]=0.;
1.186 brouard 13325: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 13326: cens=ivector(firstobs,lastobs);
13327: ageexmed=vector(firstobs,lastobs);
13328: agecens=vector(firstobs,lastobs);
13329: dcwave=ivector(firstobs,lastobs);
1.223 brouard 13330:
1.126 brouard 13331: for (i=1; i<=imx; i++){
13332: dcwave[i]=-1;
13333: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 13334: if (s[m][i]>nlstate) {
13335: dcwave[i]=m;
13336: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
13337: break;
13338: }
1.126 brouard 13339: }
1.226 brouard 13340:
1.126 brouard 13341: for (i=1; i<=imx; i++) {
13342: if (wav[i]>0){
1.226 brouard 13343: ageexmed[i]=agev[mw[1][i]][i];
13344: j=wav[i];
13345: agecens[i]=1.;
13346:
13347: if (ageexmed[i]> 1 && wav[i] > 0){
13348: agecens[i]=agev[mw[j][i]][i];
13349: cens[i]= 1;
13350: }else if (ageexmed[i]< 1)
13351: cens[i]= -1;
13352: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
13353: cens[i]=0 ;
1.126 brouard 13354: }
13355: else cens[i]=-1;
13356: }
13357:
13358: for (i=1;i<=NDIM;i++) {
13359: for (j=1;j<=NDIM;j++)
1.226 brouard 13360: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 13361: }
13362:
1.302 brouard 13363: p[1]=0.0268; p[NDIM]=0.083;
13364: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 13365:
13366:
1.136 brouard 13367: #ifdef GSL
13368: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 13369: #else
1.126 brouard 13370: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 13371: #endif
1.201 brouard 13372: strcpy(filerespow,"POW-MORT_");
13373: strcat(filerespow,fileresu);
1.126 brouard 13374: if((ficrespow=fopen(filerespow,"w"))==NULL) {
13375: printf("Problem with resultfile: %s\n", filerespow);
13376: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
13377: }
1.136 brouard 13378: #ifdef GSL
13379: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 13380: #else
1.126 brouard 13381: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 13382: #endif
1.126 brouard 13383: /* for (i=1;i<=nlstate;i++)
13384: for(j=1;j<=nlstate+ndeath;j++)
13385: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
13386: */
13387: fprintf(ficrespow,"\n");
1.136 brouard 13388: #ifdef GSL
13389: /* gsl starts here */
13390: T = gsl_multimin_fminimizer_nmsimplex;
13391: gsl_multimin_fminimizer *sfm = NULL;
13392: gsl_vector *ss, *x;
13393: gsl_multimin_function minex_func;
13394:
13395: /* Initial vertex size vector */
13396: ss = gsl_vector_alloc (NDIM);
13397:
13398: if (ss == NULL){
13399: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
13400: }
13401: /* Set all step sizes to 1 */
13402: gsl_vector_set_all (ss, 0.001);
13403:
13404: /* Starting point */
1.126 brouard 13405:
1.136 brouard 13406: x = gsl_vector_alloc (NDIM);
13407:
13408: if (x == NULL){
13409: gsl_vector_free(ss);
13410: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
13411: }
13412:
13413: /* Initialize method and iterate */
13414: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 13415: /* gsl_vector_set(x, 0, 0.0268); */
13416: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 13417: gsl_vector_set(x, 0, p[1]);
13418: gsl_vector_set(x, 1, p[2]);
13419:
13420: minex_func.f = &gompertz_f;
13421: minex_func.n = NDIM;
13422: minex_func.params = (void *)&p; /* ??? */
13423:
13424: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
13425: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
13426:
13427: printf("Iterations beginning .....\n\n");
13428: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
13429:
13430: iteri=0;
13431: while (rval == GSL_CONTINUE){
13432: iteri++;
13433: status = gsl_multimin_fminimizer_iterate(sfm);
13434:
13435: if (status) printf("error: %s\n", gsl_strerror (status));
13436: fflush(0);
13437:
13438: if (status)
13439: break;
13440:
13441: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
13442: ssval = gsl_multimin_fminimizer_size (sfm);
13443:
13444: if (rval == GSL_SUCCESS)
13445: printf ("converged to a local maximum at\n");
13446:
13447: printf("%5d ", iteri);
13448: for (it = 0; it < NDIM; it++){
13449: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
13450: }
13451: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
13452: }
13453:
13454: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
13455:
13456: gsl_vector_free(x); /* initial values */
13457: gsl_vector_free(ss); /* inital step size */
13458: for (it=0; it<NDIM; it++){
13459: p[it+1]=gsl_vector_get(sfm->x,it);
13460: fprintf(ficrespow," %.12lf", p[it]);
13461: }
13462: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
13463: #endif
13464: #ifdef POWELL
13465: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
13466: #endif
1.126 brouard 13467: fclose(ficrespow);
13468:
1.203 brouard 13469: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 13470:
13471: for(i=1; i <=NDIM; i++)
13472: for(j=i+1;j<=NDIM;j++)
1.220 brouard 13473: matcov[i][j]=matcov[j][i];
1.126 brouard 13474:
13475: printf("\nCovariance matrix\n ");
1.203 brouard 13476: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 13477: for(i=1; i <=NDIM; i++) {
13478: for(j=1;j<=NDIM;j++){
1.220 brouard 13479: printf("%f ",matcov[i][j]);
13480: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 13481: }
1.203 brouard 13482: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 13483: }
13484:
13485: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 13486: for (i=1;i<=NDIM;i++) {
1.126 brouard 13487: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 13488: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
13489: }
1.302 brouard 13490: lsurv=vector(agegomp,AGESUP);
13491: lpop=vector(agegomp,AGESUP);
13492: tpop=vector(agegomp,AGESUP);
1.126 brouard 13493: lsurv[agegomp]=100000;
13494:
13495: for (k=agegomp;k<=AGESUP;k++) {
13496: agemortsup=k;
13497: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
13498: }
13499:
13500: for (k=agegomp;k<agemortsup;k++)
13501: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
13502:
13503: for (k=agegomp;k<agemortsup;k++){
13504: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
13505: sumlpop=sumlpop+lpop[k];
13506: }
13507:
13508: tpop[agegomp]=sumlpop;
13509: for (k=agegomp;k<(agemortsup-3);k++){
13510: /* tpop[k+1]=2;*/
13511: tpop[k+1]=tpop[k]-lpop[k];
13512: }
13513:
13514:
13515: printf("\nAge lx qx dx Lx Tx e(x)\n");
13516: for (k=agegomp;k<(agemortsup-2);k++)
13517: 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]);
13518:
13519:
13520: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 13521: ageminpar=50;
13522: agemaxpar=100;
1.194 brouard 13523: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
13524: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13525: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13526: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
13527: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13528: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13529: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13530: }else{
13531: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
