Annotation of imach/src/imach.c, revision 1.342
1.342 ! brouard 1: /* $Id: imach.c,v 1.341 2022/09/11 07:58:42 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.342 ! brouard 4: Revision 1.341 2022/09/11 07:58:42 brouard
! 5: Summary: Version 0.99r38
! 6:
! 7: After adding change in cotvar.
! 8:
1.341 brouard 9: Revision 1.340 2022/09/11 07:53:11 brouard
10: Summary: Version imach 0.99r37
11:
12: * imach.c (Module): Adding timevarying products of any kinds,
13: should work before shifting cotvar from ncovcol+nqv columns in
14: order to have a correspondance between the column of cotvar and
15: the id of column.
16:
1.340 brouard 17: Revision 1.339 2022/09/09 17:55:22 brouard
18: Summary: version 0.99r37
19:
20: * imach.c (Module): Many improvements for fixing products of fixed
21: timevarying as well as fixed * fixed, and test with quantitative
22: covariate.
23:
1.339 brouard 24: Revision 1.338 2022/09/04 17:40:33 brouard
25: Summary: 0.99r36
26:
27: * imach.c (Module): Now the easy runs i.e. without result or
28: model=1+age only did not work. The defautl combination should be 1
29: and not 0 because everything hasn't been tranformed yet.
30:
1.338 brouard 31: Revision 1.337 2022/09/02 14:26:02 brouard
32: Summary: version 0.99r35
33:
34: * src/imach.c: Version 0.99r35 because it outputs same results with
35: 1+age+V1+V1*age for females and 1+age for females only
36: (education=1 noweight)
37:
1.337 brouard 38: Revision 1.336 2022/08/31 09:52:36 brouard
39: *** empty log message ***
40:
1.336 brouard 41: Revision 1.335 2022/08/31 08:23:16 brouard
42: Summary: improvements...
43:
1.335 brouard 44: Revision 1.334 2022/08/25 09:08:41 brouard
45: Summary: In progress for quantitative
46:
1.334 brouard 47: Revision 1.333 2022/08/21 09:10:30 brouard
48: * src/imach.c (Module): Version 0.99r33 A lot of changes in
49: reassigning covariates: my first idea was that people will always
50: use the first covariate V1 into the model but in fact they are
51: producing data with many covariates and can use an equation model
52: with some of the covariate; it means that in a model V2+V3 instead
53: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
54: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
55: the equation model is restricted to two variables only (V2, V3)
56: and the combination for V2 should be codtabm(k,1) instead of
57: (codtabm(k,2), and the code should be
58: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
59: made. All of these should be simplified once a day like we did in
60: hpxij() for example by using precov[nres] which is computed in
61: decoderesult for each nres of each resultline. Loop should be done
62: on the equation model globally by distinguishing only product with
63: age (which are changing with age) and no more on type of
64: covariates, single dummies, single covariates.
65:
1.333 brouard 66: Revision 1.332 2022/08/21 09:06:25 brouard
67: Summary: Version 0.99r33
68:
69: * src/imach.c (Module): Version 0.99r33 A lot of changes in
70: reassigning covariates: my first idea was that people will always
71: use the first covariate V1 into the model but in fact they are
72: producing data with many covariates and can use an equation model
73: with some of the covariate; it means that in a model V2+V3 instead
74: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
75: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
76: the equation model is restricted to two variables only (V2, V3)
77: and the combination for V2 should be codtabm(k,1) instead of
78: (codtabm(k,2), and the code should be
79: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
80: made. All of these should be simplified once a day like we did in
81: hpxij() for example by using precov[nres] which is computed in
82: decoderesult for each nres of each resultline. Loop should be done
83: on the equation model globally by distinguishing only product with
84: age (which are changing with age) and no more on type of
85: covariates, single dummies, single covariates.
86:
1.332 brouard 87: Revision 1.331 2022/08/07 05:40:09 brouard
88: *** empty log message ***
89:
1.331 brouard 90: Revision 1.330 2022/08/06 07:18:25 brouard
91: Summary: last 0.99r31
92:
93: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function
94:
1.330 brouard 95: Revision 1.329 2022/08/03 17:29:54 brouard
96: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
97:
1.329 brouard 98: Revision 1.328 2022/07/27 17:40:48 brouard
99: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
100:
1.328 brouard 101: Revision 1.327 2022/07/27 14:47:35 brouard
102: Summary: Still a problem for one-step probabilities in case of quantitative variables
103:
1.327 brouard 104: Revision 1.326 2022/07/26 17:33:55 brouard
105: Summary: some test with nres=1
106:
1.326 brouard 107: Revision 1.325 2022/07/25 14:27:23 brouard
108: Summary: r30
109:
110: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
111: coredumped, revealed by Feiuno, thank you.
112:
1.325 brouard 113: Revision 1.324 2022/07/23 17:44:26 brouard
114: *** empty log message ***
115:
1.324 brouard 116: Revision 1.323 2022/07/22 12:30:08 brouard
117: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
118:
1.323 brouard 119: Revision 1.322 2022/07/22 12:27:48 brouard
120: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
121:
1.322 brouard 122: Revision 1.321 2022/07/22 12:04:24 brouard
123: Summary: r28
124:
125: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
126:
1.321 brouard 127: Revision 1.320 2022/06/02 05:10:11 brouard
128: *** empty log message ***
129:
1.320 brouard 130: Revision 1.319 2022/06/02 04:45:11 brouard
131: * imach.c (Module): Adding the Wald tests from the log to the main
132: htm for better display of the maximum likelihood estimators.
133:
1.319 brouard 134: Revision 1.318 2022/05/24 08:10:59 brouard
135: * imach.c (Module): Some attempts to find a bug of wrong estimates
136: of confidencce intervals with product in the equation modelC
137:
1.318 brouard 138: Revision 1.317 2022/05/15 15:06:23 brouard
139: * imach.c (Module): Some minor improvements
140:
1.317 brouard 141: Revision 1.316 2022/05/11 15:11:31 brouard
142: Summary: r27
143:
1.316 brouard 144: Revision 1.315 2022/05/11 15:06:32 brouard
145: *** empty log message ***
146:
1.315 brouard 147: Revision 1.314 2022/04/13 17:43:09 brouard
148: * imach.c (Module): Adding link to text data files
149:
1.314 brouard 150: Revision 1.313 2022/04/11 15:57:42 brouard
151: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
152:
1.313 brouard 153: Revision 1.312 2022/04/05 21:24:39 brouard
154: *** empty log message ***
155:
1.312 brouard 156: Revision 1.311 2022/04/05 21:03:51 brouard
157: Summary: Fixed quantitative covariates
158:
159: Fixed covariates (dummy or quantitative)
160: with missing values have never been allowed but are ERRORS and
161: program quits. Standard deviations of fixed covariates were
162: wrongly computed. Mean and standard deviations of time varying
163: covariates are still not computed.
164:
1.311 brouard 165: Revision 1.310 2022/03/17 08:45:53 brouard
166: Summary: 99r25
167:
168: Improving detection of errors: result lines should be compatible with
169: the model.
170:
1.310 brouard 171: Revision 1.309 2021/05/20 12:39:14 brouard
172: Summary: Version 0.99r24
173:
1.309 brouard 174: Revision 1.308 2021/03/31 13:11:57 brouard
175: Summary: Version 0.99r23
176:
177:
178: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
179:
1.308 brouard 180: Revision 1.307 2021/03/08 18:11:32 brouard
181: Summary: 0.99r22 fixed bug on result:
182:
1.307 brouard 183: Revision 1.306 2021/02/20 15:44:02 brouard
184: Summary: Version 0.99r21
185:
186: * imach.c (Module): Fix bug on quitting after result lines!
187: (Module): Version 0.99r21
188:
1.306 brouard 189: Revision 1.305 2021/02/20 15:28:30 brouard
190: * imach.c (Module): Fix bug on quitting after result lines!
191:
1.305 brouard 192: Revision 1.304 2021/02/12 11:34:20 brouard
193: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
194:
1.304 brouard 195: Revision 1.303 2021/02/11 19:50:15 brouard
196: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
197:
1.303 brouard 198: Revision 1.302 2020/02/22 21:00:05 brouard
199: * (Module): imach.c Update mle=-3 (for computing Life expectancy
200: and life table from the data without any state)
201:
1.302 brouard 202: Revision 1.301 2019/06/04 13:51:20 brouard
203: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
204:
1.301 brouard 205: Revision 1.300 2019/05/22 19:09:45 brouard
206: Summary: version 0.99r19 of May 2019
207:
1.300 brouard 208: Revision 1.299 2019/05/22 18:37:08 brouard
209: Summary: Cleaned 0.99r19
210:
1.299 brouard 211: Revision 1.298 2019/05/22 18:19:56 brouard
212: *** empty log message ***
213:
1.298 brouard 214: Revision 1.297 2019/05/22 17:56:10 brouard
215: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
216:
1.297 brouard 217: Revision 1.296 2019/05/20 13:03:18 brouard
218: Summary: Projection syntax simplified
219:
220:
221: We can now start projections, forward or backward, from the mean date
222: of inteviews up to or down to a number of years of projection:
223: prevforecast=1 yearsfproj=15.3 mobil_average=0
224: or
225: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
226: or
227: prevbackcast=1 yearsbproj=12.3 mobil_average=1
228: or
229: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
230:
1.296 brouard 231: Revision 1.295 2019/05/18 09:52:50 brouard
232: Summary: doxygen tex bug
233:
1.295 brouard 234: Revision 1.294 2019/05/16 14:54:33 brouard
235: Summary: There was some wrong lines added
236:
1.294 brouard 237: Revision 1.293 2019/05/09 15:17:34 brouard
238: *** empty log message ***
239:
1.293 brouard 240: Revision 1.292 2019/05/09 14:17:20 brouard
241: Summary: Some updates
242:
1.292 brouard 243: Revision 1.291 2019/05/09 13:44:18 brouard
244: Summary: Before ncovmax
245:
1.291 brouard 246: Revision 1.290 2019/05/09 13:39:37 brouard
247: Summary: 0.99r18 unlimited number of individuals
248:
249: 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.
250:
1.290 brouard 251: Revision 1.289 2018/12/13 09:16:26 brouard
252: Summary: Bug for young ages (<-30) will be in r17
253:
1.289 brouard 254: Revision 1.288 2018/05/02 20:58:27 brouard
255: Summary: Some bugs fixed
256:
1.288 brouard 257: Revision 1.287 2018/05/01 17:57:25 brouard
258: Summary: Bug fixed by providing frequencies only for non missing covariates
259:
1.287 brouard 260: Revision 1.286 2018/04/27 14:27:04 brouard
261: Summary: some minor bugs
262:
1.286 brouard 263: Revision 1.285 2018/04/21 21:02:16 brouard
264: Summary: Some bugs fixed, valgrind tested
265:
1.285 brouard 266: Revision 1.284 2018/04/20 05:22:13 brouard
267: Summary: Computing mean and stdeviation of fixed quantitative variables
268:
1.284 brouard 269: Revision 1.283 2018/04/19 14:49:16 brouard
270: Summary: Some minor bugs fixed
271:
1.283 brouard 272: Revision 1.282 2018/02/27 22:50:02 brouard
273: *** empty log message ***
274:
1.282 brouard 275: Revision 1.281 2018/02/27 19:25:23 brouard
276: Summary: Adding second argument for quitting
277:
1.281 brouard 278: Revision 1.280 2018/02/21 07:58:13 brouard
279: Summary: 0.99r15
280:
281: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
282:
1.280 brouard 283: Revision 1.279 2017/07/20 13:35:01 brouard
284: Summary: temporary working
285:
1.279 brouard 286: Revision 1.278 2017/07/19 14:09:02 brouard
287: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
288:
1.278 brouard 289: Revision 1.277 2017/07/17 08:53:49 brouard
290: Summary: BOM files can be read now
291:
1.277 brouard 292: Revision 1.276 2017/06/30 15:48:31 brouard
293: Summary: Graphs improvements
294:
1.276 brouard 295: Revision 1.275 2017/06/30 13:39:33 brouard
296: Summary: Saito's color
297:
1.275 brouard 298: Revision 1.274 2017/06/29 09:47:08 brouard
299: Summary: Version 0.99r14
300:
1.274 brouard 301: Revision 1.273 2017/06/27 11:06:02 brouard
302: Summary: More documentation on projections
303:
1.273 brouard 304: Revision 1.272 2017/06/27 10:22:40 brouard
305: Summary: Color of backprojection changed from 6 to 5(yellow)
306:
1.272 brouard 307: Revision 1.271 2017/06/27 10:17:50 brouard
308: Summary: Some bug with rint
309:
1.271 brouard 310: Revision 1.270 2017/05/24 05:45:29 brouard
311: *** empty log message ***
312:
1.270 brouard 313: Revision 1.269 2017/05/23 08:39:25 brouard
314: Summary: Code into subroutine, cleanings
315:
1.269 brouard 316: Revision 1.268 2017/05/18 20:09:32 brouard
317: Summary: backprojection and confidence intervals of backprevalence
318:
1.268 brouard 319: Revision 1.267 2017/05/13 10:25:05 brouard
320: Summary: temporary save for backprojection
321:
1.267 brouard 322: Revision 1.266 2017/05/13 07:26:12 brouard
323: Summary: Version 0.99r13 (improvements and bugs fixed)
324:
1.266 brouard 325: Revision 1.265 2017/04/26 16:22:11 brouard
326: Summary: imach 0.99r13 Some bugs fixed
327:
1.265 brouard 328: Revision 1.264 2017/04/26 06:01:29 brouard
329: Summary: Labels in graphs
330:
1.264 brouard 331: Revision 1.263 2017/04/24 15:23:15 brouard
332: Summary: to save
333:
1.263 brouard 334: Revision 1.262 2017/04/18 16:48:12 brouard
335: *** empty log message ***
336:
1.262 brouard 337: Revision 1.261 2017/04/05 10:14:09 brouard
338: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
339:
1.261 brouard 340: Revision 1.260 2017/04/04 17:46:59 brouard
341: Summary: Gnuplot indexations fixed (humm)
342:
1.260 brouard 343: Revision 1.259 2017/04/04 13:01:16 brouard
344: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
345:
1.259 brouard 346: Revision 1.258 2017/04/03 10:17:47 brouard
347: Summary: Version 0.99r12
348:
349: Some cleanings, conformed with updated documentation.
350:
1.258 brouard 351: Revision 1.257 2017/03/29 16:53:30 brouard
352: Summary: Temp
353:
1.257 brouard 354: Revision 1.256 2017/03/27 05:50:23 brouard
355: Summary: Temporary
356:
1.256 brouard 357: Revision 1.255 2017/03/08 16:02:28 brouard
358: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
359:
1.255 brouard 360: Revision 1.254 2017/03/08 07:13:00 brouard
361: Summary: Fixing data parameter line
362:
1.254 brouard 363: Revision 1.253 2016/12/15 11:59:41 brouard
364: Summary: 0.99 in progress
365:
1.253 brouard 366: Revision 1.252 2016/09/15 21:15:37 brouard
367: *** empty log message ***
368:
1.252 brouard 369: Revision 1.251 2016/09/15 15:01:13 brouard
370: Summary: not working
371:
1.251 brouard 372: Revision 1.250 2016/09/08 16:07:27 brouard
373: Summary: continue
374:
1.250 brouard 375: Revision 1.249 2016/09/07 17:14:18 brouard
376: Summary: Starting values from frequencies
377:
1.249 brouard 378: Revision 1.248 2016/09/07 14:10:18 brouard
379: *** empty log message ***
380:
1.248 brouard 381: Revision 1.247 2016/09/02 11:11:21 brouard
382: *** empty log message ***
383:
1.247 brouard 384: Revision 1.246 2016/09/02 08:49:22 brouard
385: *** empty log message ***
386:
1.246 brouard 387: Revision 1.245 2016/09/02 07:25:01 brouard
388: *** empty log message ***
389:
1.245 brouard 390: Revision 1.244 2016/09/02 07:17:34 brouard
391: *** empty log message ***
392:
1.244 brouard 393: Revision 1.243 2016/09/02 06:45:35 brouard
394: *** empty log message ***
395:
1.243 brouard 396: Revision 1.242 2016/08/30 15:01:20 brouard
397: Summary: Fixing a lots
398:
1.242 brouard 399: Revision 1.241 2016/08/29 17:17:25 brouard
400: Summary: gnuplot problem in Back projection to fix
401:
1.241 brouard 402: Revision 1.240 2016/08/29 07:53:18 brouard
403: Summary: Better
404:
1.240 brouard 405: Revision 1.239 2016/08/26 15:51:03 brouard
406: Summary: Improvement in Powell output in order to copy and paste
407:
408: Author:
409:
1.239 brouard 410: Revision 1.238 2016/08/26 14:23:35 brouard
411: Summary: Starting tests of 0.99
412:
1.238 brouard 413: Revision 1.237 2016/08/26 09:20:19 brouard
414: Summary: to valgrind
415:
1.237 brouard 416: Revision 1.236 2016/08/25 10:50:18 brouard
417: *** empty log message ***
418:
1.236 brouard 419: Revision 1.235 2016/08/25 06:59:23 brouard
420: *** empty log message ***
421:
1.235 brouard 422: Revision 1.234 2016/08/23 16:51:20 brouard
423: *** empty log message ***
424:
1.234 brouard 425: Revision 1.233 2016/08/23 07:40:50 brouard
426: Summary: not working
427:
1.233 brouard 428: Revision 1.232 2016/08/22 14:20:21 brouard
429: Summary: not working
430:
1.232 brouard 431: Revision 1.231 2016/08/22 07:17:15 brouard
432: Summary: not working
433:
1.231 brouard 434: Revision 1.230 2016/08/22 06:55:53 brouard
435: Summary: Not working
436:
1.230 brouard 437: Revision 1.229 2016/07/23 09:45:53 brouard
438: Summary: Completing for func too
439:
1.229 brouard 440: Revision 1.228 2016/07/22 17:45:30 brouard
441: Summary: Fixing some arrays, still debugging
442:
1.227 brouard 443: Revision 1.226 2016/07/12 18:42:34 brouard
444: Summary: temp
445:
1.226 brouard 446: Revision 1.225 2016/07/12 08:40:03 brouard
447: Summary: saving but not running
448:
1.225 brouard 449: Revision 1.224 2016/07/01 13:16:01 brouard
450: Summary: Fixes
451:
1.224 brouard 452: Revision 1.223 2016/02/19 09:23:35 brouard
453: Summary: temporary
454:
1.223 brouard 455: Revision 1.222 2016/02/17 08:14:50 brouard
456: Summary: Probably last 0.98 stable version 0.98r6
457:
1.222 brouard 458: Revision 1.221 2016/02/15 23:35:36 brouard
459: Summary: minor bug
460:
1.220 brouard 461: Revision 1.219 2016/02/15 00:48:12 brouard
462: *** empty log message ***
463:
1.219 brouard 464: Revision 1.218 2016/02/12 11:29:23 brouard
465: Summary: 0.99 Back projections
466:
1.218 brouard 467: Revision 1.217 2015/12/23 17:18:31 brouard
468: Summary: Experimental backcast
469:
1.217 brouard 470: Revision 1.216 2015/12/18 17:32:11 brouard
471: Summary: 0.98r4 Warning and status=-2
472:
473: Version 0.98r4 is now:
474: - displaying an error when status is -1, date of interview unknown and date of death known;
475: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
476: Older changes concerning s=-2, dating from 2005 have been supersed.
477:
1.216 brouard 478: Revision 1.215 2015/12/16 08:52:24 brouard
479: Summary: 0.98r4 working
480:
1.215 brouard 481: Revision 1.214 2015/12/16 06:57:54 brouard
482: Summary: temporary not working
483:
1.214 brouard 484: Revision 1.213 2015/12/11 18:22:17 brouard
485: Summary: 0.98r4
486:
1.213 brouard 487: Revision 1.212 2015/11/21 12:47:24 brouard
488: Summary: minor typo
489:
1.212 brouard 490: Revision 1.211 2015/11/21 12:41:11 brouard
491: Summary: 0.98r3 with some graph of projected cross-sectional
492:
493: Author: Nicolas Brouard
494:
1.211 brouard 495: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 496: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 497: Summary: Adding ftolpl parameter
498: Author: N Brouard
499:
500: We had difficulties to get smoothed confidence intervals. It was due
501: to the period prevalence which wasn't computed accurately. The inner
502: parameter ftolpl is now an outer parameter of the .imach parameter
503: file after estepm. If ftolpl is small 1.e-4 and estepm too,
504: computation are long.