13532: 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 13533: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 13534: }
1.201 brouard 13535: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 13536: stepm, weightopt,\
13537: model,imx,p,matcov,agemortsup);
13538:
1.302 brouard 13539: free_vector(lsurv,agegomp,AGESUP);
13540: free_vector(lpop,agegomp,AGESUP);
13541: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 13542: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 13543: free_ivector(dcwave,firstobs,lastobs);
13544: free_vector(agecens,firstobs,lastobs);
13545: free_vector(ageexmed,firstobs,lastobs);
13546: free_ivector(cens,firstobs,lastobs);
1.220 brouard 13547: #ifdef GSL
1.136 brouard 13548: #endif
1.186 brouard 13549: } /* Endof if mle==-3 mortality only */
1.205 brouard 13550: /* Standard */
13551: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
13552: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13553: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 13554: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 13555: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
13556: for (k=1; k<=npar;k++)
13557: printf(" %d %8.5f",k,p[k]);
13558: printf("\n");
1.205 brouard 13559: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
13560: /* mlikeli uses func not funcone */
1.247 brouard 13561: /* for(i=1;i<nlstate;i++){ */
13562: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
13563: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
13564: /* } */
1.205 brouard 13565: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
13566: }
13567: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
13568: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13569: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
13570: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13571: }
13572: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 13573: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13574: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
1.335 brouard 13575: /* exit(0); */
1.126 brouard 13576: for (k=1; k<=npar;k++)
13577: printf(" %d %8.5f",k,p[k]);
13578: printf("\n");
13579:
13580: /*--------- results files --------------*/
1.283 brouard 13581: /* 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 13582:
13583:
13584: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13585: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 13586: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13587:
13588: printf("#model= 1 + age ");
13589: fprintf(ficres,"#model= 1 + age ");
13590: fprintf(ficlog,"#model= 1 + age ");
13591: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
13592: </ul>", model);
13593:
13594: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
13595: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
13596: if(nagesqr==1){
13597: printf(" + age*age ");
13598: fprintf(ficres," + age*age ");
13599: fprintf(ficlog," + age*age ");
13600: fprintf(fichtm, "<th>+ age*age</th>");
13601: }
13602: for(j=1;j <=ncovmodel-2;j++){
13603: if(Typevar[j]==0) {
13604: printf(" + V%d ",Tvar[j]);
13605: fprintf(ficres," + V%d ",Tvar[j]);
13606: fprintf(ficlog," + V%d ",Tvar[j]);
13607: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13608: }else if(Typevar[j]==1) {
13609: printf(" + V%d*age ",Tvar[j]);
13610: fprintf(ficres," + V%d*age ",Tvar[j]);
13611: fprintf(ficlog," + V%d*age ",Tvar[j]);
13612: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13613: }else if(Typevar[j]==2) {
13614: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13615: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13616: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13617: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13618: }
13619: }
13620: printf("\n");
13621: fprintf(ficres,"\n");
13622: fprintf(ficlog,"\n");
13623: fprintf(fichtm, "</tr>");
13624: fprintf(fichtm, "\n");
13625:
13626:
1.126 brouard 13627: for(i=1,jk=1; i <=nlstate; i++){
13628: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 13629: if (k != i) {
1.319 brouard 13630: fprintf(fichtm, "<tr>");
1.225 brouard 13631: printf("%d%d ",i,k);
13632: fprintf(ficlog,"%d%d ",i,k);
13633: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 13634: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13635: for(j=1; j <=ncovmodel; j++){
13636: printf("%12.7f ",p[jk]);
13637: fprintf(ficlog,"%12.7f ",p[jk]);
13638: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 13639: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 13640: jk++;
13641: }
13642: printf("\n");
13643: fprintf(ficlog,"\n");
13644: fprintf(ficres,"\n");
1.319 brouard 13645: fprintf(fichtm, "</tr>\n");
1.225 brouard 13646: }
1.126 brouard 13647: }
13648: }
1.319 brouard 13649: /* fprintf(fichtm,"</tr>\n"); */
13650: fprintf(fichtm,"</table>\n");
13651: fprintf(fichtm, "\n");
13652:
1.203 brouard 13653: if(mle != 0){
13654: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 13655: ftolhess=ftol; /* Usually correct */
1.203 brouard 13656: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
13657: 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");
13658: 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 13659: 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 13660: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
13661: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
13662: if(nagesqr==1){
13663: printf(" + age*age ");
13664: fprintf(ficres," + age*age ");
13665: fprintf(ficlog," + age*age ");
13666: fprintf(fichtm, "<th>+ age*age</th>");
13667: }
13668: for(j=1;j <=ncovmodel-2;j++){
13669: if(Typevar[j]==0) {
13670: printf(" + V%d ",Tvar[j]);
13671: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13672: }else if(Typevar[j]==1) {
13673: printf(" + V%d*age ",Tvar[j]);
13674: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13675: }else if(Typevar[j]==2) {
13676: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13677: }
13678: }
13679: fprintf(fichtm, "</tr>\n");
13680:
1.203 brouard 13681: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 13682: for(k=1; k <=(nlstate+ndeath); k++){
13683: if (k != i) {
1.319 brouard 13684: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 13685: printf("%d%d ",i,k);
13686: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 13687: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13688: for(j=1; j <=ncovmodel; j++){
1.319 brouard 13689: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 13690: 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]));
13691: 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 13692: if(fabs(wald) > 1.96){
1.321 brouard 13693: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 13694: }else{
13695: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
13696: }
1.324 brouard 13697: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 13698: 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 13699: jk++;
13700: }
13701: printf("\n");
13702: fprintf(ficlog,"\n");
1.319 brouard 13703: fprintf(fichtm, "</tr>\n");
1.225 brouard 13704: }
13705: }
1.193 brouard 13706: }
1.203 brouard 13707: } /* end of hesscov and Wald tests */
1.319 brouard 13708: fprintf(fichtm,"</table>\n");
1.225 brouard 13709:
1.203 brouard 13710: /* */
1.126 brouard 13711: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
13712: printf("# Scales (for hessian or gradient estimation)\n");
13713: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
13714: for(i=1,jk=1; i <=nlstate; i++){
13715: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 13716: if (j!=i) {
13717: fprintf(ficres,"%1d%1d",i,j);
13718: printf("%1d%1d",i,j);
13719: fprintf(ficlog,"%1d%1d",i,j);
13720: for(k=1; k<=ncovmodel;k++){
13721: printf(" %.5e",delti[jk]);
13722: fprintf(ficlog," %.5e",delti[jk]);
13723: fprintf(ficres," %.5e",delti[jk]);
13724: jk++;
13725: }
13726: printf("\n");
13727: fprintf(ficlog,"\n");
13728: fprintf(ficres,"\n");
13729: }
1.126 brouard 13730: }
13731: }
13732:
13733: 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 13734: if(mle >= 1) /* To big for the screen */
1.126 brouard 13735: 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");
13736: 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");
13737: /* # 121 Var(a12)\n\ */
13738: /* # 122 Cov(b12,a12) Var(b12)\n\ */
13739: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
13740: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
13741: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
13742: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
13743: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
13744: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
13745:
13746:
13747: /* Just to have a covariance matrix which will be more understandable
13748: even is we still don't want to manage dictionary of variables
13749: */
13750: for(itimes=1;itimes<=2;itimes++){
13751: jj=0;
13752: for(i=1; i <=nlstate; i++){
1.225 brouard 13753: for(j=1; j <=nlstate+ndeath; j++){
13754: if(j==i) continue;
13755: for(k=1; k<=ncovmodel;k++){
13756: jj++;
13757: ca[0]= k+'a'-1;ca[1]='\0';
13758: if(itimes==1){
13759: if(mle>=1)
13760: printf("#%1d%1d%d",i,j,k);
13761: fprintf(ficlog,"#%1d%1d%d",i,j,k);
13762: fprintf(ficres,"#%1d%1d%d",i,j,k);
13763: }else{
13764: if(mle>=1)
13765: printf("%1d%1d%d",i,j,k);
13766: fprintf(ficlog,"%1d%1d%d",i,j,k);
13767: fprintf(ficres,"%1d%1d%d",i,j,k);
13768: }
13769: ll=0;
13770: for(li=1;li <=nlstate; li++){
13771: for(lj=1;lj <=nlstate+ndeath; lj++){
13772: if(lj==li) continue;
13773: for(lk=1;lk<=ncovmodel;lk++){
13774: ll++;
13775: if(ll<=jj){
13776: cb[0]= lk +'a'-1;cb[1]='\0';
13777: if(ll<jj){
13778: if(itimes==1){
13779: if(mle>=1)
13780: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13781: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13782: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13783: }else{
13784: if(mle>=1)
13785: printf(" %.5e",matcov[jj][ll]);
13786: fprintf(ficlog," %.5e",matcov[jj][ll]);
13787: fprintf(ficres," %.5e",matcov[jj][ll]);
13788: }
13789: }else{
13790: if(itimes==1){
13791: if(mle>=1)
13792: printf(" Var(%s%1d%1d)",ca,i,j);
13793: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
13794: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
13795: }else{
13796: if(mle>=1)
13797: printf(" %.7e",matcov[jj][ll]);
13798: fprintf(ficlog," %.7e",matcov[jj][ll]);
13799: fprintf(ficres," %.7e",matcov[jj][ll]);
13800: }
13801: }
13802: }
13803: } /* end lk */
13804: } /* end lj */
13805: } /* end li */
13806: if(mle>=1)
13807: printf("\n");
13808: fprintf(ficlog,"\n");
13809: fprintf(ficres,"\n");
13810: numlinepar++;
13811: } /* end k*/
13812: } /*end j */
1.126 brouard 13813: } /* end i */
13814: } /* end itimes */
13815:
13816: fflush(ficlog);
13817: fflush(ficres);
1.225 brouard 13818: while(fgets(line, MAXLINE, ficpar)) {
13819: /* If line starts with a # it is a comment */
13820: if (line[0] == '#') {
13821: numlinepar++;
13822: fputs(line,stdout);
13823: fputs(line,ficparo);
13824: fputs(line,ficlog);
1.299 brouard 13825: fputs(line,ficres);
1.225 brouard 13826: continue;
13827: }else
13828: break;
13829: }
13830:
1.209 brouard 13831: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
13832: /* ungetc(c,ficpar); */
13833: /* fgets(line, MAXLINE, ficpar); */
13834: /* fputs(line,stdout); */
13835: /* fputs(line,ficparo); */
13836: /* } */
13837: /* ungetc(c,ficpar); */
1.126 brouard 13838:
13839: estepm=0;
1.209 brouard 13840: 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 13841:
13842: if (num_filled != 6) {
13843: 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);
13844: 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);
13845: goto end;
13846: }
13847: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
13848: }
13849: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
13850: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
13851:
1.209 brouard 13852: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 13853: if (estepm==0 || estepm < stepm) estepm=stepm;
13854: if (fage <= 2) {
13855: bage = ageminpar;
13856: fage = agemaxpar;
13857: }
13858:
13859: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 13860: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
13861: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 13862:
1.186 brouard 13863: /* Other stuffs, more or less useful */
1.254 brouard 13864: while(fgets(line, MAXLINE, ficpar)) {
13865: /* If line starts with a # it is a comment */
13866: if (line[0] == '#') {
13867: numlinepar++;
13868: fputs(line,stdout);
13869: fputs(line,ficparo);
13870: fputs(line,ficlog);
1.299 brouard 13871: fputs(line,ficres);
1.254 brouard 13872: continue;
13873: }else
13874: break;
13875: }
13876:
13877: 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){
13878:
13879: if (num_filled != 7) {
13880: 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);
13881: 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);
13882: goto end;
13883: }
13884: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13885: 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);
13886: 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);
13887: 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 13888: }
1.254 brouard 13889:
13890: while(fgets(line, MAXLINE, ficpar)) {
13891: /* If line starts with a # it is a comment */
13892: if (line[0] == '#') {
13893: numlinepar++;
13894: fputs(line,stdout);
13895: fputs(line,ficparo);
13896: fputs(line,ficlog);
1.299 brouard 13897: fputs(line,ficres);
1.254 brouard 13898: continue;
13899: }else
13900: break;
1.126 brouard 13901: }
13902:
13903:
13904: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
13905: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
13906:
1.254 brouard 13907: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
13908: if (num_filled != 1) {
13909: 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);
13910: 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);
13911: goto end;
13912: }
13913: printf("pop_based=%d\n",popbased);
13914: fprintf(ficlog,"pop_based=%d\n",popbased);
13915: fprintf(ficparo,"pop_based=%d\n",popbased);
13916: fprintf(ficres,"pop_based=%d\n",popbased);
13917: }
13918:
1.258 brouard 13919: /* Results */
1.332 brouard 13920: /* Value of covariate in each resultine will be compututed (if product) and sorted according to model rank */
13921: /* It is precov[] because we need the varying age in order to compute the real cov[] of the model equation */
13922: precov=matrix(1,MAXRESULTLINESPONE,1,NCOVMAX+1);
1.307 brouard 13923: endishere=0;
1.258 brouard 13924: nresult=0;
1.308 brouard 13925: parameterline=0;
1.258 brouard 13926: do{
13927: if(!fgets(line, MAXLINE, ficpar)){
13928: endishere=1;
1.308 brouard 13929: parameterline=15;
1.258 brouard 13930: }else if (line[0] == '#') {
13931: /* If line starts with a # it is a comment */
1.254 brouard 13932: numlinepar++;
13933: fputs(line,stdout);
13934: fputs(line,ficparo);
13935: fputs(line,ficlog);
1.299 brouard 13936: fputs(line,ficres);
1.254 brouard 13937: continue;
1.258 brouard 13938: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
13939: parameterline=11;
1.296 brouard 13940: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 13941: parameterline=12;
1.307 brouard 13942: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 13943: parameterline=13;
1.307 brouard 13944: }
1.258 brouard 13945: else{
13946: parameterline=14;
1.254 brouard 13947: }
1.308 brouard 13948: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 13949: case 11:
1.296 brouard 13950: 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)){
13951: 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 13952: 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);
13953: 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);
13954: 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);
13955: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 13956: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
13957: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 13958: prvforecast = 1;
13959: }
13960: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 13961: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13962: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13963: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 13964: prvforecast = 2;
13965: }
13966: else {
13967: 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);
13968: 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);
13969: goto end;
1.258 brouard 13970: }
1.254 brouard 13971: break;
1.258 brouard 13972: case 12:
1.296 brouard 13973: 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)){
13974: 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);
13975: 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);
13976: 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);
13977: 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);
13978: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 13979: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
13980: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 13981: prvbackcast = 1;
13982: }
13983: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 13984: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13985: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13986: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 13987: prvbackcast = 2;
13988: }
13989: else {
13990: 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);
13991: 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);
13992: goto end;
1.258 brouard 13993: }
1.230 brouard 13994: break;
1.258 brouard 13995: case 13:
1.332 brouard 13996: num_filled=sscanf(line,"result:%[^\n]\n",resultlineori);
1.307 brouard 13997: nresult++; /* Sum of resultlines */
1.342 brouard 13998: /* printf("Result %d: result:%s\n",nresult, resultlineori); */
1.332 brouard 13999: /* removefirstspace(&resultlineori); */
14000:
14001: if(strstr(resultlineori,"v") !=0){
14002: printf("Error. 'v' must be in upper case 'V' result: %s ",resultlineori);
14003: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultlineori);fflush(ficlog);
14004: return 1;
14005: }
14006: trimbb(resultline, resultlineori); /* Suppressing double blank in the resultline */
1.342 brouard 14007: /* printf("Decoderesult resultline=\"%s\" resultlineori=\"%s\"\n", resultline, resultlineori); */
1.318 brouard 14008: if(nresult > MAXRESULTLINESPONE-1){
14009: 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);
14010: 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 14011: goto end;
14012: }
1.332 brouard 14013:
1.310 brouard 14014: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 14015: fprintf(ficparo,"result: %s\n",resultline);
14016: fprintf(ficres,"result: %s\n",resultline);
14017: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 14018: } else
14019: goto end;
1.307 brouard 14020: break;
14021: case 14:
14022: printf("Error: Unknown command '%s'\n",line);
14023: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 14024: if(line[0] == ' ' || line[0] == '\n'){
14025: printf("It should not be an empty line '%s'\n",line);
14026: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
14027: }
1.307 brouard 14028: if(ncovmodel >=2 && nresult==0 ){
14029: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
14030: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 14031: }
1.307 brouard 14032: /* goto end; */
14033: break;
1.308 brouard 14034: case 15:
14035: printf("End of resultlines.\n");
14036: fprintf(ficlog,"End of resultlines.\n");
14037: break;
14038: default: /* parameterline =0 */
1.307 brouard 14039: nresult=1;
14040: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 14041: } /* End switch parameterline */
14042: }while(endishere==0); /* End do */
1.126 brouard 14043:
1.230 brouard 14044: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 14045: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 14046:
14047: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 14048: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 14049: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 14050: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
14051: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 14052: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 14053: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
14054: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 14055: }else{
1.270 brouard 14056: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 14057: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
14058: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
14059: if(prvforecast==1){
14060: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
14061: jprojd=jproj1;
14062: mprojd=mproj1;
14063: anprojd=anproj1;
14064: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
14065: jprojf=jproj2;
14066: mprojf=mproj2;
14067: anprojf=anproj2;
14068: } else if(prvforecast == 2){
14069: dateprojd=dateintmean;
14070: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
14071: dateprojf=dateintmean+yrfproj;
14072: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
14073: }
14074: if(prvbackcast==1){
14075: datebackd=(jback1+12*mback1+365*anback1)/365;
14076: jbackd=jback1;
14077: mbackd=mback1;
14078: anbackd=anback1;