505:
1.209 brouard 506: Revision 1.208 2015/11/17 14:31:57 brouard
507: Summary: temporary
508:
1.208 brouard 509: Revision 1.207 2015/10/27 17:36:57 brouard
510: *** empty log message ***
511:
1.207 brouard 512: Revision 1.206 2015/10/24 07:14:11 brouard
513: *** empty log message ***
514:
1.206 brouard 515: Revision 1.205 2015/10/23 15:50:53 brouard
516: Summary: 0.98r3 some clarification for graphs on likelihood contributions
517:
1.205 brouard 518: Revision 1.204 2015/10/01 16:20:26 brouard
519: Summary: Some new graphs of contribution to likelihood
520:
1.204 brouard 521: Revision 1.203 2015/09/30 17:45:14 brouard
522: Summary: looking at better estimation of the hessian
523:
524: Also a better criteria for convergence to the period prevalence And
525: therefore adding the number of years needed to converge. (The
526: prevalence in any alive state shold sum to one
527:
1.203 brouard 528: Revision 1.202 2015/09/22 19:45:16 brouard
529: Summary: Adding some overall graph on contribution to likelihood. Might change
530:
1.202 brouard 531: Revision 1.201 2015/09/15 17:34:58 brouard
532: Summary: 0.98r0
533:
534: - Some new graphs like suvival functions
535: - Some bugs fixed like model=1+age+V2.
536:
1.201 brouard 537: Revision 1.200 2015/09/09 16:53:55 brouard
538: Summary: Big bug thanks to Flavia
539:
540: Even model=1+age+V2. did not work anymore
541:
1.200 brouard 542: Revision 1.199 2015/09/07 14:09:23 brouard
543: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
544:
1.199 brouard 545: Revision 1.198 2015/09/03 07:14:39 brouard
546: Summary: 0.98q5 Flavia
547:
1.198 brouard 548: Revision 1.197 2015/09/01 18:24:39 brouard
549: *** empty log message ***
550:
1.197 brouard 551: Revision 1.196 2015/08/18 23:17:52 brouard
552: Summary: 0.98q5
553:
1.196 brouard 554: Revision 1.195 2015/08/18 16:28:39 brouard
555: Summary: Adding a hack for testing purpose
556:
557: After reading the title, ftol and model lines, if the comment line has
558: a q, starting with #q, the answer at the end of the run is quit. It
559: permits to run test files in batch with ctest. The former workaround was
560: $ echo q | imach foo.imach
561:
1.195 brouard 562: Revision 1.194 2015/08/18 13:32:00 brouard
563: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
564:
1.194 brouard 565: Revision 1.193 2015/08/04 07:17:42 brouard
566: Summary: 0.98q4
567:
1.193 brouard 568: Revision 1.192 2015/07/16 16:49:02 brouard
569: Summary: Fixing some outputs
570:
1.192 brouard 571: Revision 1.191 2015/07/14 10:00:33 brouard
572: Summary: Some fixes
573:
1.191 brouard 574: Revision 1.190 2015/05/05 08:51:13 brouard
575: Summary: Adding digits in output parameters (7 digits instead of 6)
576:
577: Fix 1+age+.
578:
1.190 brouard 579: Revision 1.189 2015/04/30 14:45:16 brouard
580: Summary: 0.98q2
581:
1.189 brouard 582: Revision 1.188 2015/04/30 08:27:53 brouard
583: *** empty log message ***
584:
1.188 brouard 585: Revision 1.187 2015/04/29 09:11:15 brouard
586: *** empty log message ***
587:
1.187 brouard 588: Revision 1.186 2015/04/23 12:01:52 brouard
589: Summary: V1*age is working now, version 0.98q1
590:
591: Some codes had been disabled in order to simplify and Vn*age was
592: working in the optimization phase, ie, giving correct MLE parameters,
593: but, as usual, outputs were not correct and program core dumped.
594:
1.186 brouard 595: Revision 1.185 2015/03/11 13:26:42 brouard
596: Summary: Inclusion of compile and links command line for Intel Compiler
597:
1.185 brouard 598: Revision 1.184 2015/03/11 11:52:39 brouard
599: Summary: Back from Windows 8. Intel Compiler
600:
1.184 brouard 601: Revision 1.183 2015/03/10 20:34:32 brouard
602: Summary: 0.98q0, trying with directest, mnbrak fixed
603:
604: We use directest instead of original Powell test; probably no
605: incidence on the results, but better justifications;
606: We fixed Numerical Recipes mnbrak routine which was wrong and gave
607: wrong results.
608:
1.183 brouard 609: Revision 1.182 2015/02/12 08:19:57 brouard
610: Summary: Trying to keep directest which seems simpler and more general
611: Author: Nicolas Brouard
612:
1.182 brouard 613: Revision 1.181 2015/02/11 23:22:24 brouard
614: Summary: Comments on Powell added
615:
616: Author:
617:
1.181 brouard 618: Revision 1.180 2015/02/11 17:33:45 brouard
619: Summary: Finishing move from main to function (hpijx and prevalence_limit)
620:
1.180 brouard 621: Revision 1.179 2015/01/04 09:57:06 brouard
622: Summary: back to OS/X
623:
1.179 brouard 624: Revision 1.178 2015/01/04 09:35:48 brouard
625: *** empty log message ***
626:
1.178 brouard 627: Revision 1.177 2015/01/03 18:40:56 brouard
628: Summary: Still testing ilc32 on OSX
629:
1.177 brouard 630: Revision 1.176 2015/01/03 16:45:04 brouard
631: *** empty log message ***
632:
1.176 brouard 633: Revision 1.175 2015/01/03 16:33:42 brouard
634: *** empty log message ***
635:
1.175 brouard 636: Revision 1.174 2015/01/03 16:15:49 brouard
637: Summary: Still in cross-compilation
638:
1.174 brouard 639: Revision 1.173 2015/01/03 12:06:26 brouard
640: Summary: trying to detect cross-compilation
641:
1.173 brouard 642: Revision 1.172 2014/12/27 12:07:47 brouard
643: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
644:
1.172 brouard 645: Revision 1.171 2014/12/23 13:26:59 brouard
646: Summary: Back from Visual C
647:
648: Still problem with utsname.h on Windows
649:
1.171 brouard 650: Revision 1.170 2014/12/23 11:17:12 brouard
651: Summary: Cleaning some \%% back to %%
652:
653: The escape was mandatory for a specific compiler (which one?), but too many warnings.
654:
1.170 brouard 655: Revision 1.169 2014/12/22 23:08:31 brouard
656: Summary: 0.98p
657:
658: Outputs some informations on compiler used, OS etc. Testing on different platforms.
659:
1.169 brouard 660: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 661: Summary: update
1.169 brouard 662:
1.168 brouard 663: Revision 1.167 2014/12/22 13:50:56 brouard
664: Summary: Testing uname and compiler version and if compiled 32 or 64
665:
666: Testing on Linux 64
667:
1.167 brouard 668: Revision 1.166 2014/12/22 11:40:47 brouard
669: *** empty log message ***
670:
1.166 brouard 671: Revision 1.165 2014/12/16 11:20:36 brouard
672: Summary: After compiling on Visual C
673:
674: * imach.c (Module): Merging 1.61 to 1.162
675:
1.165 brouard 676: Revision 1.164 2014/12/16 10:52:11 brouard
677: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
678:
679: * imach.c (Module): Merging 1.61 to 1.162
680:
1.164 brouard 681: Revision 1.163 2014/12/16 10:30:11 brouard
682: * imach.c (Module): Merging 1.61 to 1.162
683:
1.163 brouard 684: Revision 1.162 2014/09/25 11:43:39 brouard
685: Summary: temporary backup 0.99!
686:
1.162 brouard 687: Revision 1.1 2014/09/16 11:06:58 brouard
688: Summary: With some code (wrong) for nlopt
689:
690: Author:
691:
692: Revision 1.161 2014/09/15 20:41:41 brouard
693: Summary: Problem with macro SQR on Intel compiler
694:
1.161 brouard 695: Revision 1.160 2014/09/02 09:24:05 brouard
696: *** empty log message ***
697:
1.160 brouard 698: Revision 1.159 2014/09/01 10:34:10 brouard
699: Summary: WIN32
700: Author: Brouard
701:
1.159 brouard 702: Revision 1.158 2014/08/27 17:11:51 brouard
703: *** empty log message ***
704:
1.158 brouard 705: Revision 1.157 2014/08/27 16:26:55 brouard
706: Summary: Preparing windows Visual studio version
707: Author: Brouard
708:
709: In order to compile on Visual studio, time.h is now correct and time_t
710: and tm struct should be used. difftime should be used but sometimes I
711: just make the differences in raw time format (time(&now).
712: Trying to suppress #ifdef LINUX
713: Add xdg-open for __linux in order to open default browser.
714:
1.157 brouard 715: Revision 1.156 2014/08/25 20:10:10 brouard
716: *** empty log message ***
717:
1.156 brouard 718: Revision 1.155 2014/08/25 18:32:34 brouard
719: Summary: New compile, minor changes
720: Author: Brouard
721:
1.155 brouard 722: Revision 1.154 2014/06/20 17:32:08 brouard
723: Summary: Outputs now all graphs of convergence to period prevalence
724:
1.154 brouard 725: Revision 1.153 2014/06/20 16:45:46 brouard
726: Summary: If 3 live state, convergence to period prevalence on same graph
727: Author: Brouard
728:
1.153 brouard 729: Revision 1.152 2014/06/18 17:54:09 brouard
730: Summary: open browser, use gnuplot on same dir than imach if not found in the path
731:
1.152 brouard 732: Revision 1.151 2014/06/18 16:43:30 brouard
733: *** empty log message ***
734:
1.151 brouard 735: Revision 1.150 2014/06/18 16:42:35 brouard
736: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
737: Author: brouard
738:
1.150 brouard 739: Revision 1.149 2014/06/18 15:51:14 brouard
740: Summary: Some fixes in parameter files errors
741: Author: Nicolas Brouard
742:
1.149 brouard 743: Revision 1.148 2014/06/17 17:38:48 brouard
744: Summary: Nothing new
745: Author: Brouard
746:
747: Just a new packaging for OS/X version 0.98nS
748:
1.148 brouard 749: Revision 1.147 2014/06/16 10:33:11 brouard
750: *** empty log message ***
751:
1.147 brouard 752: Revision 1.146 2014/06/16 10:20:28 brouard
753: Summary: Merge
754: Author: Brouard
755:
756: Merge, before building revised version.
757:
1.146 brouard 758: Revision 1.145 2014/06/10 21:23:15 brouard
759: Summary: Debugging with valgrind
760: Author: Nicolas Brouard
761:
762: Lot of changes in order to output the results with some covariates
763: After the Edimburgh REVES conference 2014, it seems mandatory to
764: improve the code.
765: No more memory valgrind error but a lot has to be done in order to
766: continue the work of splitting the code into subroutines.
767: Also, decodemodel has been improved. Tricode is still not
768: optimal. nbcode should be improved. Documentation has been added in
769: the source code.
770:
1.144 brouard 771: Revision 1.143 2014/01/26 09:45:38 brouard
772: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
773:
774: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
775: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
776:
1.143 brouard 777: Revision 1.142 2014/01/26 03:57:36 brouard
778: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
779:
780: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
781:
1.142 brouard 782: Revision 1.141 2014/01/26 02:42:01 brouard
783: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
784:
1.141 brouard 785: Revision 1.140 2011/09/02 10:37:54 brouard
786: Summary: times.h is ok with mingw32 now.
787:
1.140 brouard 788: Revision 1.139 2010/06/14 07:50:17 brouard
789: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
790: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
791:
1.139 brouard 792: Revision 1.138 2010/04/30 18:19:40 brouard
793: *** empty log message ***
794:
1.138 brouard 795: Revision 1.137 2010/04/29 18:11:38 brouard
796: (Module): Checking covariates for more complex models
797: than V1+V2. A lot of change to be done. Unstable.
798:
1.137 brouard 799: Revision 1.136 2010/04/26 20:30:53 brouard
800: (Module): merging some libgsl code. Fixing computation
801: of likelione (using inter/intrapolation if mle = 0) in order to
802: get same likelihood as if mle=1.
803: Some cleaning of code and comments added.
804:
1.136 brouard 805: Revision 1.135 2009/10/29 15:33:14 brouard
806: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
807:
1.135 brouard 808: Revision 1.134 2009/10/29 13:18:53 brouard
809: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
810:
1.134 brouard 811: Revision 1.133 2009/07/06 10:21:25 brouard
812: just nforces
813:
1.133 brouard 814: Revision 1.132 2009/07/06 08:22:05 brouard
815: Many tings
816:
1.132 brouard 817: Revision 1.131 2009/06/20 16:22:47 brouard
818: Some dimensions resccaled
819:
1.131 brouard 820: Revision 1.130 2009/05/26 06:44:34 brouard
821: (Module): Max Covariate is now set to 20 instead of 8. A
822: lot of cleaning with variables initialized to 0. Trying to make
823: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
824:
1.130 brouard 825: Revision 1.129 2007/08/31 13:49:27 lievre
826: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
827:
1.129 lievre 828: Revision 1.128 2006/06/30 13:02:05 brouard
829: (Module): Clarifications on computing e.j
830:
1.128 brouard 831: Revision 1.127 2006/04/28 18:11:50 brouard
832: (Module): Yes the sum of survivors was wrong since
833: imach-114 because nhstepm was no more computed in the age
834: loop. Now we define nhstepma in the age loop.
835: (Module): In order to speed up (in case of numerous covariates) we
836: compute health expectancies (without variances) in a first step
837: and then all the health expectancies with variances or standard
838: deviation (needs data from the Hessian matrices) which slows the
839: computation.
840: In the future we should be able to stop the program is only health
841: expectancies and graph are needed without standard deviations.
842:
1.127 brouard 843: Revision 1.126 2006/04/28 17:23:28 brouard
844: (Module): Yes the sum of survivors was wrong since
845: imach-114 because nhstepm was no more computed in the age
846: loop. Now we define nhstepma in the age loop.
847: Version 0.98h
848:
1.126 brouard 849: Revision 1.125 2006/04/04 15:20:31 lievre
850: Errors in calculation of health expectancies. Age was not initialized.
851: Forecasting file added.
852:
853: Revision 1.124 2006/03/22 17:13:53 lievre
854: Parameters are printed with %lf instead of %f (more numbers after the comma).
855: The log-likelihood is printed in the log file
856:
857: Revision 1.123 2006/03/20 10:52:43 brouard
858: * imach.c (Module): <title> changed, corresponds to .htm file
859: name. <head> headers where missing.
860:
861: * imach.c (Module): Weights can have a decimal point as for
862: English (a comma might work with a correct LC_NUMERIC environment,
863: otherwise the weight is truncated).
864: Modification of warning when the covariates values are not 0 or
865: 1.
866: Version 0.98g
867:
868: Revision 1.122 2006/03/20 09:45:41 brouard
869: (Module): Weights can have a decimal point as for
870: English (a comma might work with a correct LC_NUMERIC environment,
871: otherwise the weight is truncated).
872: Modification of warning when the covariates values are not 0 or
873: 1.
874: Version 0.98g
875:
876: Revision 1.121 2006/03/16 17:45:01 lievre
877: * imach.c (Module): Comments concerning covariates added
878:
879: * imach.c (Module): refinements in the computation of lli if
880: status=-2 in order to have more reliable computation if stepm is
881: not 1 month. Version 0.98f
882:
883: Revision 1.120 2006/03/16 15:10:38 lievre
884: (Module): refinements in the computation of lli if
885: status=-2 in order to have more reliable computation if stepm is
886: not 1 month. Version 0.98f
887:
888: Revision 1.119 2006/03/15 17:42:26 brouard
889: (Module): Bug if status = -2, the loglikelihood was
890: computed as likelihood omitting the logarithm. Version O.98e
891:
892: Revision 1.118 2006/03/14 18:20:07 brouard
893: (Module): varevsij Comments added explaining the second
894: table of variances if popbased=1 .
895: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
896: (Module): Function pstamp added
897: (Module): Version 0.98d
898:
899: Revision 1.117 2006/03/14 17:16:22 brouard
900: (Module): varevsij Comments added explaining the second
901: table of variances if popbased=1 .
902: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
903: (Module): Function pstamp added
904: (Module): Version 0.98d
905:
906: Revision 1.116 2006/03/06 10:29:27 brouard
907: (Module): Variance-covariance wrong links and
908: varian-covariance of ej. is needed (Saito).
909:
910: Revision 1.115 2006/02/27 12:17:45 brouard
911: (Module): One freematrix added in mlikeli! 0.98c
912:
913: Revision 1.114 2006/02/26 12:57:58 brouard
914: (Module): Some improvements in processing parameter
915: filename with strsep.
916:
917: Revision 1.113 2006/02/24 14:20:24 brouard
918: (Module): Memory leaks checks with valgrind and:
919: datafile was not closed, some imatrix were not freed and on matrix
920: allocation too.
921:
922: Revision 1.112 2006/01/30 09:55:26 brouard
923: (Module): Back to gnuplot.exe instead of wgnuplot.exe
924:
925: Revision 1.111 2006/01/25 20:38:18 brouard
926: (Module): Lots of cleaning and bugs added (Gompertz)
927: (Module): Comments can be added in data file. Missing date values
928: can be a simple dot '.'.
929:
930: Revision 1.110 2006/01/25 00:51:50 brouard
931: (Module): Lots of cleaning and bugs added (Gompertz)
932:
933: Revision 1.109 2006/01/24 19:37:15 brouard
934: (Module): Comments (lines starting with a #) are allowed in data.
935:
936: Revision 1.108 2006/01/19 18:05:42 lievre
937: Gnuplot problem appeared...
938: To be fixed
939:
940: Revision 1.107 2006/01/19 16:20:37 brouard
941: Test existence of gnuplot in imach path
942:
943: Revision 1.106 2006/01/19 13:24:36 brouard
944: Some cleaning and links added in html output
945:
946: Revision 1.105 2006/01/05 20:23:19 lievre
947: *** empty log message ***
948:
949: Revision 1.104 2005/09/30 16:11:43 lievre
950: (Module): sump fixed, loop imx fixed, and simplifications.
951: (Module): If the status is missing at the last wave but we know
952: that the person is alive, then we can code his/her status as -2
953: (instead of missing=-1 in earlier versions) and his/her
954: contributions to the likelihood is 1 - Prob of dying from last
955: health status (= 1-p13= p11+p12 in the easiest case of somebody in
956: the healthy state at last known wave). Version is 0.98
957:
958: Revision 1.103 2005/09/30 15:54:49 lievre
959: (Module): sump fixed, loop imx fixed, and simplifications.
960:
961: Revision 1.102 2004/09/15 17:31:30 brouard
962: Add the possibility to read data file including tab characters.
963:
964: Revision 1.101 2004/09/15 10:38:38 brouard
965: Fix on curr_time
966:
967: Revision 1.100 2004/07/12 18:29:06 brouard
968: Add version for Mac OS X. Just define UNIX in Makefile
969:
970: Revision 1.99 2004/06/05 08:57:40 brouard
971: *** empty log message ***
972:
973: Revision 1.98 2004/05/16 15:05:56 brouard
974: New version 0.97 . First attempt to estimate force of mortality
975: directly from the data i.e. without the need of knowing the health
976: state at each age, but using a Gompertz model: log u =a + b*age .
977: This is the basic analysis of mortality and should be done before any
978: other analysis, in order to test if the mortality estimated from the
979: cross-longitudinal survey is different from the mortality estimated
980: from other sources like vital statistic data.
981:
982: The same imach parameter file can be used but the option for mle should be -3.
983:
1.324 brouard 984: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 985: former routines in order to include the new code within the former code.
986:
987: The output is very simple: only an estimate of the intercept and of
988: the slope with 95% confident intervals.
989:
990: Current limitations:
991: A) Even if you enter covariates, i.e. with the
992: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
993: B) There is no computation of Life Expectancy nor Life Table.
994:
995: Revision 1.97 2004/02/20 13:25:42 lievre
996: Version 0.96d. Population forecasting command line is (temporarily)
997: suppressed.
998:
999: Revision 1.96 2003/07/15 15:38:55 brouard
1000: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
1001: rewritten within the same printf. Workaround: many printfs.
1002:
1003: Revision 1.95 2003/07/08 07:54:34 brouard
1004: * imach.c (Repository):
1005: (Repository): Using imachwizard code to output a more meaningful covariance
1006: matrix (cov(a12,c31) instead of numbers.