14079: datebackf=(jback2+12*mback2+365*anback2)/365;
14080: jbackf=jback2;
14081: mbackf=mback2;
14082: anbackf=anback2;
14083: } else if(prvbackcast == 2){
14084: datebackd=dateintmean;
14085: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
14086: datebackf=dateintmean-yrbproj;
14087: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
14088: }
14089:
14090: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 14091: }
14092: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 14093: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
14094: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 14095:
1.225 brouard 14096: /*------------ free_vector -------------*/
14097: /* chdir(path); */
1.220 brouard 14098:
1.215 brouard 14099: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
14100: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
14101: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
14102: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 14103: free_lvector(num,firstobs,lastobs);
14104: free_vector(agedc,firstobs,lastobs);
1.126 brouard 14105: /*free_matrix(covar,0,NCOVMAX,1,n);*/
14106: /*free_matrix(covar,1,NCOVMAX,1,n);*/
14107: fclose(ficparo);
14108: fclose(ficres);
1.220 brouard 14109:
14110:
1.186 brouard 14111: /* Other results (useful)*/
1.220 brouard 14112:
14113:
1.126 brouard 14114: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 14115: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
14116: prlim=matrix(1,nlstate,1,nlstate);
1.332 brouard 14117: /* Computes the prevalence limit for each combination k of the dummy covariates by calling prevalim(k) */
1.209 brouard 14118: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 14119: fclose(ficrespl);
14120:
14121: /*------------- h Pij x at various ages ------------*/
1.180 brouard 14122: /*#include "hpijx.h"*/
1.332 brouard 14123: /** 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?*/
14124: /* calls hpxij with combination k */
1.180 brouard 14125: hPijx(p, bage, fage);
1.145 brouard 14126: fclose(ficrespij);
1.227 brouard 14127:
1.220 brouard 14128: /* ncovcombmax= pow(2,cptcoveff); */
1.332 brouard 14129: /*-------------- Variance of one-step probabilities for a combination ij or for nres ?---*/
1.145 brouard 14130: k=1;
1.126 brouard 14131: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 14132:
1.269 brouard 14133: /* Prevalence for each covariate combination in probs[age][status][cov] */
14134: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
14135: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 14136: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 14137: for(k=1;k<=ncovcombmax;k++)
14138: probs[i][j][k]=0.;
1.269 brouard 14139: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
14140: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 14141: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 14142: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
14143: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 14144: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 14145: for(k=1;k<=ncovcombmax;k++)
14146: mobaverages[i][j][k]=0.;
1.219 brouard 14147: mobaverage=mobaverages;
14148: if (mobilav!=0) {
1.235 brouard 14149: printf("Movingaveraging observed prevalence\n");
1.258 brouard 14150: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 14151: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
14152: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
14153: printf(" Error in movingaverage mobilav=%d\n",mobilav);
14154: }
1.269 brouard 14155: } else if (mobilavproj !=0) {
1.235 brouard 14156: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 14157: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 14158: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
14159: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
14160: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
14161: }
1.269 brouard 14162: }else{
14163: printf("Internal error moving average\n");
14164: fflush(stdout);
14165: exit(1);
1.219 brouard 14166: }
14167: }/* end if moving average */
1.227 brouard 14168:
1.126 brouard 14169: /*---------- Forecasting ------------------*/
1.296 brouard 14170: if(prevfcast==1){
14171: /* /\* if(stepm ==1){*\/ */
14172: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
14173: /*This done previously after freqsummary.*/
14174: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
14175: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
14176:
14177: /* } else if (prvforecast==2){ */
14178: /* /\* if(stepm ==1){*\/ */
14179: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
14180: /* } */
14181: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
14182: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 14183: }
1.269 brouard 14184:
1.296 brouard 14185: /* Prevbcasting */
14186: if(prevbcast==1){
1.219 brouard 14187: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14188: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14189: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14190:
14191: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
14192:
14193: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 14194:
1.219 brouard 14195: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
14196: fclose(ficresplb);
14197:
1.222 brouard 14198: hBijx(p, bage, fage, mobaverage);
14199: fclose(ficrespijb);
1.219 brouard 14200:
1.296 brouard 14201: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
14202: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
14203: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
14204: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
14205: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
14206: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
14207:
14208:
1.269 brouard 14209: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14210:
14211:
1.269 brouard 14212: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 14213: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
14214: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
14215: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 14216: } /* end Prevbcasting */
1.268 brouard 14217:
1.186 brouard 14218:
14219: /* ------ Other prevalence ratios------------ */
1.126 brouard 14220:
1.215 brouard 14221: free_ivector(wav,1,imx);
14222: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
14223: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
14224: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 14225:
14226:
1.127 brouard 14227: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 14228:
1.201 brouard 14229: strcpy(filerese,"E_");
14230: strcat(filerese,fileresu);
1.126 brouard 14231: if((ficreseij=fopen(filerese,"w"))==NULL) {
14232: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
14233: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
14234: }
1.208 brouard 14235: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
14236: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 14237:
14238: pstamp(ficreseij);
1.219 brouard 14239:
1.235 brouard 14240: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
14241: if (cptcovn < 1){i1=1;}
14242:
14243: for(nres=1; nres <= nresult; nres++) /* For each resultline */
14244: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 14245: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 14246: continue;
1.219 brouard 14247: fprintf(ficreseij,"\n#****** ");
1.235 brouard 14248: printf("\n#****** ");
1.225 brouard 14249: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 14250: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
14251: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 14252: }
14253: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.337 brouard 14254: printf(" V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]); /* TvarsQ[j] gives the name of the jth quantitative (fixed or time v) */
14255: fprintf(ficreseij,"V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]);
1.219 brouard 14256: }
14257: fprintf(ficreseij,"******\n");
1.235 brouard 14258: printf("******\n");
1.219 brouard 14259:
14260: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14261: oldm=oldms;savm=savms;
1.330 brouard 14262: /* 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 14263: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 14264:
1.219 brouard 14265: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 14266: }
14267: fclose(ficreseij);
1.208 brouard 14268: printf("done evsij\n");fflush(stdout);
14269: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 14270:
1.218 brouard 14271:
1.227 brouard 14272: /*---------- State-specific expectancies and variances ------------*/
1.336 brouard 14273: /* Should be moved in a function */
1.201 brouard 14274: strcpy(filerest,"T_");
14275: strcat(filerest,fileresu);
1.127 brouard 14276: if((ficrest=fopen(filerest,"w"))==NULL) {
14277: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
14278: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
14279: }
1.208 brouard 14280: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
14281: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 14282: strcpy(fileresstde,"STDE_");
14283: strcat(fileresstde,fileresu);
1.126 brouard 14284: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 14285: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
14286: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 14287: }
1.227 brouard 14288: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
14289: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 14290:
1.201 brouard 14291: strcpy(filerescve,"CVE_");
14292: strcat(filerescve,fileresu);
1.126 brouard 14293: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 14294: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
14295: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 14296: }
1.227 brouard 14297: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
14298: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 14299:
1.201 brouard 14300: strcpy(fileresv,"V_");
14301: strcat(fileresv,fileresu);
1.126 brouard 14302: if((ficresvij=fopen(fileresv,"w"))==NULL) {
14303: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
14304: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
14305: }
1.227 brouard 14306: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
14307: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 14308:
1.235 brouard 14309: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
14310: if (cptcovn < 1){i1=1;}
14311:
1.334 brouard 14312: for(nres=1; nres <= nresult; nres++) /* For each resultline, find the combination and output results according to the values of dummies and then quanti. */
14313: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying. For each nres and each value at position k
14314: * we know Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline
14315: * Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline
14316: * and Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
14317: /* */
14318: 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 14319: continue;
1.321 brouard 14320: printf("\n# model %s \n#****** Result for:", model);
14321: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
14322: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.334 brouard 14323: /* It might not be a good idea to mix dummies and quantitative */
14324: /* 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 *\/ */
14325: 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 */
14326: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* Output by variables in the resultline *\/ */
14327: /* Tvaraff[j] is the name of the dummy variable in position j in the equation model:
14328: * Tvaraff[1]@9={4, 3, 0, 0, 0, 0, 0, 0, 0}, in model=V5+V4+V3+V4*V3+V5*age
14329: * (V5 is quanti) V4 and V3 are dummies
14330: * TnsdVar[4] is the position 1 and TnsdVar[3]=2 in codtabm(k,l)(V4 V3)=V4 V3
14331: * l=1 l=2
14332: * k=1 1 1 0 0
14333: * k=2 2 1 1 0
14334: * k=3 [1] [2] 0 1
14335: * k=4 2 2 1 1
14336: * If nres=1 result: V3=1 V4=0 then k=3 and outputs
14337: * If nres=2 result: V4=1 V3=0 then k=2 and outputs
14338: * nres=1 =>k=3 j=1 V4= nbcode[4][codtabm(3,1)=1)=0; j=2 V3= nbcode[3][codtabm(3,2)=2]=1
14339: * nres=2 =>k=2 j=1 V4= nbcode[4][codtabm(2,1)=2)=1; j=2 V3= nbcode[3][codtabm(2,2)=1]=0
14340: */
14341: /* Tvresult[nres][j] Name of the variable at position j in this resultline */
14342: /* Tresult[nres][j] Value of this variable at position j could be a float if quantitative */
14343: /* We give up with the combinations!! */
1.342 brouard 14344: /* if(debugILK) */
14345: /* 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 14346:
14347: if(Dummy[modelresult[nres][j]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to j in resultline */
1.337 brouard 14348: 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 */
14349: fprintf(ficlog,"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 */
14350: fprintf(ficrest,"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 */
1.334 brouard 14351: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
14352: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
14353: }else{
14354: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
14355: }
14356: /* fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14357: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14358: }else if(Dummy[modelresult[nres][j]]==1){ /* Quanti variable */
14359: /* For each selected (single) quantitative value */
1.337 brouard 14360: printf(" V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
14361: fprintf(ficlog," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
14362: fprintf(ficrest," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
1.334 brouard 14363: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
14364: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
14365: }else{
14366: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
14367: }
14368: }else{
14369: 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 */
14370: 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 */
14371: exit(1);
14372: }
1.335 brouard 14373: } /* End loop for each variable in the resultline */