1007:
1008: Revision 1.94 2003/06/27 13:00:02 brouard
1009: Just cleaning
1010:
1011: Revision 1.93 2003/06/25 16:33:55 brouard
1012: (Module): On windows (cygwin) function asctime_r doesn't
1013: exist so I changed back to asctime which exists.
1014: (Module): Version 0.96b
1015:
1016: Revision 1.92 2003/06/25 16:30:45 brouard
1017: (Module): On windows (cygwin) function asctime_r doesn't
1018: exist so I changed back to asctime which exists.
1019:
1020: Revision 1.91 2003/06/25 15:30:29 brouard
1021: * imach.c (Repository): Duplicated warning errors corrected.
1022: (Repository): Elapsed time after each iteration is now output. It
1023: helps to forecast when convergence will be reached. Elapsed time
1024: is stamped in powell. We created a new html file for the graphs
1025: concerning matrix of covariance. It has extension -cov.htm.
1026:
1027: Revision 1.90 2003/06/24 12:34:15 brouard
1028: (Module): Some bugs corrected for windows. Also, when
1029: mle=-1 a template is output in file "or"mypar.txt with the design
1030: of the covariance matrix to be input.
1031:
1032: Revision 1.89 2003/06/24 12:30:52 brouard
1033: (Module): Some bugs corrected for windows. Also, when
1034: mle=-1 a template is output in file "or"mypar.txt with the design
1035: of the covariance matrix to be input.
1036:
1037: Revision 1.88 2003/06/23 17:54:56 brouard
1038: * 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.
1039:
1040: Revision 1.87 2003/06/18 12:26:01 brouard
1041: Version 0.96
1042:
1043: Revision 1.86 2003/06/17 20:04:08 brouard
1044: (Module): Change position of html and gnuplot routines and added
1045: routine fileappend.
1046:
1047: Revision 1.85 2003/06/17 13:12:43 brouard
1048: * imach.c (Repository): Check when date of death was earlier that
1049: current date of interview. It may happen when the death was just
1050: prior to the death. In this case, dh was negative and likelihood
1051: was wrong (infinity). We still send an "Error" but patch by
1052: assuming that the date of death was just one stepm after the
1053: interview.
1054: (Repository): Because some people have very long ID (first column)
1055: we changed int to long in num[] and we added a new lvector for
1056: memory allocation. But we also truncated to 8 characters (left
1057: truncation)
1058: (Repository): No more line truncation errors.
1059:
1060: Revision 1.84 2003/06/13 21:44:43 brouard
1061: * imach.c (Repository): Replace "freqsummary" at a correct
1062: place. It differs from routine "prevalence" which may be called
1063: many times. Probs is memory consuming and must be used with
1064: parcimony.
1065: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
1066:
1067: Revision 1.83 2003/06/10 13:39:11 lievre
1068: *** empty log message ***
1069:
1070: Revision 1.82 2003/06/05 15:57:20 brouard
1071: Add log in imach.c and fullversion number is now printed.
1072:
1073: */
1074: /*
1075: Interpolated Markov Chain
1076:
1077: Short summary of the programme:
1078:
1.227 brouard 1079: This program computes Healthy Life Expectancies or State-specific
1080: (if states aren't health statuses) Expectancies from
1081: cross-longitudinal data. Cross-longitudinal data consist in:
1082:
1083: -1- a first survey ("cross") where individuals from different ages
1084: are interviewed on their health status or degree of disability (in
1085: the case of a health survey which is our main interest)
1086:
1087: -2- at least a second wave of interviews ("longitudinal") which
1088: measure each change (if any) in individual health status. Health
1089: expectancies are computed from the time spent in each health state
1090: according to a model. More health states you consider, more time is
1091: necessary to reach the Maximum Likelihood of the parameters involved
1092: in the model. The simplest model is the multinomial logistic model
1093: where pij is the probability to be observed in state j at the second
1094: wave conditional to be observed in state i at the first
1095: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1096: etc , where 'age' is age and 'sex' is a covariate. If you want to
1097: have a more complex model than "constant and age", you should modify
1098: the program where the markup *Covariates have to be included here
1099: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1100: convergence.
1101:
1102: The advantage of this computer programme, compared to a simple
1103: multinomial logistic model, is clear when the delay between waves is not
1104: identical for each individual. Also, if a individual missed an
1105: intermediate interview, the information is lost, but taken into
1106: account using an interpolation or extrapolation.
1107:
1108: hPijx is the probability to be observed in state i at age x+h
1109: conditional to the observed state i at age x. The delay 'h' can be
1110: split into an exact number (nh*stepm) of unobserved intermediate
1111: states. This elementary transition (by month, quarter,
1112: semester or year) is modelled as a multinomial logistic. The hPx
1113: matrix is simply the matrix product of nh*stepm elementary matrices
1114: and the contribution of each individual to the likelihood is simply
1115: hPijx.
1116:
1117: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1118: of the life expectancies. It also computes the period (stable) prevalence.
1119:
1120: Back prevalence and projections:
1.227 brouard 1121:
1122: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1123: double agemaxpar, double ftolpl, int *ncvyearp, double
1124: dateprev1,double dateprev2, int firstpass, int lastpass, int
1125: mobilavproj)
1126:
1127: Computes the back prevalence limit for any combination of
1128: covariate values k at any age between ageminpar and agemaxpar and
1129: returns it in **bprlim. In the loops,
1130:
1131: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1132: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1133:
1134: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1135: Computes for any combination of covariates k and any age between bage and fage
1136: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1137: oldm=oldms;savm=savms;
1.227 brouard 1138:
1.267 brouard 1139: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1140: Computes the transition matrix starting at age 'age' over
1141: 'nhstepm*hstepm*stepm' months (i.e. until
1142: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1143: nhstepm*hstepm matrices.
1144:
1145: Returns p3mat[i][j][h] after calling
1146: p3mat[i][j][h]=matprod2(newm,
1147: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1148: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1149: oldm);
1.226 brouard 1150:
1151: Important routines
1152:
1153: - func (or funcone), computes logit (pij) distinguishing
1154: o fixed variables (single or product dummies or quantitative);
1155: o varying variables by:
1156: (1) wave (single, product dummies, quantitative),
1157: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1158: % fixed dummy (treated) or quantitative (not done because time-consuming);
1159: % varying dummy (not done) or quantitative (not done);
1160: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1161: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1162: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1163: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1164: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1165:
1.226 brouard 1166:
1167:
1.324 brouard 1168: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1169: Institut national d'études démographiques, Paris.
1.126 brouard 1170: This software have been partly granted by Euro-REVES, a concerted action
1171: from the European Union.
1172: It is copyrighted identically to a GNU software product, ie programme and
1173: software can be distributed freely for non commercial use. Latest version
1174: can be accessed at http://euroreves.ined.fr/imach .
1175:
1176: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1177: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1178:
1179: **********************************************************************/
1180: /*
1181: main
1182: read parameterfile
1183: read datafile
1184: concatwav
1185: freqsummary
1186: if (mle >= 1)
1187: mlikeli
1188: print results files
1189: if mle==1
1190: computes hessian
1191: read end of parameter file: agemin, agemax, bage, fage, estepm
1192: begin-prev-date,...
1193: open gnuplot file
1194: open html file
1.145 brouard 1195: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1196: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1197: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1198: freexexit2 possible for memory heap.
1199:
1200: h Pij x | pij_nom ficrestpij
1201: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1202: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1203: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1204:
1205: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1206: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1207: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1208: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1209: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1210:
1.126 brouard 1211: forecasting if prevfcast==1 prevforecast call prevalence()
1212: health expectancies
1213: Variance-covariance of DFLE
1214: prevalence()
1215: movingaverage()
1216: varevsij()
1217: if popbased==1 varevsij(,popbased)
1218: total life expectancies
1219: Variance of period (stable) prevalence
1220: end
1221: */
1222:
1.187 brouard 1223: /* #define DEBUG */
1224: /* #define DEBUGBRENT */
1.203 brouard 1225: /* #define DEBUGLINMIN */
1226: /* #define DEBUGHESS */
1227: #define DEBUGHESSIJ
1.224 brouard 1228: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1229: #define POWELL /* Instead of NLOPT */
1.224 brouard 1230: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1231: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1232: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1233: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1234:
1235: #include <math.h>
1236: #include <stdio.h>
1237: #include <stdlib.h>
1238: #include <string.h>
1.226 brouard 1239: #include <ctype.h>
1.159 brouard 1240:
1241: #ifdef _WIN32
1242: #include <io.h>
1.172 brouard 1243: #include <windows.h>
1244: #include <tchar.h>
1.159 brouard 1245: #else
1.126 brouard 1246: #include <unistd.h>
1.159 brouard 1247: #endif
1.126 brouard 1248:
1249: #include <limits.h>
1250: #include <sys/types.h>
1.171 brouard 1251:
1252: #if defined(__GNUC__)
1253: #include <sys/utsname.h> /* Doesn't work on Windows */
1254: #endif
1255:
1.126 brouard 1256: #include <sys/stat.h>
1257: #include <errno.h>
1.159 brouard 1258: /* extern int errno; */
1.126 brouard 1259:
1.157 brouard 1260: /* #ifdef LINUX */
1261: /* #include <time.h> */
1262: /* #include "timeval.h" */
1263: /* #else */
1264: /* #include <sys/time.h> */
1265: /* #endif */
1266:
1.126 brouard 1267: #include <time.h>
1268:
1.136 brouard 1269: #ifdef GSL
1270: #include <gsl/gsl_errno.h>
1271: #include <gsl/gsl_multimin.h>
1272: #endif
1273:
1.167 brouard 1274:
1.162 brouard 1275: #ifdef NLOPT
1276: #include <nlopt.h>
1277: typedef struct {
1278: double (* function)(double [] );
1279: } myfunc_data ;
1280: #endif
1281:
1.126 brouard 1282: /* #include <libintl.h> */
1283: /* #define _(String) gettext (String) */
1284:
1.251 brouard 1285: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1286:
1287: #define GNUPLOTPROGRAM "gnuplot"
1288: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1289: #define FILENAMELENGTH 256
1.126 brouard 1290:
1291: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1292: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1293:
1.144 brouard 1294: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1295: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1296:
1297: #define NINTERVMAX 8
1.144 brouard 1298: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1299: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1300: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1301: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1302: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1303: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1304: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1305: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1306: /* #define AGESUP 130 */
1.288 brouard 1307: /* #define AGESUP 150 */
1308: #define AGESUP 200
1.268 brouard 1309: #define AGEINF 0
1.218 brouard 1310: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1311: #define AGEBASE 40
1.194 brouard 1312: #define AGEOVERFLOW 1.e20
1.164 brouard 1313: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1314: #ifdef _WIN32
1315: #define DIRSEPARATOR '\\'
1316: #define CHARSEPARATOR "\\"
1317: #define ODIRSEPARATOR '/'
1318: #else
1.126 brouard 1319: #define DIRSEPARATOR '/'
1320: #define CHARSEPARATOR "/"
1321: #define ODIRSEPARATOR '\\'
1322: #endif
1323:
1.342 ! brouard 1324: /* $Id: imach.c,v 1.341 2022/09/11 07:58:42 brouard Exp $ */
1.126 brouard 1325: /* $State: Exp $ */
1.196 brouard 1326: #include "version.h"
1327: char version[]=__IMACH_VERSION__;
1.337 brouard 1328: 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";
1.342 ! brouard 1329: char fullversion[]="$Revision: 1.341 $ $Date: 2022/09/11 07:58:42 $";
1.126 brouard 1330: char strstart[80];
1331: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1332: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.342 ! brouard 1333: int debugILK=0; /* debugILK is set by a #d in a comment line */
1.187 brouard 1334: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 brouard 1335: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
1336: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
1.335 brouard 1337: 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 1338: 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 1339: int cptcovs=0; /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
1340: int cptcovsnq=0; /**< cptcovsnq number of SIMPLE covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1341: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1342: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.335 brouard 1343: 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 1344: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1345: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.339 brouard 1346: int ncovvt=0; /* Total number of effective (wave) varying covariates (dummy or quantitative or products [without age]) in the model */
1.232 brouard 1347: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1348: int nsd=0; /**< Total number of single dummy variables (output) */
1349: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1350: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1351: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1352: int ntveff=0; /**< ntveff number of effective time varying variables */
1353: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1354: int cptcov=0; /* Working variable */
1.334 brouard 1355: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs+1 declared globally ;*/
1.290 brouard 1356: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1357: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1358: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1359: int nlstate=2; /* Number of live states */
1360: int ndeath=1; /* Number of dead states */
1.130 brouard 1361: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.339 brouard 1362: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable*/
1363: int ncovcolt=0; /* ncovcolt=ncovcol+nqv+ntv+nqtv; total of covariates in the data, not in the model equation*/
1.126 brouard 1364: int popbased=0;
1365:
1366: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1367: int maxwav=0; /* Maxim number of waves */
1368: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1369: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1370: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1371: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1372: int mle=1, weightopt=0;
1.126 brouard 1373: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1374: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1375: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1376: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1377: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1378: int selected(int kvar); /* Is covariate kvar selected for printing results */
1379:
1.130 brouard 1380: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1381: double **matprod2(); /* test */
1.126 brouard 1382: double **oldm, **newm, **savm; /* Working pointers to matrices */
1383: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1384: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1385:
1.136 brouard 1386: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1387: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1388: FILE *ficlog, *ficrespow;
1.130 brouard 1389: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1390: double fretone; /* Only one call to likelihood */
1.130 brouard 1391: long ipmx=0; /* Number of contributions */
1.126 brouard 1392: double sw; /* Sum of weights */
1393: char filerespow[FILENAMELENGTH];
1394: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1395: FILE *ficresilk;
1396: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1397: FILE *ficresprobmorprev;
1398: FILE *fichtm, *fichtmcov; /* Html File */
1399: FILE *ficreseij;
1400: char filerese[FILENAMELENGTH];
1401: FILE *ficresstdeij;
1402: char fileresstde[FILENAMELENGTH];
1403: FILE *ficrescveij;
1404: char filerescve[FILENAMELENGTH];
1405: FILE *ficresvij;
1406: char fileresv[FILENAMELENGTH];
1.269 brouard 1407:
1.126 brouard 1408: char title[MAXLINE];
1.234 brouard 1409: char model[MAXLINE]; /**< The model line */
1.217 brouard 1410: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1411: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1412: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1413: char command[FILENAMELENGTH];
1414: int outcmd=0;
1415:
1.217 brouard 1416: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1417: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1418: char filelog[FILENAMELENGTH]; /* Log file */
1419: char filerest[FILENAMELENGTH];
1420: char fileregp[FILENAMELENGTH];
1421: char popfile[FILENAMELENGTH];
1422:
1423: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1424:
1.157 brouard 1425: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1426: /* struct timezone tzp; */
1427: /* extern int gettimeofday(); */
1428: struct tm tml, *gmtime(), *localtime();
1429:
1430: extern time_t time();
1431:
1432: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1433: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1434: struct tm tm;
1435:
1.126 brouard 1436: char strcurr[80], strfor[80];
1437:
1438: char *endptr;
1439: long lval;
1440: double dval;
1441:
1442: #define NR_END 1
1443: #define FREE_ARG char*
1444: #define FTOL 1.0e-10
1445:
1446: #define NRANSI
1.240 brouard 1447: #define ITMAX 200
1448: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1449:
1450: #define TOL 2.0e-4
1451:
1452: #define CGOLD 0.3819660
1453: #define ZEPS 1.0e-10
1454: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1455:
1456: #define GOLD 1.618034
1457: #define GLIMIT 100.0
1458: #define TINY 1.0e-20
1459:
1460: static double maxarg1,maxarg2;
1461: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1462: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1463:
1464: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1465: #define rint(a) floor(a+0.5)
1.166 brouard 1466: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1467: #define mytinydouble 1.0e-16
1.166 brouard 1468: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1469: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1470: /* static double dsqrarg; */
1471: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1472: static double sqrarg;
1473: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1474: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1475: int agegomp= AGEGOMP;
1476:
1477: int imx;
1478: int stepm=1;
1479: /* Stepm, step in month: minimum step interpolation*/
1480:
1481: int estepm;
1482: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1483:
1484: int m,nb;
1485: long *num;
1.197 brouard 1486: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1487: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1488: covariate for which somebody answered excluding
1489: undefined. Usually 2: 0 and 1. */
1490: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1491: covariate for which somebody answered including
1492: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1493: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1494: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1495: double ***mobaverage, ***mobaverages; /* New global variable */
1.332 brouard 1496: 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 1497: double *ageexmed,*agecens;
1498: double dateintmean=0;
1.296 brouard 1499: double anprojd, mprojd, jprojd; /* For eventual projections */
1500: double anprojf, mprojf, jprojf;
1.126 brouard 1501:
1.296 brouard 1502: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1503: double anbackf, mbackf, jbackf;
1504: double jintmean,mintmean,aintmean;
1.126 brouard 1505: double *weight;
1506: int **s; /* Status */
1.141 brouard 1507: double *agedc;
1.145 brouard 1508: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1509: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1510: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1511: double **coqvar; /* Fixed quantitative covariate nqv */
1.341 brouard 1512: double ***cotvar; /* Time varying covariate start at ncovcol + nqv + (1 to ntv) */
1.225 brouard 1513: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1514: double idx;
1515: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1516: /* Some documentation */
1517: /* Design original data
1518: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1519: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1520: * ntv=3 nqtv=1
1.330 brouard 1521: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1522: * For time varying covariate, quanti or dummies
1523: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1.341 brouard 1524: * cotvar[wav][ncovcol+nqv+ iv(1 to nqtv)][i]= [(1 to nqtv)][i]=(V12) quanti
1.319 brouard 1525: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1526: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1.332 brouard 1527: * covar[Vk,i], value of the Vkth fixed covariate dummy or quanti for individual i:
1.319 brouard 1528: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1529: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1530: * k= 1 2 3 4 5 6 7 8 9 10 11
1531: */
1532: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1533: /* 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
1534: # States 1=Coresidence, 2 Living alone, 3 Institution
1535: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1536: */
1.319 brouard 1537: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1538: /* k 1 2 3 4 5 6 7 8 9 */
1539: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1540: /* fixed or varying), 1 for age product, 2 for*/
1541: /* product */
1542: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1543: /*(single or product without age), 2 dummy*/
1544: /* with age product, 3 quant with age product*/
1545: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1546: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1.330 brouard 1547: /*TnsdVar[Tvar] 1 2 3 */
1.337 brouard 1548: /*Tvaraff[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1.319 brouard 1549: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1.338 brouard 1550: /*TvarsDind[nsd] 2 3 9 */ /* position K of single dummy cova */
1.319 brouard 1551: /* nsq 1 2 */ /* Counting single quantit tv */
1552: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1553: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1554: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1555: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1556: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1557: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 brouard 1558: /* 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 1559: /* 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 1560: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1561: /* Type */
1562: /* V 1 2 3 4 5 */
1563: /* F F V V V */
1564: /* D Q D D Q */
1565: /* */
1566: int *TvarsD;
1.330 brouard 1567: int *TnsdVar;
1.234 brouard 1568: int *TvarsDind;
1569: int *TvarsQ;
1570: int *TvarsQind;
1571:
1.318 brouard 1572: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1573: int nresult=0;
1.258 brouard 1574: int parameterline=0; /* # of the parameter (type) line */
1.334 brouard 1575: int TKresult[MAXRESULTLINESPONE]; /* TKresult[nres]=k for each resultline nres give the corresponding combination of dummies */
1576: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model corresponds to the k3 position in the resultline */
1577: int modelresult[MAXRESULTLINESPONE][NCOVMAX];/* modelresult[k3]=k1: k1th position in the model corresponds to the k3 position in the resultline */
1578: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1579: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1580: double TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.334 brouard 1581: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1582: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.318 brouard 1583: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1.332 brouard 1584: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.318 brouard 1585:
1586: /* 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
1587: # States 1=Coresidence, 2 Living alone, 3 Institution
1588: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1589: */
1.234 brouard 1590: /* 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 1591: 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 */
1592: 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 */
1593: 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 */
1594: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1595: 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 */
1596: 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 1597: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1598: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1599: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1600: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1601: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1602: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1603: 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 */
1604: 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 1605: int *TvarVV; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1606: int *TvarVVind; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1607: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
1608: /* model V1+V3+age*V1+age*V3+V1*V3 */
1609: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
1610: /* TvarVV={3,1,3}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1611: /* TvarVVind={2,5,5}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1.230 brouard 1612: int *Tvarsel; /**< Selected covariates for output */
1613: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1614: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1615: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1616: 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 1617: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1618: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1619: int *Tage;
1.227 brouard 1620: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1621: 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 1622: 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*/
1623: 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 1624: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1625: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1626: int **Tvard;
1.330 brouard 1627: int **Tvardk;
1.227 brouard 1628: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1629: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1630: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1631: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1632: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1633: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1634: double *lsurv, *lpop, *tpop;
1635:
1.231 brouard 1636: #define FD 1; /* Fixed dummy covariate */
1637: #define FQ 2; /* Fixed quantitative covariate */
1638: #define FP 3; /* Fixed product covariate */
1639: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1640: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1641: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1642: #define VD 10; /* Varying dummy covariate */
1643: #define VQ 11; /* Varying quantitative covariate */
1644: #define VP 12; /* Varying product covariate */
1645: #define VPDD 13; /* Varying product dummy*dummy covariate */
1646: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1647: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1648: #define APFD 16; /* Age product * fixed dummy covariate */
1649: #define APFQ 17; /* Age product * fixed quantitative covariate */
1650: #define APVD 18; /* Age product * varying dummy covariate */
1651: #define APVQ 19; /* Age product * varying quantitative covariate */
1652:
1653: #define FTYPE 1; /* Fixed covariate */
1654: #define VTYPE 2; /* Varying covariate (loop in wave) */
1655: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1656:
1657: struct kmodel{
1658: int maintype; /* main type */
1659: int subtype; /* subtype */
1660: };
1661: struct kmodel modell[NCOVMAX];
1662:
1.143 brouard 1663: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1664: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1665:
1666: /**************** split *************************/
1667: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1668: {
1669: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1670: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1671: */
1672: char *ss; /* pointer */
1.186 brouard 1673: int l1=0, l2=0; /* length counters */
1.126 brouard 1674:
1675: l1 = strlen(path ); /* length of path */
1676: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1677: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1678: if ( ss == NULL ) { /* no directory, so determine current directory */
1679: strcpy( name, path ); /* we got the fullname name because no directory */
1680: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1681: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1682: /* get current working directory */
1683: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1684: #ifdef WIN32
1685: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1686: #else
1687: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1688: #endif
1.126 brouard 1689: return( GLOCK_ERROR_GETCWD );
1690: }
1691: /* got dirc from getcwd*/
1692: printf(" DIRC = %s \n",dirc);
1.205 brouard 1693: } else { /* strip directory from path */
1.126 brouard 1694: ss++; /* after this, the filename */
1695: l2 = strlen( ss ); /* length of filename */
1696: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1697: strcpy( name, ss ); /* save file name */
1698: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1699: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1700: printf(" DIRC2 = %s \n",dirc);
1701: }
1702: /* We add a separator at the end of dirc if not exists */
1703: l1 = strlen( dirc ); /* length of directory */
1704: if( dirc[l1-1] != DIRSEPARATOR ){
1705: dirc[l1] = DIRSEPARATOR;
1706: dirc[l1+1] = 0;
1707: printf(" DIRC3 = %s \n",dirc);
1708: }
1709: ss = strrchr( name, '.' ); /* find last / */
1710: if (ss >0){
1711: ss++;
1712: strcpy(ext,ss); /* save extension */
1713: l1= strlen( name);
1714: l2= strlen(ss)+1;
1715: strncpy( finame, name, l1-l2);
1716: finame[l1-l2]= 0;
1717: }
1718:
1719: return( 0 ); /* we're done */
1720: }
1721:
1722:
1723: /******************************************/
1724:
1725: void replace_back_to_slash(char *s, char*t)
1726: {
1727: int i;
1728: int lg=0;
1729: i=0;
1730: lg=strlen(t);
1731: for(i=0; i<= lg; i++) {
1732: (s[i] = t[i]);
1733: if (t[i]== '\\') s[i]='/';
1734: }
1735: }
1736:
1.132 brouard 1737: char *trimbb(char *out, char *in)
1.137 brouard 1738: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1739: char *s;
1740: s=out;
1741: while (*in != '\0'){
1.137 brouard 1742: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1743: in++;
1744: }
1745: *out++ = *in++;
1746: }
1747: *out='\0';
1748: return s;
1749: }
1750:
1.187 brouard 1751: /* char *substrchaine(char *out, char *in, char *chain) */
1752: /* { */
1753: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1754: /* char *s, *t; */
1755: /* t=in;s=out; */
1756: /* while ((*in != *chain) && (*in != '\0')){ */
1757: /* *out++ = *in++; */
1758: /* } */
1759:
1760: /* /\* *in matches *chain *\/ */
1761: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1762: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1763: /* } */
1764: /* in--; chain--; */
1765: /* while ( (*in != '\0')){ */
1766: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1767: /* *out++ = *in++; */
1768: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1769: /* } */
1770: /* *out='\0'; */
1771: /* out=s; */
1772: /* return out; */
1773: /* } */
1774: char *substrchaine(char *out, char *in, char *chain)
1775: {
1776: /* Substract chain 'chain' from 'in', return and output 'out' */
1777: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1778:
1779: char *strloc;
1780:
1781: strcpy (out, in);
1782: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1783: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1784: if(strloc != NULL){
1785: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1786: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1787: /* strcpy (strloc, strloc +strlen(chain));*/
1788: }
1789: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1790: return out;
1791: }
1792:
1793:
1.145 brouard 1794: char *cutl(char *blocc, char *alocc, char *in, char occ)
1795: {
1.187 brouard 1796: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1797: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1798: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1799: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1800: */
1.160 brouard 1801: char *s, *t;
1.145 brouard 1802: t=in;s=in;
1803: while ((*in != occ) && (*in != '\0')){
1804: *alocc++ = *in++;
1805: }
1806: if( *in == occ){
1807: *(alocc)='\0';
1808: s=++in;
1809: }
1810:
1811: if (s == t) {/* occ not found */
1812: *(alocc-(in-s))='\0';
1813: in=s;
1814: }
1815: while ( *in != '\0'){
1816: *blocc++ = *in++;
1817: }
1818:
1819: *blocc='\0';
1820: return t;
1821: }
1.137 brouard 1822: char *cutv(char *blocc, char *alocc, char *in, char occ)
1823: {
1.187 brouard 1824: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1825: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1826: gives blocc="abcdef2ghi" and alocc="j".