1.334 brouard 14374: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
14375: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /\* Wrong j is not in the equation model *\/ */
14376: /* fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14377: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14378: /* } */
1.208 brouard 14379: fprintf(ficrest,"******\n");
1.227 brouard 14380: fprintf(ficlog,"******\n");
14381: printf("******\n");
1.208 brouard 14382:
14383: fprintf(ficresstdeij,"\n#****** ");
14384: fprintf(ficrescveij,"\n#****** ");
1.337 brouard 14385: /* It could have been: for(j=1;j<=cptcoveff;j++) {printf("V=%d=%lg",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);} */
14386: /* But it won't be sorted and depends on how the resultline is ordered */
1.225 brouard 14387: for(j=1;j<=cptcoveff;j++) {
1.334 brouard 14388: fprintf(ficresstdeij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14389: /* fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14390: /* fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14391: }
14392: 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 14393: fprintf(ficresstdeij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
14394: fprintf(ficrescveij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
1.235 brouard 14395: }
1.208 brouard 14396: fprintf(ficresstdeij,"******\n");
14397: fprintf(ficrescveij,"******\n");
14398:
14399: fprintf(ficresvij,"\n#****** ");
1.238 brouard 14400: /* pstamp(ficresvij); */
1.225 brouard 14401: for(j=1;j<=cptcoveff;j++)
1.335 brouard 14402: fprintf(ficresvij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14403: /* fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[TnsdVar[Tvaraff[j]]])]); */
1.235 brouard 14404: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 14405: /* fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]); /\* To solve *\/ */
1.337 brouard 14406: fprintf(ficresvij," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /* Solved */
1.235 brouard 14407: }
1.208 brouard 14408: fprintf(ficresvij,"******\n");
14409:
14410: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14411: oldm=oldms;savm=savms;
1.235 brouard 14412: printf(" cvevsij ");
14413: fprintf(ficlog, " cvevsij ");
14414: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 14415: printf(" end cvevsij \n ");
14416: fprintf(ficlog, " end cvevsij \n ");
14417:
14418: /*
14419: */
14420: /* goto endfree; */
14421:
14422: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14423: pstamp(ficrest);
14424:
1.269 brouard 14425: epj=vector(1,nlstate+1);
1.208 brouard 14426: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 14427: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
14428: cptcod= 0; /* To be deleted */
14429: printf("varevsij vpopbased=%d \n",vpopbased);
14430: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 14431: 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 14432: 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 ");
14433: if(vpopbased==1)
14434: 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);
14435: else
1.288 brouard 14436: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.335 brouard 14437: fprintf(ficrest,"# Age popbased mobilav e.. (std) "); /* Adding covariate values? */
1.227 brouard 14438: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
14439: fprintf(ficrest,"\n");
14440: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 14441: printf("Computing age specific forward period (stable) prevalences in each health state \n");
14442: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 14443: for(age=bage; age <=fage ;age++){
1.235 brouard 14444: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 14445: if (vpopbased==1) {
14446: if(mobilav ==0){
14447: for(i=1; i<=nlstate;i++)
14448: prlim[i][i]=probs[(int)age][i][k];
14449: }else{ /* mobilav */
14450: for(i=1; i<=nlstate;i++)
14451: prlim[i][i]=mobaverage[(int)age][i][k];
14452: }
14453: }
1.219 brouard 14454:
1.227 brouard 14455: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
14456: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
14457: /* printf(" age %4.0f ",age); */
14458: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
14459: for(i=1, epj[j]=0.;i <=nlstate;i++) {
14460: epj[j] += prlim[i][i]*eij[i][j][(int)age];
14461: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
14462: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
14463: }
14464: epj[nlstate+1] +=epj[j];
14465: }
14466: /* printf(" age %4.0f \n",age); */
1.219 brouard 14467:
1.227 brouard 14468: for(i=1, vepp=0.;i <=nlstate;i++)
14469: for(j=1;j <=nlstate;j++)
14470: vepp += vareij[i][j][(int)age];
14471: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
14472: for(j=1;j <=nlstate;j++){
14473: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
14474: }
14475: fprintf(ficrest,"\n");
14476: }
1.208 brouard 14477: } /* End vpopbased */
1.269 brouard 14478: free_vector(epj,1,nlstate+1);
1.208 brouard 14479: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
14480: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 14481: printf("done selection\n");fflush(stdout);
14482: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 14483:
1.335 brouard 14484: } /* End k selection or end covariate selection for nres */
1.227 brouard 14485:
14486: printf("done State-specific expectancies\n");fflush(stdout);
14487: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
14488:
1.335 brouard 14489: /* variance-covariance of forward period prevalence */
1.269 brouard 14490: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14491:
1.227 brouard 14492:
1.290 brouard 14493: free_vector(weight,firstobs,lastobs);
1.330 brouard 14494: free_imatrix(Tvardk,1,NCOVMAX,1,2);
1.227 brouard 14495: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 14496: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
14497: free_matrix(anint,1,maxwav,firstobs,lastobs);
14498: free_matrix(mint,1,maxwav,firstobs,lastobs);
14499: free_ivector(cod,firstobs,lastobs);
1.227 brouard 14500: free_ivector(tab,1,NCOVMAX);
14501: fclose(ficresstdeij);
14502: fclose(ficrescveij);
14503: fclose(ficresvij);
14504: fclose(ficrest);
14505: fclose(ficpar);
14506:
14507:
1.126 brouard 14508: /*---------- End : free ----------------*/
1.219 brouard 14509: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 14510: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
14511: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 14512: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
14513: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 14514: } /* mle==-3 arrives here for freeing */
1.227 brouard 14515: /* endfree:*/
14516: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
14517: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
14518: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.341 brouard 14519: /* if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs); */
14520: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs);
1.290 brouard 14521: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