1827: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1828: */
1829: char *s, *t;
1830: t=in;s=in;
1831: while (*in != '\0'){
1832: while( *in == occ){
1833: *blocc++ = *in++;
1834: s=in;
1835: }
1836: *blocc++ = *in++;
1837: }
1838: if (s == t) /* occ not found */
1839: *(blocc-(in-s))='\0';
1840: else
1841: *(blocc-(in-s)-1)='\0';
1842: in=s;
1843: while ( *in != '\0'){
1844: *alocc++ = *in++;
1845: }
1846:
1847: *alocc='\0';
1848: return s;
1849: }
1850:
1.126 brouard 1851: int nbocc(char *s, char occ)
1852: {
1853: int i,j=0;
1854: int lg=20;
1855: i=0;
1856: lg=strlen(s);
1857: for(i=0; i<= lg; i++) {
1.234 brouard 1858: if (s[i] == occ ) j++;
1.126 brouard 1859: }
1860: return j;
1861: }
1862:
1.137 brouard 1863: /* void cutv(char *u,char *v, char*t, char occ) */
1864: /* { */
1865: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1866: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1867: /* gives u="abcdef2ghi" and v="j" *\/ */
1868: /* int i,lg,j,p=0; */
1869: /* i=0; */
1870: /* lg=strlen(t); */
1871: /* for(j=0; j<=lg-1; j++) { */
1872: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1873: /* } */
1.126 brouard 1874:
1.137 brouard 1875: /* for(j=0; j<p; j++) { */
1876: /* (u[j] = t[j]); */
1877: /* } */
1878: /* u[p]='\0'; */
1.126 brouard 1879:
1.137 brouard 1880: /* for(j=0; j<= lg; j++) { */
1881: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1882: /* } */
1883: /* } */
1.126 brouard 1884:
1.160 brouard 1885: #ifdef _WIN32
1886: char * strsep(char **pp, const char *delim)
1887: {
1888: char *p, *q;
1889:
1890: if ((p = *pp) == NULL)
1891: return 0;
1892: if ((q = strpbrk (p, delim)) != NULL)
1893: {
1894: *pp = q + 1;
1895: *q = '\0';
1896: }
1897: else
1898: *pp = 0;
1899: return p;
1900: }
1901: #endif
1902:
1.126 brouard 1903: /********************** nrerror ********************/
1904:
1905: void nrerror(char error_text[])
1906: {
1907: fprintf(stderr,"ERREUR ...\n");
1908: fprintf(stderr,"%s\n",error_text);
1909: exit(EXIT_FAILURE);
1910: }
1911: /*********************** vector *******************/
1912: double *vector(int nl, int nh)
1913: {
1914: double *v;
1915: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1916: if (!v) nrerror("allocation failure in vector");
1917: return v-nl+NR_END;
1918: }
1919:
1920: /************************ free vector ******************/
1921: void free_vector(double*v, int nl, int nh)
1922: {
1923: free((FREE_ARG)(v+nl-NR_END));
1924: }
1925:
1926: /************************ivector *******************************/
1927: int *ivector(long nl,long nh)
1928: {
1929: int *v;
1930: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1931: if (!v) nrerror("allocation failure in ivector");
1932: return v-nl+NR_END;
1933: }
1934:
1935: /******************free ivector **************************/
1936: void free_ivector(int *v, long nl, long nh)
1937: {
1938: free((FREE_ARG)(v+nl-NR_END));
1939: }
1940:
1941: /************************lvector *******************************/
1942: long *lvector(long nl,long nh)
1943: {
1944: long *v;
1945: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1946: if (!v) nrerror("allocation failure in ivector");
1947: return v-nl+NR_END;
1948: }
1949:
1950: /******************free lvector **************************/
1951: void free_lvector(long *v, long nl, long nh)
1952: {
1953: free((FREE_ARG)(v+nl-NR_END));
1954: }
1955:
1956: /******************* imatrix *******************************/
1957: int **imatrix(long nrl, long nrh, long ncl, long nch)
1958: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1959: {
1960: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1961: int **m;
1962:
1963: /* allocate pointers to rows */
1964: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1965: if (!m) nrerror("allocation failure 1 in matrix()");
1966: m += NR_END;
1967: m -= nrl;
1968:
1969:
1970: /* allocate rows and set pointers to them */
1971: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1972: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1973: m[nrl] += NR_END;
1974: m[nrl] -= ncl;
1975:
1976: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1977:
1978: /* return pointer to array of pointers to rows */
1979: return m;
1980: }
1981:
1982: /****************** free_imatrix *************************/
1983: void free_imatrix(m,nrl,nrh,ncl,nch)
1984: int **m;
1985: long nch,ncl,nrh,nrl;
1986: /* free an int matrix allocated by imatrix() */
1987: {
1988: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1989: free((FREE_ARG) (m+nrl-NR_END));
1990: }
1991:
1992: /******************* matrix *******************************/
1993: double **matrix(long nrl, long nrh, long ncl, long nch)
1994: {
1995: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1996: double **m;
1997:
1998: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1999: if (!m) nrerror("allocation failure 1 in matrix()");
2000: m += NR_END;
2001: m -= nrl;
2002:
2003: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2004: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2005: m[nrl] += NR_END;
2006: m[nrl] -= ncl;
2007:
2008: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2009: return m;
1.145 brouard 2010: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
2011: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
2012: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 2013: */
2014: }
2015:
2016: /*************************free matrix ************************/
2017: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
2018: {
2019: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2020: free((FREE_ARG)(m+nrl-NR_END));
2021: }
2022:
2023: /******************* ma3x *******************************/
2024: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
2025: {
2026: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
2027: double ***m;
2028:
2029: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2030: if (!m) nrerror("allocation failure 1 in matrix()");
2031: m += NR_END;
2032: m -= nrl;
2033:
2034: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2035: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2036: m[nrl] += NR_END;
2037: m[nrl] -= ncl;
2038:
2039: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2040:
2041: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
2042: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
2043: m[nrl][ncl] += NR_END;
2044: m[nrl][ncl] -= nll;
2045: for (j=ncl+1; j<=nch; j++)
2046: m[nrl][j]=m[nrl][j-1]+nlay;
2047:
2048: for (i=nrl+1; i<=nrh; i++) {
2049: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
2050: for (j=ncl+1; j<=nch; j++)
2051: m[i][j]=m[i][j-1]+nlay;
2052: }
2053: return m;
2054: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
2055: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
2056: */
2057: }
2058:
2059: /*************************free ma3x ************************/
2060: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
2061: {
2062: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
2063: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2064: free((FREE_ARG)(m+nrl-NR_END));
2065: }
2066:
2067: /*************** function subdirf ***********/
2068: char *subdirf(char fileres[])
2069: {
2070: /* Caution optionfilefiname is hidden */
2071: strcpy(tmpout,optionfilefiname);
2072: strcat(tmpout,"/"); /* Add to the right */
2073: strcat(tmpout,fileres);
2074: return tmpout;
2075: }
2076:
2077: /*************** function subdirf2 ***********/
2078: char *subdirf2(char fileres[], char *preop)
2079: {
1.314 brouard 2080: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
2081: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 2082: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 2083: /* Caution optionfilefiname is hidden */
2084: strcpy(tmpout,optionfilefiname);
2085: strcat(tmpout,"/");
2086: strcat(tmpout,preop);
2087: strcat(tmpout,fileres);
2088: return tmpout;
2089: }
2090:
2091: /*************** function subdirf3 ***********/
2092: char *subdirf3(char fileres[], char *preop, char *preop2)
2093: {
2094:
2095: /* Caution optionfilefiname is hidden */
2096: strcpy(tmpout,optionfilefiname);
2097: strcat(tmpout,"/");
2098: strcat(tmpout,preop);
2099: strcat(tmpout,preop2);
2100: strcat(tmpout,fileres);
2101: return tmpout;
2102: }
1.213 brouard 2103:
2104: /*************** function subdirfext ***********/
2105: char *subdirfext(char fileres[], char *preop, char *postop)
2106: {
2107:
2108: strcpy(tmpout,preop);
2109: strcat(tmpout,fileres);
2110: strcat(tmpout,postop);
2111: return tmpout;
2112: }
1.126 brouard 2113:
1.213 brouard 2114: /*************** function subdirfext3 ***********/
2115: char *subdirfext3(char fileres[], char *preop, char *postop)
2116: {
2117:
2118: /* Caution optionfilefiname is hidden */
2119: strcpy(tmpout,optionfilefiname);
2120: strcat(tmpout,"/");
2121: strcat(tmpout,preop);
2122: strcat(tmpout,fileres);
2123: strcat(tmpout,postop);
2124: return tmpout;
2125: }
2126:
1.162 brouard 2127: char *asc_diff_time(long time_sec, char ascdiff[])
2128: {
2129: long sec_left, days, hours, minutes;
2130: days = (time_sec) / (60*60*24);
2131: sec_left = (time_sec) % (60*60*24);
2132: hours = (sec_left) / (60*60) ;
2133: sec_left = (sec_left) %(60*60);
2134: minutes = (sec_left) /60;
2135: sec_left = (sec_left) % (60);
2136: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2137: return ascdiff;
2138: }
2139:
1.126 brouard 2140: /***************** f1dim *************************/
2141: extern int ncom;
2142: extern double *pcom,*xicom;
2143: extern double (*nrfunc)(double []);
2144:
2145: double f1dim(double x)
2146: {
2147: int j;
2148: double f;
2149: double *xt;
2150:
2151: xt=vector(1,ncom);
2152: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2153: f=(*nrfunc)(xt);
2154: free_vector(xt,1,ncom);
2155: return f;
2156: }
2157:
2158: /*****************brent *************************/
2159: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2160: {
2161: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2162: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2163: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2164: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2165: * returned function value.
2166: */
1.126 brouard 2167: int iter;
2168: double a,b,d,etemp;
1.159 brouard 2169: double fu=0,fv,fw,fx;
1.164 brouard 2170: double ftemp=0.;
1.126 brouard 2171: double p,q,r,tol1,tol2,u,v,w,x,xm;
2172: double e=0.0;
2173:
2174: a=(ax < cx ? ax : cx);
2175: b=(ax > cx ? ax : cx);
2176: x=w=v=bx;
2177: fw=fv=fx=(*f)(x);
2178: for (iter=1;iter<=ITMAX;iter++) {
2179: xm=0.5*(a+b);
2180: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2181: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2182: printf(".");fflush(stdout);
2183: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2184: #ifdef DEBUGBRENT
1.126 brouard 2185: 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);
2186: 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);
2187: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2188: #endif
2189: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2190: *xmin=x;
2191: return fx;
2192: }
2193: ftemp=fu;
2194: if (fabs(e) > tol1) {
2195: r=(x-w)*(fx-fv);
2196: q=(x-v)*(fx-fw);
2197: p=(x-v)*q-(x-w)*r;
2198: q=2.0*(q-r);
2199: if (q > 0.0) p = -p;
2200: q=fabs(q);
2201: etemp=e;
2202: e=d;
2203: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2204: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2205: else {
1.224 brouard 2206: d=p/q;
2207: u=x+d;
2208: if (u-a < tol2 || b-u < tol2)
2209: d=SIGN(tol1,xm-x);
1.126 brouard 2210: }
2211: } else {
2212: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2213: }
2214: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2215: fu=(*f)(u);
2216: if (fu <= fx) {
2217: if (u >= x) a=x; else b=x;
2218: SHFT(v,w,x,u)
1.183 brouard 2219: SHFT(fv,fw,fx,fu)
2220: } else {
2221: if (u < x) a=u; else b=u;
2222: if (fu <= fw || w == x) {
1.224 brouard 2223: v=w;
2224: w=u;
2225: fv=fw;
2226: fw=fu;
1.183 brouard 2227: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2228: v=u;
2229: fv=fu;
1.183 brouard 2230: }
2231: }
1.126 brouard 2232: }
2233: nrerror("Too many iterations in brent");
2234: *xmin=x;
2235: return fx;
2236: }
2237:
2238: /****************** mnbrak ***********************/
2239:
2240: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2241: double (*func)(double))
1.183 brouard 2242: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2243: the downhill direction (defined by the function as evaluated at the initial points) and returns
2244: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2245: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2246: */
1.126 brouard 2247: double ulim,u,r,q, dum;
2248: double fu;
1.187 brouard 2249:
2250: double scale=10.;
2251: int iterscale=0;
2252:
2253: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2254: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2255:
2256:
2257: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2258: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2259: /* *bx = *ax - (*ax - *bx)/scale; */
2260: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2261: /* } */
2262:
1.126 brouard 2263: if (*fb > *fa) {
2264: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2265: SHFT(dum,*fb,*fa,dum)
2266: }
1.126 brouard 2267: *cx=(*bx)+GOLD*(*bx-*ax);
2268: *fc=(*func)(*cx);
1.183 brouard 2269: #ifdef DEBUG
1.224 brouard 2270: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2271: 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 2272: #endif
1.224 brouard 2273: 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 2274: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2275: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2276: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2277: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2278: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2279: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2280: fu=(*func)(u);
1.163 brouard 2281: #ifdef DEBUG
2282: /* f(x)=A(x-u)**2+f(u) */
2283: double A, fparabu;
2284: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2285: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2286: 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);
2287: 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 2288: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2289: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2290: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2291: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2292: #endif
1.184 brouard 2293: #ifdef MNBRAKORIGINAL
1.183 brouard 2294: #else
1.191 brouard 2295: /* if (fu > *fc) { */
2296: /* #ifdef DEBUG */
2297: /* printf("mnbrak4 fu > fc \n"); */
2298: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2299: /* #endif */
2300: /* /\* 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 *\\/ *\/ */
2301: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2302: /* dum=u; /\* Shifting c and u *\/ */
2303: /* u = *cx; */
2304: /* *cx = dum; */
2305: /* dum = fu; */
2306: /* fu = *fc; */
2307: /* *fc =dum; */
2308: /* } else { /\* end *\/ */
2309: /* #ifdef DEBUG */
2310: /* printf("mnbrak3 fu < fc \n"); */
2311: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2312: /* #endif */
2313: /* dum=u; /\* Shifting c and u *\/ */
2314: /* u = *cx; */
2315: /* *cx = dum; */
2316: /* dum = fu; */
2317: /* fu = *fc; */
2318: /* *fc =dum; */
2319: /* } */
1.224 brouard 2320: #ifdef DEBUGMNBRAK
2321: double A, fparabu;
2322: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2323: fparabu= *fa - A*(*ax-u)*(*ax-u);
2324: 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);
2325: 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 2326: #endif
1.191 brouard 2327: dum=u; /* Shifting c and u */
2328: u = *cx;
2329: *cx = dum;
2330: dum = fu;
2331: fu = *fc;
2332: *fc =dum;
1.183 brouard 2333: #endif
1.162 brouard 2334: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2335: #ifdef DEBUG
1.224 brouard 2336: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2337: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2338: #endif
1.126 brouard 2339: fu=(*func)(u);
2340: if (fu < *fc) {
1.183 brouard 2341: #ifdef DEBUG
1.224 brouard 2342: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2343: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2344: #endif
2345: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2346: SHFT(*fb,*fc,fu,(*func)(u))
2347: #ifdef DEBUG
2348: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2349: #endif
2350: }
1.162 brouard 2351: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2352: #ifdef DEBUG
1.224 brouard 2353: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2354: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2355: #endif
1.126 brouard 2356: u=ulim;
2357: fu=(*func)(u);
1.183 brouard 2358: } else { /* u could be left to b (if r > q parabola has a maximum) */
2359: #ifdef DEBUG
1.224 brouard 2360: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2361: 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 2362: #endif
1.126 brouard 2363: u=(*cx)+GOLD*(*cx-*bx);
2364: fu=(*func)(u);
1.224 brouard 2365: #ifdef DEBUG
2366: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2367: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2368: #endif
1.183 brouard 2369: } /* end tests */
1.126 brouard 2370: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2371: SHFT(*fa,*fb,*fc,fu)
2372: #ifdef DEBUG
1.224 brouard 2373: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2374: 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 2375: #endif
2376: } /* 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 2377: }
2378:
2379: /*************** linmin ************************/
1.162 brouard 2380: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2381: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2382: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2383: the value of func at the returned location p . This is actually all accomplished by calling the
2384: routines mnbrak and brent .*/
1.126 brouard 2385: int ncom;
2386: double *pcom,*xicom;
2387: double (*nrfunc)(double []);
2388:
1.224 brouard 2389: #ifdef LINMINORIGINAL
1.126 brouard 2390: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2391: #else
2392: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2393: #endif
1.126 brouard 2394: {
2395: double brent(double ax, double bx, double cx,
2396: double (*f)(double), double tol, double *xmin);
2397: double f1dim(double x);
2398: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2399: double *fc, double (*func)(double));
2400: int j;
2401: double xx,xmin,bx,ax;
2402: double fx,fb,fa;
1.187 brouard 2403:
1.203 brouard 2404: #ifdef LINMINORIGINAL
2405: #else
2406: double scale=10., axs, xxs; /* Scale added for infinity */
2407: #endif
2408:
1.126 brouard 2409: ncom=n;
2410: pcom=vector(1,n);
2411: xicom=vector(1,n);
2412: nrfunc=func;
2413: for (j=1;j<=n;j++) {
2414: pcom[j]=p[j];
1.202 brouard 2415: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2416: }
1.187 brouard 2417:
1.203 brouard 2418: #ifdef LINMINORIGINAL
2419: xx=1.;
2420: #else
2421: axs=0.0;
2422: xxs=1.;
2423: do{
2424: xx= xxs;
2425: #endif
1.187 brouard 2426: ax=0.;
2427: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2428: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2429: /* 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)) */
2430: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2431: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2432: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2433: /* 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 2434: #ifdef LINMINORIGINAL
2435: #else
2436: if (fx != fx){
1.224 brouard 2437: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2438: printf("|");
2439: fprintf(ficlog,"|");
1.203 brouard 2440: #ifdef DEBUGLINMIN
1.224 brouard 2441: 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 2442: #endif
2443: }
1.224 brouard 2444: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2445: #endif
2446:
1.191 brouard 2447: #ifdef DEBUGLINMIN
2448: 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 2449: 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 2450: #endif
1.224 brouard 2451: #ifdef LINMINORIGINAL
2452: #else
1.317 brouard 2453: if(fb == fx){ /* Flat function in the direction */
2454: xmin=xx;
1.224 brouard 2455: *flat=1;
1.317 brouard 2456: }else{
1.224 brouard 2457: *flat=0;
2458: #endif
2459: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2460: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2461: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2462: /* fmin = f(p[j] + xmin * xi[j]) */
2463: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2464: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2465: #ifdef DEBUG
1.224 brouard 2466: 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);
2467: 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);
2468: #endif
2469: #ifdef LINMINORIGINAL
2470: #else
2471: }
1.126 brouard 2472: #endif
1.191 brouard 2473: #ifdef DEBUGLINMIN
2474: printf("linmin end ");
1.202 brouard 2475: fprintf(ficlog,"linmin end ");
1.191 brouard 2476: #endif
1.126 brouard 2477: for (j=1;j<=n;j++) {
1.203 brouard 2478: #ifdef LINMINORIGINAL
2479: xi[j] *= xmin;
2480: #else
2481: #ifdef DEBUGLINMIN
2482: if(xxs <1.0)
2483: printf(" before xi[%d]=%12.8f", j,xi[j]);
2484: #endif
2485: 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) */
2486: #ifdef DEBUGLINMIN
2487: if(xxs <1.0)
2488: 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 );
2489: #endif
2490: #endif
1.187 brouard 2491: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2492: }
1.191 brouard 2493: #ifdef DEBUGLINMIN
1.203 brouard 2494: printf("\n");
1.191 brouard 2495: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2496: 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 2497: for (j=1;j<=n;j++) {
1.202 brouard 2498: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2499: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2500: if(j % ncovmodel == 0){
1.191 brouard 2501: printf("\n");
1.202 brouard 2502: fprintf(ficlog,"\n");
2503: }
1.191 brouard 2504: }
1.203 brouard 2505: #else
1.191 brouard 2506: #endif
1.126 brouard 2507: free_vector(xicom,1,n);
2508: free_vector(pcom,1,n);
2509: }
2510:
2511:
2512: /*************** powell ************************/
1.162 brouard 2513: /*
1.317 brouard 2514: Minimization of a function func of n variables. Input consists in an initial starting point
2515: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2516: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2517: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2518: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2519: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2520: */
1.224 brouard 2521: #ifdef LINMINORIGINAL
2522: #else
2523: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2524: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2525: #endif
1.126 brouard 2526: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2527: double (*func)(double []))
2528: {
1.224 brouard 2529: #ifdef LINMINORIGINAL
2530: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2531: double (*func)(double []));
1.224 brouard 2532: #else
1.241 brouard 2533: void linmin(double p[], double xi[], int n, double *fret,
2534: double (*func)(double []),int *flat);
1.224 brouard 2535: #endif
1.239 brouard 2536: int i,ibig,j,jk,k;
1.126 brouard 2537: double del,t,*pt,*ptt,*xit;
1.181 brouard 2538: double directest;
1.126 brouard 2539: double fp,fptt;
2540: double *xits;
2541: int niterf, itmp;
2542:
2543: pt=vector(1,n);
2544: ptt=vector(1,n);
2545: xit=vector(1,n);
2546: xits=vector(1,n);
2547: *fret=(*func)(p);
2548: for (j=1;j<=n;j++) pt[j]=p[j];
1.338 brouard 2549: rcurr_time = time(NULL);
2550: fp=(*fret); /* Initialisation */
1.126 brouard 2551: for (*iter=1;;++(*iter)) {
2552: ibig=0;
2553: del=0.0;
1.157 brouard 2554: rlast_time=rcurr_time;
2555: /* (void) gettimeofday(&curr_time,&tzp); */
2556: rcurr_time = time(NULL);
2557: curr_time = *localtime(&rcurr_time);
1.337 brouard 2558: /* 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); */
2559: /* 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); */
2560: 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);
2561: 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 2562: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2563: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2564: for (i=1;i<=n;i++) {
1.126 brouard 2565: fprintf(ficrespow," %.12lf", p[i]);
2566: }
1.239 brouard 2567: fprintf(ficrespow,"\n");fflush(ficrespow);
2568: printf("\n#model= 1 + age ");
2569: fprintf(ficlog,"\n#model= 1 + age ");
2570: if(nagesqr==1){
1.241 brouard 2571: printf(" + age*age ");
2572: fprintf(ficlog," + age*age ");
1.239 brouard 2573: }
2574: for(j=1;j <=ncovmodel-2;j++){
2575: if(Typevar[j]==0) {
2576: printf(" + V%d ",Tvar[j]);
2577: fprintf(ficlog," + V%d ",Tvar[j]);
2578: }else if(Typevar[j]==1) {
2579: printf(" + V%d*age ",Tvar[j]);
2580: fprintf(ficlog," + V%d*age ",Tvar[j]);
2581: }else if(Typevar[j]==2) {
2582: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2583: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2584: }
2585: }
1.126 brouard 2586: printf("\n");
1.239 brouard 2587: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2588: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2589: fprintf(ficlog,"\n");
1.239 brouard 2590: for(i=1,jk=1; i <=nlstate; i++){
2591: for(k=1; k <=(nlstate+ndeath); k++){
2592: if (k != i) {
2593: printf("%d%d ",i,k);
2594: fprintf(ficlog,"%d%d ",i,k);
2595: for(j=1; j <=ncovmodel; j++){
2596: printf("%12.7f ",p[jk]);
2597: fprintf(ficlog,"%12.7f ",p[jk]);
2598: jk++;
2599: }
2600: printf("\n");
2601: fprintf(ficlog,"\n");
2602: }
2603: }
2604: }
1.241 brouard 2605: if(*iter <=3 && *iter >1){
1.157 brouard 2606: tml = *localtime(&rcurr_time);
2607: strcpy(strcurr,asctime(&tml));
2608: rforecast_time=rcurr_time;
1.126 brouard 2609: itmp = strlen(strcurr);
2610: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2611: strcurr[itmp-1]='\0';
1.162 brouard 2612: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2613: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2614: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2615: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2616: forecast_time = *localtime(&rforecast_time);
2617: strcpy(strfor,asctime(&forecast_time));
2618: itmp = strlen(strfor);
2619: if(strfor[itmp-1]=='\n')
2620: strfor[itmp-1]='\0';
2621: 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);
2622: 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 2623: }
2624: }
1.187 brouard 2625: for (i=1;i<=n;i++) { /* For each direction i */
2626: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2627: fptt=(*fret);
2628: #ifdef DEBUG
1.203 brouard 2629: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2630: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2631: #endif
1.203 brouard 2632: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2633: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2634: #ifdef LINMINORIGINAL
1.188 brouard 2635: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2636: #else
2637: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2638: flatdir[i]=flat; /* Function is vanishing in that direction i */
2639: #endif
2640: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2641: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2642: /* because that direction will be replaced unless the gain del is small */
2643: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2644: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2645: /* with the new direction. */
2646: del=fabs(fptt-(*fret));
2647: ibig=i;
1.126 brouard 2648: }
2649: #ifdef DEBUG
2650: printf("%d %.12e",i,(*fret));
2651: fprintf(ficlog,"%d %.12e",i,(*fret));
2652: for (j=1;j<=n;j++) {
1.224 brouard 2653: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2654: printf(" x(%d)=%.12e",j,xit[j]);
2655: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2656: }
2657: for(j=1;j<=n;j++) {
1.225 brouard 2658: printf(" p(%d)=%.12e",j,p[j]);
2659: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2660: }
2661: printf("\n");
2662: fprintf(ficlog,"\n");
2663: #endif
1.187 brouard 2664: } /* end loop on each direction i */
2665: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2666: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2667: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2668: for(j=1;j<=n;j++) {
2669: if(flatdir[j] >0){
2670: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2671: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2672: }
1.319 brouard 2673: /* printf("\n"); */
2674: /* fprintf(ficlog,"\n"); */
2675: }
1.243 brouard 2676: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2677: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2678: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2679: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2680: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2681: /* decreased of more than 3.84 */
2682: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2683: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2684: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2685:
1.188 brouard 2686: /* Starting the program with initial values given by a former maximization will simply change */
2687: /* the scales of the directions and the directions, because the are reset to canonical directions */
2688: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2689: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2690: #ifdef DEBUG
2691: int k[2],l;
2692: k[0]=1;
2693: k[1]=-1;
2694: printf("Max: %.12e",(*func)(p));
2695: fprintf(ficlog,"Max: %.12e",(*func)(p));
2696: for (j=1;j<=n;j++) {
2697: printf(" %.12e",p[j]);
2698: fprintf(ficlog," %.12e",p[j]);
2699: }
2700: printf("\n");
2701: fprintf(ficlog,"\n");
2702: for(l=0;l<=1;l++) {
2703: for (j=1;j<=n;j++) {
2704: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2705: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2706: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2707: }
2708: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2709: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2710: }
2711: #endif
2712:
2713: free_vector(xit,1,n);
2714: free_vector(xits,1,n);
2715: free_vector(ptt,1,n);
2716: free_vector(pt,1,n);
2717: return;
1.192 brouard 2718: } /* enough precision */
1.240 brouard 2719: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2720: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2721: ptt[j]=2.0*p[j]-pt[j];
2722: xit[j]=p[j]-pt[j];
2723: pt[j]=p[j];
2724: }
1.181 brouard 2725: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2726: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2727: if (*iter <=4) {
1.225 brouard 2728: #else
2729: #endif
1.224 brouard 2730: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2731: #else
1.161 brouard 2732: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2733: #endif
1.162 brouard 2734: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2735: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2736: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2737: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2738: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2739: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2740: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2741: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2742: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2743: /* Even if f3 <f1, directest can be negative and t >0 */
2744: /* mu² and del² are equal when f3=f1 */
2745: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2746: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2747: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2748: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2749: #ifdef NRCORIGINAL
2750: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2751: #else
2752: 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 2753: t= t- del*SQR(fp-fptt);
1.183 brouard 2754: #endif
1.202 brouard 2755: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2756: #ifdef DEBUG
1.181 brouard 2757: 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);
2758: 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 2759: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2760: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2761: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2762: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2763: 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);
2764: 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);
2765: #endif
1.183 brouard 2766: #ifdef POWELLORIGINAL
2767: if (t < 0.0) { /* Then we use it for new direction */
2768: #else
1.182 brouard 2769: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2770: 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 2771: 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 2772: 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 2773: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2774: }
1.181 brouard 2775: if (directest < 0.0) { /* Then we use it for new direction */
2776: #endif
1.191 brouard 2777: #ifdef DEBUGLINMIN
1.234 brouard 2778: printf("Before linmin in direction P%d-P0\n",n);
2779: for (j=1;j<=n;j++) {
2780: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2781: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2782: if(j % ncovmodel == 0){
2783: printf("\n");
2784: fprintf(ficlog,"\n");
2785: }
2786: }
1.224 brouard 2787: #endif
2788: #ifdef LINMINORIGINAL
1.234 brouard 2789: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2790: #else
1.234 brouard 2791: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2792: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2793: #endif
1.234 brouard 2794:
1.191 brouard 2795: #ifdef DEBUGLINMIN
1.234 brouard 2796: for (j=1;j<=n;j++) {
2797: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2798: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2799: if(j % ncovmodel == 0){
2800: printf("\n");
2801: fprintf(ficlog,"\n");
2802: }
2803: }
1.224 brouard 2804: #endif
1.234 brouard 2805: for (j=1;j<=n;j++) {
2806: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2807: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2808: }
1.224 brouard 2809: #ifdef LINMINORIGINAL
2810: #else
1.234 brouard 2811: for (j=1, flatd=0;j<=n;j++) {
2812: if(flatdir[j]>0)
2813: flatd++;
2814: }
2815: if(flatd >0){
1.255 brouard 2816: printf("%d flat directions: ",flatd);
2817: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2818: for (j=1;j<=n;j++) {
2819: if(flatdir[j]>0){
2820: printf("%d ",j);
2821: fprintf(ficlog,"%d ",j);
2822: }
2823: }
2824: printf("\n");
2825: fprintf(ficlog,"\n");
1.319 brouard 2826: #ifdef FLATSUP
2827: free_vector(xit,1,n);
2828: free_vector(xits,1,n);
2829: free_vector(ptt,1,n);
2830: free_vector(pt,1,n);
2831: return;
2832: #endif
1.234 brouard 2833: }
1.191 brouard 2834: #endif
1.234 brouard 2835: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2836: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2837:
1.126 brouard 2838: #ifdef DEBUG
1.234 brouard 2839: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2840: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2841: for(j=1;j<=n;j++){
2842: printf(" %lf",xit[j]);
2843: fprintf(ficlog," %lf",xit[j]);
2844: }
2845: printf("\n");
2846: fprintf(ficlog,"\n");
1.126 brouard 2847: #endif
1.192 brouard 2848: } /* end of t or directest negative */
1.224 brouard 2849: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2850: #else
1.234 brouard 2851: } /* end if (fptt < fp) */
1.192 brouard 2852: #endif
1.225 brouard 2853: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2854: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2855: #else
1.224 brouard 2856: #endif
1.234 brouard 2857: } /* loop iteration */
1.126 brouard 2858: }
1.234 brouard 2859:
1.126 brouard 2860: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2861:
1.235 brouard 2862: 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 2863: {
1.338 brouard 2864: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij . Nicely done
1.279 brouard 2865: * (and selected quantitative values in nres)
2866: * by left multiplying the unit
2867: * matrix by transitions matrix until convergence is reached with precision ftolpl
2868: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2869: * Wx is row vector: population in state 1, population in state 2, population dead
2870: * or prevalence in state 1, prevalence in state 2, 0
2871: * newm is the matrix after multiplications, its rows are identical at a factor.
2872: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2873: * Output is prlim.