14522: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
14523: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 14524: free_matrix(matcov,1,npar,1,npar);
14525: free_matrix(hess,1,npar,1,npar);
14526: /*free_vector(delti,1,npar);*/
14527: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14528: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 14529: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 14530: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14531:
14532: free_ivector(ncodemax,1,NCOVMAX);
14533: free_ivector(ncodemaxwundef,1,NCOVMAX);
14534: free_ivector(Dummy,-1,NCOVMAX);
14535: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 14536: free_ivector(DummyV,1,NCOVMAX);
14537: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 14538: free_ivector(Typevar,-1,NCOVMAX);
14539: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 14540: free_ivector(TvarsQ,1,NCOVMAX);
14541: free_ivector(TvarsQind,1,NCOVMAX);
14542: free_ivector(TvarsD,1,NCOVMAX);
1.330 brouard 14543: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 14544: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 14545: free_ivector(TvarFD,1,NCOVMAX);
14546: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 14547: free_ivector(TvarF,1,NCOVMAX);
14548: free_ivector(TvarFind,1,NCOVMAX);
14549: free_ivector(TvarV,1,NCOVMAX);
14550: free_ivector(TvarVind,1,NCOVMAX);
14551: free_ivector(TvarA,1,NCOVMAX);
14552: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 14553: free_ivector(TvarFQ,1,NCOVMAX);
14554: free_ivector(TvarFQind,1,NCOVMAX);
14555: free_ivector(TvarVD,1,NCOVMAX);
14556: free_ivector(TvarVDind,1,NCOVMAX);
14557: free_ivector(TvarVQ,1,NCOVMAX);
14558: free_ivector(TvarVQind,1,NCOVMAX);
1.339 brouard 14559: free_ivector(TvarVV,1,NCOVMAX);
14560: free_ivector(TvarVVind,1,NCOVMAX);
14561:
1.230 brouard 14562: free_ivector(Tvarsel,1,NCOVMAX);
14563: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 14564: free_ivector(Tposprod,1,NCOVMAX);
14565: free_ivector(Tprod,1,NCOVMAX);
14566: free_ivector(Tvaraff,1,NCOVMAX);
1.338 brouard 14567: free_ivector(invalidvarcomb,0,ncovcombmax);
1.227 brouard 14568: free_ivector(Tage,1,NCOVMAX);
14569: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 14570: free_ivector(TmodelInvind,1,NCOVMAX);
14571: free_ivector(TmodelInvQind,1,NCOVMAX);
1.332 brouard 14572:
14573: free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /* Could be elsewhere ?*/
14574:
1.227 brouard 14575: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
14576: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 14577: fflush(fichtm);
14578: fflush(ficgp);
14579:
1.227 brouard 14580:
1.126 brouard 14581: if((nberr >0) || (nbwarn>0)){
1.216 brouard 14582: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
14583: 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 14584: }else{
14585: printf("End of Imach\n");
14586: fprintf(ficlog,"End of Imach\n");
14587: }
14588: printf("See log file on %s\n",filelog);
14589: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 14590: /*(void) gettimeofday(&end_time,&tzp);*/
14591: rend_time = time(NULL);
14592: end_time = *localtime(&rend_time);
14593: /* tml = *localtime(&end_time.tm_sec); */
14594: strcpy(strtend,asctime(&end_time));
1.126 brouard 14595: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
14596: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 14597: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 14598:
1.157 brouard 14599: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
14600: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
14601: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 14602: /* printf("Total time was %d uSec.\n", total_usecs);*/
14603: /* if(fileappend(fichtm,optionfilehtm)){ */
14604: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14605: fclose(fichtm);
14606: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14607: fclose(fichtmcov);
14608: fclose(ficgp);
14609: fclose(ficlog);
14610: /*------ End -----------*/
1.227 brouard 14611:
1.281 brouard 14612:
14613: /* Executes gnuplot */
1.227 brouard 14614:
14615: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 14616: #ifdef WIN32
1.227 brouard 14617: if (_chdir(pathcd) != 0)
14618: printf("Can't move to directory %s!\n",path);
14619: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 14620: #else
1.227 brouard 14621: if(chdir(pathcd) != 0)
14622: printf("Can't move to directory %s!\n", path);
14623: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 14624: #endif
1.126 brouard 14625: printf("Current directory %s!\n",pathcd);
14626: /*strcat(plotcmd,CHARSEPARATOR);*/
14627: sprintf(plotcmd,"gnuplot");
1.157 brouard 14628: #ifdef _WIN32
1.126 brouard 14629: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
14630: #endif
14631: if(!stat(plotcmd,&info)){
1.158 brouard 14632: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14633: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 14634: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 14635: }else
14636: strcpy(pplotcmd,plotcmd);
1.157 brouard 14637: #ifdef __unix
1.126 brouard 14638: strcpy(plotcmd,GNUPLOTPROGRAM);
14639: if(!stat(plotcmd,&info)){
1.158 brouard 14640: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14641: }else
14642: strcpy(pplotcmd,plotcmd);
14643: #endif
14644: }else
14645: strcpy(pplotcmd,plotcmd);
14646:
14647: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 14648: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 14649: strcpy(pplotcmd,plotcmd);
1.227 brouard 14650:
1.126 brouard 14651: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 14652: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 14653: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 14654: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 14655: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 14656: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 14657: strcpy(plotcmd,pplotcmd);
14658: }
1.126 brouard 14659: }
1.158 brouard 14660: printf(" Successful, please wait...");
1.126 brouard 14661: while (z[0] != 'q') {
14662: /* chdir(path); */
1.154 brouard 14663: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 14664: scanf("%s",z);
14665: /* if (z[0] == 'c') system("./imach"); */
14666: if (z[0] == 'e') {
1.158 brouard 14667: #ifdef __APPLE__
1.152 brouard 14668: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 14669: #elif __linux
14670: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 14671: #else
1.152 brouard 14672: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 14673: #endif
14674: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
14675: system(pplotcmd);
1.126 brouard 14676: }
14677: else if (z[0] == 'g') system(plotcmd);
14678: else if (z[0] == 'q') exit(0);
14679: }
1.227 brouard 14680: end:
1.126 brouard 14681: while (z[0] != 'q') {
1.195 brouard 14682: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 14683: scanf("%s",z);
14684: }
1.283 brouard 14685: printf("End\n");
1.282 brouard 14686: exit(0);
1.126 brouard 14687: }
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