2874: * Initial matrix pimij
2875: */
1.206 brouard 2876: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2877: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2878: /* 0, 0 , 1} */
2879: /*
2880: * and after some iteration: */
2881: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2882: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2883: /* 0, 0 , 1} */
2884: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2885: /* {0.51571254859325999, 0.4842874514067399, */
2886: /* 0.51326036147820708, 0.48673963852179264} */
2887: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2888:
1.332 brouard 2889: int i, ii,j,k, k1;
1.209 brouard 2890: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2891: /* double **matprod2(); */ /* test */
1.218 brouard 2892: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2893: double **newm;
1.209 brouard 2894: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2895: int ncvloop=0;
1.288 brouard 2896: int first=0;
1.169 brouard 2897:
1.209 brouard 2898: min=vector(1,nlstate);
2899: max=vector(1,nlstate);
2900: meandiff=vector(1,nlstate);
2901:
1.218 brouard 2902: /* Starting with matrix unity */
1.126 brouard 2903: for (ii=1;ii<=nlstate+ndeath;ii++)
2904: for (j=1;j<=nlstate+ndeath;j++){
2905: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2906: }
1.169 brouard 2907:
2908: cov[1]=1.;
2909:
2910: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2911: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2912: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2913: ncvloop++;
1.126 brouard 2914: newm=savm;
2915: /* Covariates have to be included here again */
1.138 brouard 2916: cov[2]=agefin;
1.319 brouard 2917: if(nagesqr==1){
2918: cov[3]= agefin*agefin;
2919: }
1.332 brouard 2920: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
2921: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
2922: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
2923: if(Typevar[k1]==1){ /* A product with age */
2924: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
2925: }else{
2926: cov[2+nagesqr+k1]=precov[nres][k1];
2927: }
2928: }/* End of loop on model equation */
2929:
2930: /* Start of old code (replaced by a loop on position in the model equation */
2931: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only of the model *\/ */
2932: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
2933: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; *\/ */
2934: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])]; */
2935: /* /\* model = 1 +age + V1*V3 + age*V1 + V2 + V1 + age*V2 + V3 + V3*age + V1*V2 */
2936: /* * k 1 2 3 4 5 6 7 8 */
2937: /* *cov[] 1 2 3 4 5 6 7 8 9 10 */
2938: /* *TypeVar[k] 2 1 0 0 1 0 1 2 */
2939: /* *Dummy[k] 0 2 0 0 2 0 2 0 */
2940: /* *Tvar[k] 4 1 2 1 2 3 3 5 */
2941: /* *nsd=3 (1) (2) (3) */
2942: /* *TvarsD[nsd] [1]=2 1 3 */
2943: /* *TnsdVar [2]=2 [1]=1 [3]=3 */
2944: /* *TvarsDind[nsd](=k) [1]=3 [2]=4 [3]=6 */
2945: /* *Tage[] [1]=1 [2]=2 [3]=3 */
2946: /* *Tvard[] [1][1]=1 [2][1]=1 */
2947: /* * [1][2]=3 [2][2]=2 */
2948: /* *Tprod[](=k) [1]=1 [2]=8 */
2949: /* *TvarsDp(=Tvar) [1]=1 [2]=2 [3]=3 [4]=5 */
2950: /* *TvarD (=k) [1]=1 [2]=3 [3]=4 [3]=6 [4]=6 */
2951: /* *TvarsDpType */
2952: /* *si model= 1 + age + V3 + V2*age + V2 + V3*age */
2953: /* * nsd=1 (1) (2) */
2954: /* *TvarsD[nsd] 3 2 */
2955: /* *TnsdVar (3)=1 (2)=2 */
2956: /* *TvarsDind[nsd](=k) [1]=1 [2]=3 */
2957: /* *Tage[] [1]=2 [2]= 3 */
2958: /* *\/ */
2959: /* /\* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; *\/ */
2960: /* /\* 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)); *\/ */
2961: /* } */
2962: /* for (k=1; k<=nsq;k++) { /\* For single quantitative varying covariates only of the model *\/ */
2963: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
2964: /* /\* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline *\/ */
2965: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
2966: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][resultmodel[nres][k1]] */
2967: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2968: /* /\* 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]); *\/ */
2969: /* } */
2970: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
2971: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
2972: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
2973: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
2974: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
2975: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
2976: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2977: /* } */
2978: /* /\* 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]); *\/ */
2979: /* } */
2980: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
2981: /* /\* 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]); *\/ */
2982: /* if(Dummy[Tvard[k][1]]==0){ */
2983: /* if(Dummy[Tvard[k][2]]==0){ */
2984: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
2985: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
2986: /* }else{ */
2987: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
2988: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
2989: /* } */
2990: /* }else{ */
2991: /* if(Dummy[Tvard[k][2]]==0){ */
2992: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
2993: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
2994: /* }else{ */
2995: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
2996: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
2997: /* } */
2998: /* } */
2999: /* } /\* End product without age *\/ */
3000: /* ENd of old code */
1.138 brouard 3001: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3002: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3003: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 3004: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3005: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 3006: /* age and covariate values of ij are in 'cov' */
1.142 brouard 3007: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 3008:
1.126 brouard 3009: savm=oldm;
3010: oldm=newm;
1.209 brouard 3011:
3012: for(j=1; j<=nlstate; j++){
3013: max[j]=0.;
3014: min[j]=1.;
3015: }
3016: for(i=1;i<=nlstate;i++){
3017: sumnew=0;
3018: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
3019: for(j=1; j<=nlstate; j++){
3020: prlim[i][j]= newm[i][j]/(1-sumnew);
3021: max[j]=FMAX(max[j],prlim[i][j]);
3022: min[j]=FMIN(min[j],prlim[i][j]);
3023: }
3024: }
3025:
1.126 brouard 3026: maxmax=0.;
1.209 brouard 3027: for(j=1; j<=nlstate; j++){
3028: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
3029: maxmax=FMAX(maxmax,meandiff[j]);
3030: /* 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 3031: } /* j loop */
1.203 brouard 3032: *ncvyear= (int)age- (int)agefin;
1.208 brouard 3033: /* 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 3034: if(maxmax < ftolpl){
1.209 brouard 3035: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
3036: free_vector(min,1,nlstate);
3037: free_vector(max,1,nlstate);
3038: free_vector(meandiff,1,nlstate);
1.126 brouard 3039: return prlim;
3040: }
1.288 brouard 3041: } /* agefin loop */
1.208 brouard 3042: /* After some age loop it doesn't converge */
1.288 brouard 3043: if(!first){
3044: first=1;
3045: 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 3046: 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);
3047: }else if (first >=1 && first <10){
3048: 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);
3049: first++;
3050: }else if (first ==10){
3051: 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);
3052: 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");
3053: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
3054: first++;
1.288 brouard 3055: }
3056:
1.209 brouard 3057: /* 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); */
3058: free_vector(min,1,nlstate);
3059: free_vector(max,1,nlstate);
3060: free_vector(meandiff,1,nlstate);
1.208 brouard 3061:
1.169 brouard 3062: return prlim; /* should not reach here */
1.126 brouard 3063: }
3064:
1.217 brouard 3065:
3066: /**** Back Prevalence limit (stable or period prevalence) ****************/
3067:
1.218 brouard 3068: /* 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) */
3069: /* 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 3070: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 3071: {
1.264 brouard 3072: /* 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 3073: matrix by transitions matrix until convergence is reached with precision ftolpl */
3074: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
3075: /* Wx is row vector: population in state 1, population in state 2, population dead */
3076: /* or prevalence in state 1, prevalence in state 2, 0 */
3077: /* newm is the matrix after multiplications, its rows are identical at a factor */
3078: /* Initial matrix pimij */
3079: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
3080: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
3081: /* 0, 0 , 1} */
3082: /*
3083: * and after some iteration: */
3084: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
3085: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
3086: /* 0, 0 , 1} */
3087: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
3088: /* {0.51571254859325999, 0.4842874514067399, */
3089: /* 0.51326036147820708, 0.48673963852179264} */
3090: /* If we start from prlim again, prlim tends to a constant matrix */
3091:
1.332 brouard 3092: int i, ii,j,k, k1;
1.247 brouard 3093: int first=0;
1.217 brouard 3094: double *min, *max, *meandiff, maxmax,sumnew=0.;
3095: /* double **matprod2(); */ /* test */
3096: double **out, cov[NCOVMAX+1], **bmij();
3097: double **newm;
1.218 brouard 3098: double **dnewm, **doldm, **dsavm; /* for use */
3099: double **oldm, **savm; /* for use */
3100:
1.217 brouard 3101: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
3102: int ncvloop=0;
3103:
3104: min=vector(1,nlstate);
3105: max=vector(1,nlstate);
3106: meandiff=vector(1,nlstate);
3107:
1.266 brouard 3108: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
3109: oldm=oldms; savm=savms;
3110:
3111: /* Starting with matrix unity */
3112: for (ii=1;ii<=nlstate+ndeath;ii++)
3113: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 3114: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3115: }
3116:
3117: cov[1]=1.;
3118:
3119: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3120: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 3121: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 3122: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
3123: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 3124: ncvloop++;
1.218 brouard 3125: newm=savm; /* oldm should be kept from previous iteration or unity at start */
3126: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 3127: /* Covariates have to be included here again */
3128: cov[2]=agefin;
1.319 brouard 3129: if(nagesqr==1){
1.217 brouard 3130: cov[3]= agefin*agefin;;
1.319 brouard 3131: }
1.332 brouard 3132: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3133: if(Typevar[k1]==1){ /* A product with age */
3134: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.242 brouard 3135: }else{
1.332 brouard 3136: cov[2+nagesqr+k1]=precov[nres][k1];
1.242 brouard 3137: }
1.332 brouard 3138: }/* End of loop on model equation */
3139:
3140: /* Old code */
3141:
3142: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
3143: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
3144: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; */
3145: /* /\* 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)); *\/ */
3146: /* } */
3147: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
3148: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3149: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3150: /* /\* /\\* 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])]); *\\/ *\/ */
3151: /* /\* } *\/ */
3152: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3153: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3154: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3155: /* /\* 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]); *\/ */
3156: /* } */
3157: /* /\* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; *\/ */
3158: /* /\* for (k=1; k<=cptcovprod;k++) /\\* Useless *\\/ *\/ */
3159: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\\/ *\/ */
3160: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3161: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
3162: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ ERROR ???*\/ */
3163: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3164: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3165: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3166: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3167: /* } */
3168: /* /\* 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]); *\/ */
3169: /* } */
3170: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
3171: /* /\* 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]); *\/ */
3172: /* if(Dummy[Tvard[k][1]]==0){ */
3173: /* if(Dummy[Tvard[k][2]]==0){ */
3174: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3175: /* }else{ */
3176: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3177: /* } */
3178: /* }else{ */
3179: /* if(Dummy[Tvard[k][2]]==0){ */
3180: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3181: /* }else{ */
3182: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3183: /* } */
3184: /* } */
3185: /* } */
1.217 brouard 3186:
3187: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3188: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3189: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3190: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3191: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3192: /* ij should be linked to the correct index of cov */
3193: /* age and covariate values ij are in 'cov', but we need to pass
3194: * ij for the observed prevalence at age and status and covariate
3195: * number: prevacurrent[(int)agefin][ii][ij]
3196: */
3197: /* 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 *\/ */
3198: /* 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 *\/ */
3199: 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 3200: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3201: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3202: /* for(i=1; i<=nlstate+ndeath; i++) { */
3203: /* printf("%d newm= ",i); */
3204: /* for(j=1;j<=nlstate+ndeath;j++) { */
3205: /* printf("%f ",newm[i][j]); */
3206: /* } */
3207: /* printf("oldm * "); */
3208: /* for(j=1;j<=nlstate+ndeath;j++) { */
3209: /* printf("%f ",oldm[i][j]); */
3210: /* } */
1.268 brouard 3211: /* printf(" bmmij "); */
1.266 brouard 3212: /* for(j=1;j<=nlstate+ndeath;j++) { */
3213: /* printf("%f ",pmmij[i][j]); */
3214: /* } */
3215: /* printf("\n"); */
3216: /* } */
3217: /* } */
1.217 brouard 3218: savm=oldm;
3219: oldm=newm;
1.266 brouard 3220:
1.217 brouard 3221: for(j=1; j<=nlstate; j++){
3222: max[j]=0.;
3223: min[j]=1.;
3224: }
3225: for(j=1; j<=nlstate; j++){
3226: for(i=1;i<=nlstate;i++){
1.234 brouard 3227: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3228: bprlim[i][j]= newm[i][j];
3229: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3230: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3231: }
3232: }
1.218 brouard 3233:
1.217 brouard 3234: maxmax=0.;
3235: for(i=1; i<=nlstate; i++){
1.318 brouard 3236: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3237: maxmax=FMAX(maxmax,meandiff[i]);
3238: /* 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 3239: } /* i loop */
1.217 brouard 3240: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3241: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3242: if(maxmax < ftolpl){
1.220 brouard 3243: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3244: free_vector(min,1,nlstate);
3245: free_vector(max,1,nlstate);
3246: free_vector(meandiff,1,nlstate);
3247: return bprlim;
3248: }
1.288 brouard 3249: } /* agefin loop */
1.217 brouard 3250: /* After some age loop it doesn't converge */
1.288 brouard 3251: if(!first){
1.247 brouard 3252: first=1;
3253: 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\
3254: 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);
3255: }
3256: 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 3257: 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);
3258: /* 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); */
3259: free_vector(min,1,nlstate);
3260: free_vector(max,1,nlstate);
3261: free_vector(meandiff,1,nlstate);
3262:
3263: return bprlim; /* should not reach here */
3264: }
3265:
1.126 brouard 3266: /*************** transition probabilities ***************/
3267:
3268: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3269: {
1.138 brouard 3270: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3271: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3272: model to the ncovmodel covariates (including constant and age).
3273: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3274: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3275: ncth covariate in the global vector x is given by the formula:
3276: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3277: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3278: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3279: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3280: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3281: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3282: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3283: */
3284: double s1, lnpijopii;
1.126 brouard 3285: /*double t34;*/
1.164 brouard 3286: int i,j, nc, ii, jj;
1.126 brouard 3287:
1.223 brouard 3288: for(i=1; i<= nlstate; i++){
3289: for(j=1; j<i;j++){
3290: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3291: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3292: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3293: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3294: }
3295: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3296: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3297: }
3298: for(j=i+1; j<=nlstate+ndeath;j++){
3299: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3300: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3301: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3302: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3303: }
3304: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3305: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3306: }
3307: }
1.218 brouard 3308:
1.223 brouard 3309: for(i=1; i<= nlstate; i++){
3310: s1=0;
3311: for(j=1; j<i; j++){
1.339 brouard 3312: /* printf("debug1 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 3313: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3314: }
3315: for(j=i+1; j<=nlstate+ndeath; j++){
1.339 brouard 3316: /* printf("debug2 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 3317: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3318: }
3319: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3320: ps[i][i]=1./(s1+1.);
3321: /* Computing other pijs */
3322: for(j=1; j<i; j++)
1.325 brouard 3323: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3324: for(j=i+1; j<=nlstate+ndeath; j++)
3325: ps[i][j]= exp(ps[i][j])*ps[i][i];
3326: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3327: } /* end i */
1.218 brouard 3328:
1.223 brouard 3329: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3330: for(jj=1; jj<= nlstate+ndeath; jj++){
3331: ps[ii][jj]=0;
3332: ps[ii][ii]=1;
3333: }
3334: }
1.294 brouard 3335:
3336:
1.223 brouard 3337: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3338: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3339: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3340: /* } */
3341: /* printf("\n "); */
3342: /* } */
3343: /* printf("\n ");printf("%lf ",cov[2]);*/
3344: /*
3345: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3346: goto end;*/
1.266 brouard 3347: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3348: }
3349:
1.218 brouard 3350: /*************** backward transition probabilities ***************/
3351:
3352: /* 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 ) */
3353: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3354: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3355: {
1.302 brouard 3356: /* 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 3357: * 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 3358: */
1.218 brouard 3359: int i, ii, j,k;
1.222 brouard 3360:
3361: double **out, **pmij();
3362: double sumnew=0.;
1.218 brouard 3363: double agefin;
1.292 brouard 3364: 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 3365: double **dnewm, **dsavm, **doldm;
3366: double **bbmij;
3367:
1.218 brouard 3368: doldm=ddoldms; /* global pointers */
1.222 brouard 3369: dnewm=ddnewms;
3370: dsavm=ddsavms;
1.318 brouard 3371:
3372: /* Debug */
3373: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3374: agefin=cov[2];
1.268 brouard 3375: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3376: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3377: the observed prevalence (with this covariate ij) at beginning of transition */
3378: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3379:
3380: /* P_x */
1.325 brouard 3381: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3382: /* outputs pmmij which is a stochastic matrix in row */
3383:
3384: /* Diag(w_x) */
1.292 brouard 3385: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3386: sumnew=0.;
1.269 brouard 3387: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3388: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3389: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3390: sumnew+=prevacurrent[(int)agefin][ii][ij];
3391: }
3392: if(sumnew >0.01){ /* At least some value in the prevalence */
3393: for (ii=1;ii<=nlstate+ndeath;ii++){
3394: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3395: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3396: }
3397: }else{
3398: for (ii=1;ii<=nlstate+ndeath;ii++){
3399: for (j=1;j<=nlstate+ndeath;j++)
3400: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3401: }
3402: /* if(sumnew <0.9){ */
3403: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3404: /* } */
3405: }
3406: k3=0.0; /* We put the last diagonal to 0 */
3407: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3408: doldm[ii][ii]= k3;
3409: }
3410: /* End doldm, At the end doldm is diag[(w_i)] */
3411:
1.292 brouard 3412: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3413: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3414:
1.292 brouard 3415: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3416: /* 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 3417: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3418: sumnew=0.;
1.222 brouard 3419: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3420: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3421: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3422: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3423: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3424: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3425: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3426: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3427: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3428: /* }else */
1.268 brouard 3429: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3430: } /*End ii */
3431: } /* 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 */
3432:
1.292 brouard 3433: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3434: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3435: /* end bmij */
1.266 brouard 3436: return ps; /*pointer is unchanged */
1.218 brouard 3437: }
1.217 brouard 3438: /*************** transition probabilities ***************/
3439:
1.218 brouard 3440: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3441: {
3442: /* According to parameters values stored in x and the covariate's values stored in cov,
3443: computes the probability to be observed in state j being in state i by appying the
3444: model to the ncovmodel covariates (including constant and age).
3445: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3446: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3447: ncth covariate in the global vector x is given by the formula:
3448: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3449: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3450: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3451: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3452: Outputs ps[i][j] the probability to be observed in j being in j according to
3453: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3454: */
3455: double s1, lnpijopii;
3456: /*double t34;*/
3457: int i,j, nc, ii, jj;
3458:
1.234 brouard 3459: for(i=1; i<= nlstate; i++){
3460: for(j=1; j<i;j++){
3461: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3462: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3463: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3464: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3465: }
3466: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3467: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3468: }
3469: for(j=i+1; j<=nlstate+ndeath;j++){
3470: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3471: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3472: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3473: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3474: }
3475: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3476: }
3477: }
3478:
3479: for(i=1; i<= nlstate; i++){
3480: s1=0;
3481: for(j=1; j<i; j++){
3482: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3483: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3484: }
3485: for(j=i+1; j<=nlstate+ndeath; j++){
3486: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3487: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3488: }
3489: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3490: ps[i][i]=1./(s1+1.);
3491: /* Computing other pijs */
3492: for(j=1; j<i; j++)
3493: ps[i][j]= exp(ps[i][j])*ps[i][i];
3494: for(j=i+1; j<=nlstate+ndeath; j++)
3495: ps[i][j]= exp(ps[i][j])*ps[i][i];
3496: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3497: } /* end i */
3498:
3499: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3500: for(jj=1; jj<= nlstate+ndeath; jj++){
3501: ps[ii][jj]=0;
3502: ps[ii][ii]=1;
3503: }
3504: }
1.296 brouard 3505: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3506: for(jj=1; jj<= nlstate+ndeath; jj++){
3507: s1=0.;
3508: for(ii=1; ii<= nlstate+ndeath; ii++){
3509: s1+=ps[ii][jj];
3510: }
3511: for(ii=1; ii<= nlstate; ii++){
3512: ps[ii][jj]=ps[ii][jj]/s1;
3513: }
3514: }
3515: /* Transposition */
3516: for(jj=1; jj<= nlstate+ndeath; jj++){
3517: for(ii=jj; ii<= nlstate+ndeath; ii++){
3518: s1=ps[ii][jj];
3519: ps[ii][jj]=ps[jj][ii];
3520: ps[jj][ii]=s1;
3521: }
3522: }
3523: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3524: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3525: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3526: /* } */
3527: /* printf("\n "); */
3528: /* } */
3529: /* printf("\n ");printf("%lf ",cov[2]);*/
3530: /*
3531: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3532: goto end;*/
3533: return ps;
1.217 brouard 3534: }
3535:
3536:
1.126 brouard 3537: /**************** Product of 2 matrices ******************/
3538:
1.145 brouard 3539: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3540: {
3541: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3542: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3543: /* in, b, out are matrice of pointers which should have been initialized
3544: before: only the contents of out is modified. The function returns
3545: a pointer to pointers identical to out */
1.145 brouard 3546: int i, j, k;
1.126 brouard 3547: for(i=nrl; i<= nrh; i++)
1.145 brouard 3548: for(k=ncolol; k<=ncoloh; k++){
3549: out[i][k]=0.;
3550: for(j=ncl; j<=nch; j++)
3551: out[i][k] +=in[i][j]*b[j][k];
3552: }
1.126 brouard 3553: return out;
3554: }
3555:
3556:
3557: /************* Higher Matrix Product ***************/
3558:
1.235 brouard 3559: 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 3560: {
1.336 brouard 3561: /* Already optimized with precov.
3562: 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 3563: 'nhstepm*hstepm*stepm' months (i.e. until
3564: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3565: nhstepm*hstepm matrices.
3566: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3567: (typically every 2 years instead of every month which is too big
3568: for the memory).
3569: Model is determined by parameters x and covariates have to be
3570: included manually here.
3571:
3572: */
3573:
1.330 brouard 3574: int i, j, d, h, k, k1;
1.131 brouard 3575: double **out, cov[NCOVMAX+1];
1.126 brouard 3576: double **newm;
1.187 brouard 3577: double agexact;
1.214 brouard 3578: double agebegin, ageend;
1.126 brouard 3579:
3580: /* Hstepm could be zero and should return the unit matrix */
3581: for (i=1;i<=nlstate+ndeath;i++)
3582: for (j=1;j<=nlstate+ndeath;j++){
3583: oldm[i][j]=(i==j ? 1.0 : 0.0);
3584: po[i][j][0]=(i==j ? 1.0 : 0.0);
3585: }
3586: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3587: for(h=1; h <=nhstepm; h++){
3588: for(d=1; d <=hstepm; d++){
3589: newm=savm;
3590: /* Covariates have to be included here again */
3591: cov[1]=1.;
1.214 brouard 3592: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3593: cov[2]=agexact;
1.319 brouard 3594: if(nagesqr==1){
1.227 brouard 3595: cov[3]= agexact*agexact;
1.319 brouard 3596: }
1.330 brouard 3597: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
3598: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
3599: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.332 brouard 3600: if(Typevar[k1]==1){ /* A product with age */
3601: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
3602: }else{
3603: cov[2+nagesqr+k1]=precov[nres][k1];
3604: }
3605: }/* End of loop on model equation */
3606: /* Old code */
3607: /* if( Dummy[k1]==0 && Typevar[k1]==0 ){ /\* Single dummy *\/ */
3608: /* /\* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) *\/ */
3609: /* /\* for (k=1; k<=nsd;k++) { /\\* For single dummy covariates only *\\/ *\/ */
3610: /* /\* /\\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\\/ *\/ */
3611: /* /\* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 *\/ */
3612: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3613: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3614: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
3615: /* /\* nsd 1 2 3 *\/ /\* Counting single dummies covar fixed or tv *\/ */
3616: /* /\*TvarsD[nsd] 4 3 1 *\/ /\* ID of single dummy cova fixed or timevary*\/ */
3617: /* /\*TvarsDind[k] 2 3 9 *\/ /\* position K of single dummy cova *\/ */
3618: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] *\/ */
3619: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3620: /* /\* 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]])); *\/ */
3621: /* 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); */
3622: /* printf("hpxij new Dummy precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3623: /* }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /\* Single quantitative variables *\/ */
3624: /* /\* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline *\/ */
3625: /* cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]]; */
3626: /* /\* for (k=1; k<=nsq;k++) { /\\* For single varying covariates only *\\/ *\/ */
3627: /* /\* /\\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\\/ *\/ */
3628: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
3629: /* 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]]); */
3630: /* printf("hpxij new Quanti precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3631: /* }else if( Dummy[k1]==2 ){ /\* For dummy with age product *\/ */
3632: /* /\* Tvar[k1] Variable in the age product age*V1 is 1 *\/ */
3633: /* /\* [Tinvresult[nres][V1] is its value in the resultline nres *\/ */
3634: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvar[k1]]*cov[2]; */
3635: /* 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]); */
3636: /* printf("hpxij new Dummy with age product precov[nres=%d][k1=%d]=%.4f * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3637:
3638: /* /\* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; *\/ */
3639: /* /\* for (k=1; k<=cptcovage;k++){ /\\* For product with age V1+V1*age +V4 +age*V3 *\\/ *\/ */
3640: /* /\* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*\/ */
3641: /* /\* *\/ */
1.330 brouard 3642: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3643: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3644: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
1.332 brouard 3645: /* /\*cptcovage=2 1 2 *\/ */
3646: /* /\*Tage[k]= 5 8 *\/ */
3647: /* }else if( Dummy[k1]==3 ){ /\* For quant with age product *\/ */
3648: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3649: /* 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]]); */
3650: /* printf("hpxij new Quanti with age product precov[nres=%d][k1=%d] * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3651: /* /\* if(Dummy[Tage[k]]== 2){ /\\* dummy with age *\\/ *\/ */
3652: /* /\* /\\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\\* dummy with age *\\\/ *\\/ *\/ */
3653: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\\/ *\/ */
3654: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\\/ *\/ */
3655: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
3656: /* /\* 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); *\/ */
3657: /* /\* } else if(Dummy[Tage[k]]== 3){ /\\* quantitative with age *\\/ *\/ */
3658: /* /\* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; *\/ */
3659: /* /\* } *\/ */
3660: /* /\* 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]); *\/ */
3661: /* }else if(Typevar[k1]==2 ){ /\* For product (not with age) *\/ */
3662: /* /\* for (k=1; k<=cptcovprod;k++){ /\\* For product without age *\\/ *\/ */
3663: /* /\* /\\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\\/ *\/ */
3664: /* /\* /\\* k 1 2 3 4 5 6 7 8 9 *\\/ *\/ */
3665: /* /\* /\\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\\/ *\/ */
3666: /* /\* /\\*cptcovprod=1 1 2 *\\/ *\/ */
3667: /* /\* /\\*Tprod[]= 4 7 *\\/ *\/ */
3668: /* /\* /\\*Tvard[][1] 4 1 *\\/ *\/ */
3669: /* /\* /\\*Tvard[][2] 3 2 *\\/ *\/ */
1.330 brouard 3670:
1.332 brouard 3671: /* /\* 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])]); *\/ */
3672: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3673: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]]; */
3674: /* 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]]); */
3675: /* printf("hpxij new Product no age product precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3676:
3677: /* /\* if(Dummy[Tvardk[k1][1]]==0){ *\/ */
3678: /* /\* if(Dummy[Tvardk[k1][2]]==0){ /\\* Product of dummies *\\/ *\/ */
3679: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3680: /* /\* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; *\/ */
3681: /* /\* 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]])]; *\/ */
3682: /* /\* }else{ /\\* Product of dummy by quantitative *\\/ *\/ */
3683: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; *\/ */
3684: /* /\* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; *\/ */
3685: /* /\* } *\/ */
3686: /* /\* }else{ /\\* Product of quantitative by...*\\/ *\/ */
3687: /* /\* if(Dummy[Tvard[k][2]]==0){ /\\* quant by dummy *\\/ *\/ */
3688: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\\/ *\/ */
3689: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3690: /* /\* }else{ /\\* Product of two quant *\\/ *\/ */
3691: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\\/ *\/ */
3692: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3693: /* /\* } *\/ */
3694: /* /\* }/\\*end of products quantitative *\\/ *\/ */
3695: /* }/\*end of products *\/ */
3696: /* } /\* End of loop on model equation *\/ */
1.235 brouard 3697: /* for (k=1; k<=cptcovn;k++) */
3698: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3699: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3700: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3701: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3702: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3703:
3704:
1.126 brouard 3705: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3706: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3707: /* right multiplication of oldm by the current matrix */
1.126 brouard 3708: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3709: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3710: /* if((int)age == 70){ */
3711: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3712: /* for(i=1; i<=nlstate+ndeath; i++) { */
3713: /* printf("%d pmmij ",i); */
3714: /* for(j=1;j<=nlstate+ndeath;j++) { */
3715: /* printf("%f ",pmmij[i][j]); */
3716: /* } */
3717: /* printf(" oldm "); */
3718: /* for(j=1;j<=nlstate+ndeath;j++) { */
3719: /* printf("%f ",oldm[i][j]); */
3720: /* } */
3721: /* printf("\n"); */
3722: /* } */
3723: /* } */
1.126 brouard 3724: savm=oldm;
3725: oldm=newm;
3726: }
3727: for(i=1; i<=nlstate+ndeath; i++)
3728: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3729: po[i][j][h]=newm[i][j];
3730: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3731: }
1.128 brouard 3732: /*printf("h=%d ",h);*/
1.126 brouard 3733: } /* end h */
1.267 brouard 3734: /* printf("\n H=%d \n",h); */
1.126 brouard 3735: return po;
3736: }
3737:
1.217 brouard 3738: /************* Higher Back Matrix Product ***************/
1.218 brouard 3739: /* 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 3740: 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 3741: {
1.332 brouard 3742: /* For dummy covariates given in each resultline (for historical, computes the corresponding combination ij),
3743: computes the transition matrix starting at age 'age' over
1.217 brouard 3744: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3745: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3746: nhstepm*hstepm matrices.
3747: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3748: (typically every 2 years instead of every month which is too big
1.217 brouard 3749: for the memory).
1.218 brouard 3750: Model is determined by parameters x and covariates have to be
1.266 brouard 3751: included manually here. Then we use a call to bmij(x and cov)
3752: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3753: */
1.217 brouard 3754:
1.332 brouard 3755: int i, j, d, h, k, k1;
1.266 brouard 3756: double **out, cov[NCOVMAX+1], **bmij();
3757: double **newm, ***newmm;
1.217 brouard 3758: double agexact;
3759: double agebegin, ageend;
1.222 brouard 3760: double **oldm, **savm;
1.217 brouard 3761:
1.266 brouard 3762: newmm=po; /* To be saved */
3763: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3764: /* Hstepm could be zero and should return the unit matrix */
3765: for (i=1;i<=nlstate+ndeath;i++)
3766: for (j=1;j<=nlstate+ndeath;j++){
3767: oldm[i][j]=(i==j ? 1.0 : 0.0);
3768: po[i][j][0]=(i==j ? 1.0 : 0.0);
3769: }
3770: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3771: for(h=1; h <=nhstepm; h++){
3772: for(d=1; d <=hstepm; d++){
3773: newm=savm;
3774: /* Covariates have to be included here again */
3775: cov[1]=1.;
1.271 brouard 3776: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3777: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3778: /* Debug */
3779: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3780: cov[2]=agexact;
1.332 brouard 3781: if(nagesqr==1){
1.222 brouard 3782: cov[3]= agexact*agexact;
1.332 brouard 3783: }
3784: /** New code */
3785: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3786: if(Typevar[k1]==1){ /* A product with age */
3787: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.325 brouard 3788: }else{
1.332 brouard 3789: cov[2+nagesqr+k1]=precov[nres][k1];
1.325 brouard 3790: }
1.332 brouard 3791: }/* End of loop on model equation */
3792: /** End of new code */
3793: /** This was old code */
3794: /* for (k=1; k<=nsd;k++){ /\* For single dummy covariates only *\//\* cptcovn error *\/ */
3795: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3796: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3797: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/\* Bug valgrind *\/ */
3798: /* /\* 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)); *\/ */
3799: /* } */
3800: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3801: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3802: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3803: /* /\* 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]); *\/ */
3804: /* } */
3805: /* for (k=1; k<=cptcovage;k++){ /\* Should start at cptcovn+1 *\//\* For product with age *\/ */
3806: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age error!!!*\\/ *\/ */
3807: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3808: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3809: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3810: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3811: /* } */
3812: /* /\* 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]); *\/ */
3813: /* } */
3814: /* for (k=1; k<=cptcovprod;k++){ /\* Useless because included in cptcovn *\/ */
3815: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3816: /* if(Dummy[Tvard[k][1]]==0){ */
3817: /* if(Dummy[Tvard[k][2]]==0){ */
3818: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]; */
3819: /* }else{ */
3820: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3821: /* } */
3822: /* }else{ */
3823: /* if(Dummy[Tvard[k][2]]==0){ */
3824: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3825: /* }else{ */
3826: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3827: /* } */
3828: /* } */
3829: /* } */
3830: /* /\*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*\/ */
3831: /* /\*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*\/ */
3832: /** End of old code */
3833:
1.218 brouard 3834: /* Careful transposed matrix */
1.266 brouard 3835: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3836: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3837: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3838: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3839: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3840: /* if((int)age == 70){ */
3841: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3842: /* for(i=1; i<=nlstate+ndeath; i++) { */
3843: /* printf("%d pmmij ",i); */
3844: /* for(j=1;j<=nlstate+ndeath;j++) { */
3845: /* printf("%f ",pmmij[i][j]); */
3846: /* } */
3847: /* printf(" oldm "); */
3848: /* for(j=1;j<=nlstate+ndeath;j++) { */
3849: /* printf("%f ",oldm[i][j]); */
3850: /* } */
3851: /* printf("\n"); */
3852: /* } */
3853: /* } */
3854: savm=oldm;
3855: oldm=newm;
3856: }
3857: for(i=1; i<=nlstate+ndeath; i++)
3858: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3859: po[i][j][h]=newm[i][j];
1.268 brouard 3860: /* if(h==nhstepm) */
3861: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3862: }
1.268 brouard 3863: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3864: } /* end h */
1.268 brouard 3865: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3866: return po;
3867: }
3868:
3869:
1.162 brouard 3870: #ifdef NLOPT
3871: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3872: double fret;
3873: double *xt;
3874: int j;
3875: myfunc_data *d2 = (myfunc_data *) pd;
3876: /* xt = (p1-1); */
3877: xt=vector(1,n);
3878: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3879:
3880: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3881: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3882: printf("Function = %.12lf ",fret);
3883: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3884: printf("\n");
3885: free_vector(xt,1,n);
3886: return fret;
3887: }
3888: #endif
1.126 brouard 3889:
3890: /*************** log-likelihood *************/
3891: double func( double *x)
3892: {
1.336 brouard 3893: int i, ii, j, k, mi, d, kk, kf=0;
1.226 brouard 3894: int ioffset=0;
1.339 brouard 3895: int ipos=0,iposold=0,ncovv=0;
3896:
1.340 brouard 3897: double cotvarv, cotvarvold;
1.226 brouard 3898: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3899: double **out;
3900: double lli; /* Individual log likelihood */
3901: int s1, s2;
1.228 brouard 3902: 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 3903:
1.226 brouard 3904: double bbh, survp;
3905: double agexact;
1.336 brouard 3906: double agebegin, ageend;
1.226 brouard 3907: /*extern weight */
3908: /* We are differentiating ll according to initial status */
3909: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3910: /*for(i=1;i<imx;i++)
3911: printf(" %d\n",s[4][i]);
3912: */
1.162 brouard 3913:
1.226 brouard 3914: ++countcallfunc;
1.162 brouard 3915:
1.226 brouard 3916: cov[1]=1.;
1.126 brouard 3917:
1.226 brouard 3918: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3919: ioffset=0;
1.226 brouard 3920: if(mle==1){
3921: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3922: /* Computes the values of the ncovmodel covariates of the model
3923: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3924: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3925: to be observed in j being in i according to the model.
3926: */
1.243 brouard 3927: ioffset=2+nagesqr ;
1.233 brouard 3928: /* Fixed */
1.336 brouard 3929: for (kf=1; kf<=ncovf;kf++){ /* For each fixed covariate dummu or quant or prod */
1.319 brouard 3930: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3931: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3932: /* 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 3933: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.336 brouard 3934: 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 3935: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3936: }
1.226 brouard 3937: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3938: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3939: has been calculated etc */
3940: /* For an individual i, wav[i] gives the number of effective waves */
3941: /* We compute the contribution to Likelihood of each effective transition
3942: mw[mi][i] is real wave of the mi th effectve wave */
3943: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3944: s2=s[mw[mi+1][i]][i];
1.341 brouard 3945: 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 3946: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3947: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3948: */
1.336 brouard 3949: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
3950: /* Wave varying (but not age varying) */
1.339 brouard 3951: /* 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*\/ */
3952: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? *\/ */
3953: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
3954: /* } */
1.340 brouard 3955: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age )*/
3956: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
3957: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
3958: if(TvarFind[itv]==0){ /* Not a fixed covariate */
1.341 brouard 3959: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* cotvar[wav][ncovcol+nqv+iv][i] */
1.340 brouard 3960: }else{ /* fixed covariate */
3961: cotvarv=covar[Tvar[TvarFind[itv]]][i];
3962: }
1.339 brouard 3963: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 3964: cotvarvold=cotvarv;
3965: }else{ /* A second product */
3966: cotvarv=cotvarv*cotvarvold;
1.339 brouard 3967: }
3968: iposold=ipos;
1.340 brouard 3969: cov[ioffset+ipos]=cotvarv;
1.234 brouard 3970: }
1.339 brouard 3971: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
3972: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3973: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3974: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3975: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3976: /* 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]); */
3977: /* } */
3978: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
3979: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3980: /* /\* 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]); *\/ */
3981: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
3982: /* } */
3983: /* for products of time varying to be done */
1.234 brouard 3984: for (ii=1;ii<=nlstate+ndeath;ii++)
3985: for (j=1;j<=nlstate+ndeath;j++){
3986: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3987: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3988: }
1.336 brouard 3989:
3990: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3991: 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 3992: for(d=0; d<dh[mi][i]; d++){
3993: newm=savm;
3994: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3995: cov[2]=agexact;
3996: if(nagesqr==1)
3997: cov[3]= agexact*agexact; /* Should be changed here */
3998: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 3999: if(!FixedV[Tvar[Tage[kk]]])
4000: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4001: else
1.341 brouard 4002: 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 4003: }
4004: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4005: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4006: savm=oldm;
4007: oldm=newm;
4008: } /* end mult */
4009:
4010: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
4011: /* But now since version 0.9 we anticipate for bias at large stepm.
4012: * If stepm is larger than one month (smallest stepm) and if the exact delay
4013: * (in months) between two waves is not a multiple of stepm, we rounded to
4014: * the nearest (and in case of equal distance, to the lowest) interval but now
4015: * we keep into memory the bias bh[mi][i] and also the previous matrix product
4016: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
4017: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 4018: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
4019: * -stepm/2 to stepm/2 .
4020: * For stepm=1 the results are the same as for previous versions of Imach.
4021: * For stepm > 1 the results are less biased than in previous versions.
4022: */
1.234 brouard 4023: s1=s[mw[mi][i]][i];
4024: s2=s[mw[mi+1][i]][i];
4025: bbh=(double)bh[mi][i]/(double)stepm;
4026: /* bias bh is positive if real duration
4027: * is higher than the multiple of stepm and negative otherwise.
4028: */
4029: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
4030: if( s2 > nlstate){
4031: /* i.e. if s2 is a death state and if the date of death is known
4032: then the contribution to the likelihood is the probability to
4033: die between last step unit time and current step unit time,
4034: which is also equal to probability to die before dh
4035: minus probability to die before dh-stepm .
4036: In version up to 0.92 likelihood was computed
4037: as if date of death was unknown. Death was treated as any other
4038: health state: the date of the interview describes the actual state
4039: and not the date of a change in health state. The former idea was
4040: to consider that at each interview the state was recorded
4041: (healthy, disable or death) and IMaCh was corrected; but when we
4042: introduced the exact date of death then we should have modified
4043: the contribution of an exact death to the likelihood. This new
4044: contribution is smaller and very dependent of the step unit
4045: stepm. It is no more the probability to die between last interview
4046: and month of death but the probability to survive from last
4047: interview up to one month before death multiplied by the
4048: probability to die within a month. Thanks to Chris
4049: Jackson for correcting this bug. Former versions increased
4050: mortality artificially. The bad side is that we add another loop
4051: which slows down the processing. The difference can be up to 10%
4052: lower mortality.
4053: */
4054: /* If, at the beginning of the maximization mostly, the
4055: cumulative probability or probability to be dead is
4056: constant (ie = 1) over time d, the difference is equal to
4057: 0. out[s1][3] = savm[s1][3]: probability, being at state
4058: s1 at precedent wave, to be dead a month before current
4059: wave is equal to probability, being at state s1 at
4060: precedent wave, to be dead at mont of the current
4061: wave. Then the observed probability (that this person died)
4062: is null according to current estimated parameter. In fact,
4063: it should be very low but not zero otherwise the log go to
4064: infinity.
4065: */
1.183 brouard 4066: /* #ifdef INFINITYORIGINAL */
4067: /* lli=log(out[s1][s2] - savm[s1][s2]); */
4068: /* #else */
4069: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
4070: /* lli=log(mytinydouble); */
4071: /* else */
4072: /* lli=log(out[s1][s2] - savm[s1][s2]); */
4073: /* #endif */
1.226 brouard 4074: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4075:
1.226 brouard 4076: } else if ( s2==-1 ) { /* alive */
4077: for (j=1,survp=0. ; j<=nlstate; j++)
4078: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4079: /*survp += out[s1][j]; */
4080: lli= log(survp);
4081: }
1.336 brouard 4082: /* else if (s2==-4) { */
4083: /* for (j=3,survp=0. ; j<=nlstate; j++) */
4084: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4085: /* lli= log(survp); */
4086: /* } */
4087: /* else if (s2==-5) { */
4088: /* for (j=1,survp=0. ; j<=2; j++) */
4089: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4090: /* lli= log(survp); */
4091: /* } */
1.226 brouard 4092: else{
4093: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4094: /* 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 */
4095: }
4096: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
4097: /*if(lli ==000.0)*/
1.340 brouard 4098: /* 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 4099: ipmx +=1;
4100: sw += weight[i];
4101: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4102: /* if (lli < log(mytinydouble)){ */
4103: /* 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); */
4104: /* 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]); */
4105: /* } */
4106: } /* end of wave */
4107: } /* end of individual */
4108: } else if(mle==2){
4109: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 4110: ioffset=2+nagesqr ;
4111: for (k=1; k<=ncovf;k++)
4112: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 4113: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 4114: for(k=1; k <= ncovv ; k++){
1.341 brouard 4115: 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 4116: }
1.226 brouard 4117: for (ii=1;ii<=nlstate+ndeath;ii++)
4118: for (j=1;j<=nlstate+ndeath;j++){
4119: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4120: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4121: }
4122: for(d=0; d<=dh[mi][i]; d++){
4123: newm=savm;
4124: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4125: cov[2]=agexact;
4126: if(nagesqr==1)
4127: cov[3]= agexact*agexact;
4128: for (kk=1; kk<=cptcovage;kk++) {
4129: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4130: }
4131: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4132: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4133: savm=oldm;
4134: oldm=newm;
4135: } /* end mult */
4136:
4137: s1=s[mw[mi][i]][i];
4138: s2=s[mw[mi+1][i]][i];
4139: bbh=(double)bh[mi][i]/(double)stepm;
4140: 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 */
4141: ipmx +=1;
4142: sw += weight[i];
4143: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4144: } /* end of wave */
4145: } /* end of individual */
4146: } else if(mle==3){ /* exponential inter-extrapolation */
4147: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4148: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4149: for(mi=1; mi<= wav[i]-1; mi++){
4150: for (ii=1;ii<=nlstate+ndeath;ii++)
4151: for (j=1;j<=nlstate+ndeath;j++){
4152: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4153: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4154: }
4155: for(d=0; d<dh[mi][i]; d++){
4156: newm=savm;
4157: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4158: cov[2]=agexact;
4159: if(nagesqr==1)
4160: cov[3]= agexact*agexact;
4161: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 4162: if(!FixedV[Tvar[Tage[kk]]])
4163: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4164: else
1.341 brouard 4165: 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 4166: }
4167: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4168: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4169: savm=oldm;
4170: oldm=newm;
4171: } /* end mult */
4172:
4173: s1=s[mw[mi][i]][i];
4174: s2=s[mw[mi+1][i]][i];
4175: bbh=(double)bh[mi][i]/(double)stepm;
4176: 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 */
4177: ipmx +=1;
4178: sw += weight[i];
4179: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4180: } /* end of wave */
4181: } /* end of individual */
4182: }else if (mle==4){ /* ml=4 no inter-extrapolation */
4183: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4184: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4185: for(mi=1; mi<= wav[i]-1; mi++){
4186: for (ii=1;ii<=nlstate+ndeath;ii++)
4187: for (j=1;j<=nlstate+ndeath;j++){
4188: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4189: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4190: }
4191: for(d=0; d<dh[mi][i]; d++){
4192: newm=savm;
4193: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4194: cov[2]=agexact;
4195: if(nagesqr==1)
4196: cov[3]= agexact*agexact;
4197: for (kk=1; kk<=cptcovage;kk++) {
4198: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4199: }
1.126 brouard 4200:
1.226 brouard 4201: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4202: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4203: savm=oldm;
4204: oldm=newm;
4205: } /* end mult */
4206:
4207: s1=s[mw[mi][i]][i];
4208: s2=s[mw[mi+1][i]][i];
4209: if( s2 > nlstate){
4210: lli=log(out[s1][s2] - savm[s1][s2]);
4211: } else if ( s2==-1 ) { /* alive */
4212: for (j=1,survp=0. ; j<=nlstate; j++)
4213: survp += out[s1][j];
4214: lli= log(survp);
4215: }else{
4216: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4217: }
4218: ipmx +=1;
4219: sw += weight[i];
4220: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.340 brouard 4221: /* 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]); */
1.226 brouard 4222: } /* end of wave */
4223: } /* end of individual */
4224: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
4225: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4226: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4227: for(mi=1; mi<= wav[i]-1; mi++){
4228: for (ii=1;ii<=nlstate+ndeath;ii++)
4229: for (j=1;j<=nlstate+ndeath;j++){
4230: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4231: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4232: }
4233: for(d=0; d<dh[mi][i]; d++){
4234: newm=savm;
4235: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4236: cov[2]=agexact;
4237: if(nagesqr==1)
4238: cov[3]= agexact*agexact;
4239: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 4240: if(!FixedV[Tvar[Tage[kk]]])
4241: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4242: else
1.341 brouard 4243: 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 4244: }
1.126 brouard 4245:
1.226 brouard 4246: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4247: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4248: savm=oldm;
4249: oldm=newm;
4250: } /* end mult */
4251:
4252: s1=s[mw[mi][i]][i];
4253: s2=s[mw[mi+1][i]][i];
4254: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4255: ipmx +=1;
4256: sw += weight[i];
4257: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4258: /*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]);*/
4259: } /* end of wave */
4260: } /* end of individual */
4261: } /* End of if */
4262: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4263: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4264: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4265: return -l;
1.126 brouard 4266: }
4267:
4268: /*************** log-likelihood *************/
4269: double funcone( double *x)
4270: {
1.228 brouard 4271: /* Same as func but slower because of a lot of printf and if */
1.335 brouard 4272: int i, ii, j, k, mi, d, kk, kf=0;
1.228 brouard 4273: int ioffset=0;
1.339 brouard 4274: int ipos=0,iposold=0,ncovv=0;
4275:
1.340 brouard 4276: double cotvarv, cotvarvold;
1.131 brouard 4277: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 4278: double **out;
4279: double lli; /* Individual log likelihood */
4280: double llt;
4281: int s1, s2;
1.228 brouard 4282: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
4283:
1.126 brouard 4284: double bbh, survp;
1.187 brouard 4285: double agexact;
1.214 brouard 4286: double agebegin, ageend;
1.126 brouard 4287: /*extern weight */
4288: /* We are differentiating ll according to initial status */
4289: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4290: /*for(i=1;i<imx;i++)
4291: printf(" %d\n",s[4][i]);
4292: */
4293: cov[1]=1.;
4294:
4295: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4296: ioffset=0;
4297: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.336 brouard 4298: /* Computes the values of the ncovmodel covariates of the model
4299: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
4300: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
4301: to be observed in j being in i according to the model.
4302: */
1.243 brouard 4303: /* ioffset=2+nagesqr+cptcovage; */
4304: ioffset=2+nagesqr;
1.232 brouard 4305: /* Fixed */
1.224 brouard 4306: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4307: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.335 brouard 4308: 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 4309: /* printf("Debug3 TvarFind[%d]=%d",kf, TvarFind[kf]); */
4310: /* printf(" Tvar[TvarFind[kf]]=%d", Tvar[TvarFind[kf]]); */
4311: /* printf(" i=%d covar[Tvar[TvarFind[kf]]][i]=%f\n",i,covar[Tvar[TvarFind[kf]]][i]); */
1.335 brouard 4312: 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 4313: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4314: /* cov[2+6]=covar[Tvar[6]][i]; */
4315: /* cov[2+6]=covar[2][i]; V2 */
4316: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4317: /* cov[2+7]=covar[Tvar[7]][i]; */
4318: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4319: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4320: /* cov[2+9]=covar[Tvar[9]][i]; */
4321: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4322: }
1.336 brouard 4323: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
4324: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
4325: has been calculated etc */
4326: /* For an individual i, wav[i] gives the number of effective waves */
4327: /* We compute the contribution to Likelihood of each effective transition
4328: mw[mi][i] is real wave of the mi th effectve wave */
4329: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
4330: s2=s[mw[mi+1][i]][i];
1.341 brouard 4331: And the iv th varying covariate in the DATA is the cotvar[mw[mi+1][i]][ncovcol+nqv+iv][i]
1.336 brouard 4332: */
4333: /* This part may be useless now because everythin should be in covar */
1.232 brouard 4334: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4335: /* 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?)*\/ */
4336: /* } */
1.231 brouard 4337: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4338: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4339: /* } */
1.225 brouard 4340:
1.233 brouard 4341:
4342: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.339 brouard 4343: /* 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 */
4344: /* for(k=1; k <= ncovv ; k++){ /\* Varying covariates (single and product but no age )*\/ */
4345: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; *\/ */
4346: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
4347: /* } */
4348:
4349: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
4350: /* model V1+V3+age*V1+age*V3+V1*V3 */
4351: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
4352: /* TvarVV[1]=V3 (first time varying in the model equation, TvarVV[2]=V1 (in V1*V3) TvarVV[3]=3(V3) */
4353: /* We need the position of the time varying or product in the model */
4354: /* 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 */
4355: /* TvarVV gives the variable name */
1.340 brouard 4356: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
4357: * k= 1 2 3 4 5 6 7 8 9
4358: * varying 1 2 3 4 5
4359: * ncovv 1 2 3 4 5 6 7 8
4360: * TvarVV V3 5 1 3 3 5 1 5
4361: * TvarVVind 2 3 7 7 8 8 9 9
4362: * TvarFind[k] 1 0 0 0 0 0 0 0 0
4363: * cotvar starts at ntv=2 (because of V3 V4)
4364: */
4365: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
4366: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
4367: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
4368: if(TvarFind[itv]==0){ /* Not a fixed covariate */
1.341 brouard 4369: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.340 brouard 4370: }else{ /* fixed covariate */
4371: cotvarv=covar[Tvar[TvarFind[itv]]][i];
4372: }
1.339 brouard 4373: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 4374: cotvarvold=cotvarv;
4375: }else{ /* A second product */
4376: cotvarv=cotvarv*cotvarvold;
1.339 brouard 4377: }
4378: iposold=ipos;
1.340 brouard 4379: cov[ioffset+ipos]=cotvarv;
1.339 brouard 4380: /* For products */
4381: }
4382: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates single *\/ */
4383: /* iv=TvarVDind[itv]; /\* iv, position in the model equation of time varying covariate itv *\/ */
4384: /* /\* "V1+V3+age*V1+age*V3+V1*V3" with V3 time varying *\/ */
4385: /* /\* 1 2 3 4 5 *\/ */
4386: /* /\*itv 1 *\/ */
4387: /* /\* TvarVInd[1]= 2 *\/ */
4388: /* /\* iv= Tvar[Tmodelind[itv]]-ncovcol-nqv; /\\* Counting the # varying covariate from 1 to ntveff *\\/ *\/ */
4389: /* /\* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; *\/ */
4390: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; *\/ */
4391: /* /\* k=ioffset-2-nagesqr-cptcovage+itv; /\\* position in simple model *\\/ *\/ */
4392: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; *\/ */
4393: /* cov[ioffset+iv]=cotvar[mw[mi][i]][itv][i]; */
4394: /* /\* 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]); *\/ */
4395: /* } */
1.232 brouard 4396: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4397: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4398: /* /\* 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]); *\/ */
4399: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4400: /* } */
1.126 brouard 4401: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4402: for (j=1;j<=nlstate+ndeath;j++){
4403: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4404: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4405: }
1.214 brouard 4406:
4407: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4408: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4409: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4410: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4411: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4412: and mw[mi+1][i]. dh depends on stepm.*/
4413: newm=savm;
1.247 brouard 4414: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4415: cov[2]=agexact;
4416: if(nagesqr==1)
4417: cov[3]= agexact*agexact;
4418: for (kk=1; kk<=cptcovage;kk++) {
4419: if(!FixedV[Tvar[Tage[kk]]])
4420: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4421: else
1.341 brouard 4422: 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 4423: }
4424: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4425: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4426: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4427: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4428: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4429: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4430: savm=oldm;
4431: oldm=newm;
1.126 brouard 4432: } /* end mult */
1.336 brouard 4433: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
4434: /* But now since version 0.9 we anticipate for bias at large stepm.
4435: * If stepm is larger than one month (smallest stepm) and if the exact delay
4436: * (in months) between two waves is not a multiple of stepm, we rounded to
4437: * the nearest (and in case of equal distance, to the lowest) interval but now
4438: * we keep into memory the bias bh[mi][i] and also the previous matrix product
4439: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
4440: * probability in order to take into account the bias as a fraction of the way
4441: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
4442: * -stepm/2 to stepm/2 .
4443: * For stepm=1 the results are the same as for previous versions of Imach.
4444: * For stepm > 1 the results are less biased than in previous versions.
4445: */
1.126 brouard 4446: s1=s[mw[mi][i]][i];
4447: s2=s[mw[mi+1][i]][i];
1.217 brouard 4448: /* if(s2==-1){ */
1.268 brouard 4449: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4450: /* /\* exit(1); *\/ */
4451: /* } */
1.126 brouard 4452: bbh=(double)bh[mi][i]/(double)stepm;
4453: /* bias is positive if real duration
4454: * is higher than the multiple of stepm and negative otherwise.
4455: */
4456: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4457: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4458: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4459: for (j=1,survp=0. ; j<=nlstate; j++)
4460: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4461: lli= log(survp);
1.126 brouard 4462: }else if (mle==1){
1.242 brouard 4463: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4464: } else if(mle==2){
1.242 brouard 4465: 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 4466: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4467: 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 4468: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4469: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4470: } else{ /* mle=0 back to 1 */
1.242 brouard 4471: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4472: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4473: } /* End of if */
4474: ipmx +=1;
4475: sw += weight[i];
4476: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.342 ! brouard 4477: /* Printing covariates values for each contribution for checking */
! 4478: /* printf(" s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",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 4479: if(globpr){
1.246 brouard 4480: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4481: %11.6f %11.6f %11.6f ", \
1.242 brouard 4482: 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 4483: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.335 brouard 4484: /* printf("%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\ */
4485: /* %11.6f %11.6f %11.6f ", \ */
4486: /* num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw, */
4487: /* 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.242 brouard 4488: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4489: llt +=ll[k]*gipmx/gsw;
4490: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
1.335 brouard 4491: /* printf(" %10.6f",-ll[k]*gipmx/gsw); */
1.242 brouard 4492: }
1.342 ! brouard 4493: fprintf(ficresilk," %10.6f", -llt);
1.335 brouard 4494: /* printf(" %10.6f\n", -llt); */
1.342 ! brouard 4495: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
! 4496: fprintf(ficresilk,"%09ld ", num[i]);
! 4497: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
! 4498: fprintf(ficresilk," %g",covar[Tvar[TvarFind[kf]]][i]);
! 4499: }
! 4500: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
! 4501: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
! 4502: if(ipos!=iposold){ /* Not a product or first of a product */
! 4503: fprintf(ficresilk," %g",cov[ioffset+ipos]);
! 4504: /* printf(" %g",cov[ioffset+ipos]); */
! 4505: }else{
! 4506: fprintf(ficresilk,"*");
! 4507: /* printf("*"); */
! 4508: }
! 4509: iposold=ipos;
! 4510: }
! 4511: for (kk=1; kk<=cptcovage;kk++) {
! 4512: if(!FixedV[Tvar[Tage[kk]]]){
! 4513: fprintf(ficresilk," %g*age",covar[Tvar[Tage[kk]]][i]);
! 4514: /* printf(" %g*age",covar[Tvar[Tage[kk]]][i]); */
! 4515: }else{
! 4516: fprintf(ficresilk," %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
! 4517: /* printf(" %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
! 4518: }
! 4519: }
! 4520: /* printf("\n"); */
! 4521: /* } /\* End debugILK *\/ */
! 4522: fprintf(ficresilk,"\n");
! 4523: } /* End if globpr */
1.335 brouard 4524: } /* end of wave */
4525: } /* end of individual */
4526: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
1.232 brouard 4527: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
1.335 brouard 4528: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4529: if(globpr==0){ /* First time we count the contributions and weights */
4530: gipmx=ipmx;
4531: gsw=sw;
4532: }
1.232 brouard 4533: return -l;
1.126 brouard 4534: }
4535:
4536:
4537: /*************** function likelione ***********/
1.292 brouard 4538: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4539: {
4540: /* This routine should help understanding what is done with
4541: the selection of individuals/waves and
4542: to check the exact contribution to the likelihood.
4543: Plotting could be done.
1.342 ! brouard 4544: */
! 4545: void pstamp(FILE *ficres);
! 4546: int k, kf, kk, ncovv, iposold, ipos;
1.126 brouard 4547:
4548: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4549: strcpy(fileresilk,"ILK_");
1.202 brouard 4550: strcat(fileresilk,fileresu);
1.126 brouard 4551: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4552: printf("Problem with resultfile: %s\n", fileresilk);
4553: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4554: }
1.342 ! brouard 4555: pstamp(ficresilk);fprintf(ficresilk,"# model=1+age+%s\n",model);
1.214 brouard 4556: 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");
4557: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4558: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4559: for(k=1; k<=nlstate; k++)
4560: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
1.342 ! brouard 4561: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total) ");
! 4562:
! 4563: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
! 4564: for(kf=1;kf <= ncovf; kf++){
! 4565: fprintf(ficresilk,"V%d",Tvar[TvarFind[kf]]);
! 4566: /* printf("V%d",Tvar[TvarFind[kf]]); */
! 4567: }
! 4568: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){
! 4569: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
! 4570: if(ipos!=iposold){ /* Not a product or first of a product */
! 4571: /* printf(" %d",ipos); */
! 4572: fprintf(ficresilk," V%d",TvarVV[ncovv]);
! 4573: }else{
! 4574: /* printf("*"); */
! 4575: fprintf(ficresilk,"*");
! 4576: }
! 4577: iposold=ipos;
! 4578: }
! 4579: for (kk=1; kk<=cptcovage;kk++) {
! 4580: if(!FixedV[Tvar[Tage[kk]]]){
! 4581: /* printf(" %d*age(Fixed)",Tvar[Tage[kk]]); */
! 4582: fprintf(ficresilk," %d*age(Fixed)",Tvar[Tage[kk]]);
! 4583: }else{
! 4584: fprintf(ficresilk," %d*age(Varying)",Tvar[Tage[kk]]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
! 4585: /* printf(" %d*age(Varying)",Tvar[Tage[kk]]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
! 4586: }
! 4587: }
! 4588: /* } /\* End if debugILK *\/ */
! 4589: /* printf("\n"); */
! 4590: fprintf(ficresilk,"\n");
! 4591: } /* End glogpri */
1.126 brouard 4592:
1.292 brouard 4593: *fretone=(*func)(p);
1.126 brouard 4594: if(*globpri !=0){
4595: fclose(ficresilk);
1.205 brouard 4596: if (mle ==0)
4597: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4598: else if(mle >=1)
4599: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4600: 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 4601: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4602:
4603: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4604: 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> \
1.208 brouard 4605: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4606: }
1.207 brouard 4607: 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> \
1.204 brouard 4608: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4609: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4610: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4611: fflush(fichtm);
1.205 brouard 4612: }
1.126 brouard 4613: return;
4614: }
4615:
4616:
4617: /*********** Maximum Likelihood Estimation ***************/
4618:
4619: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4620: {
1.319 brouard 4621: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4622: double **xi;
4623: double fret;
4624: double fretone; /* Only one call to likelihood */
4625: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4626:
4627: #ifdef NLOPT
4628: int creturn;
4629: nlopt_opt opt;
4630: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4631: double *lb;
4632: double minf; /* the minimum objective value, upon return */
4633: double * p1; /* Shifted parameters from 0 instead of 1 */
4634: myfunc_data dinst, *d = &dinst;
4635: #endif
4636:
4637:
1.126 brouard 4638: xi=matrix(1,npar,1,npar);
4639: for (i=1;i<=npar;i++)
4640: for (j=1;j<=npar;j++)
4641: xi[i][j]=(i==j ? 1.0 : 0.0);
4642: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4643: strcpy(filerespow,"POW_");
1.126 brouard 4644: strcat(filerespow,fileres);
4645: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4646: printf("Problem with resultfile: %s\n", filerespow);
4647: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4648: }
4649: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4650: for (i=1;i<=nlstate;i++)
4651: for(j=1;j<=nlstate+ndeath;j++)
4652: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4653: fprintf(ficrespow,"\n");
1.162 brouard 4654: #ifdef POWELL
1.319 brouard 4655: #ifdef LINMINORIGINAL
4656: #else /* LINMINORIGINAL */
4657:
4658: flatdir=ivector(1,npar);
4659: for (j=1;j<=npar;j++) flatdir[j]=0;
4660: #endif /*LINMINORIGINAL */
4661:
4662: #ifdef FLATSUP
4663: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4664: /* reorganizing p by suppressing flat directions */
4665: for(i=1, jk=1; i <=nlstate; i++){
4666: for(k=1; k <=(nlstate+ndeath); k++){
4667: if (k != i) {
4668: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4669: if(flatdir[jk]==1){
4670: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4671: }
4672: for(j=1; j <=ncovmodel; j++){
4673: printf("%12.7f ",p[jk]);
4674: jk++;
4675: }
4676: printf("\n");
4677: }
4678: }
4679: }
4680: /* skipping */
4681: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4682: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4683: for(k=1; k <=(nlstate+ndeath); k++){
4684: if (k != i) {
4685: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4686: if(flatdir[jk]==1){
4687: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4688: for(j=1; j <=ncovmodel; jk++,j++){
4689: printf(" p[%d]=%12.7f",jk, p[jk]);
4690: /*q[jjk]=p[jk];*/
4691: }
4692: }else{
4693: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4694: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4695: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4696: /*q[jjk]=p[jk];*/
4697: }
4698: }
4699: printf("\n");
4700: }
4701: fflush(stdout);
4702: }
4703: }
4704: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4705: #else /* FLATSUP */
1.126 brouard 4706: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4707: #endif /* FLATSUP */
4708:
4709: #ifdef LINMINORIGINAL
4710: #else
4711: free_ivector(flatdir,1,npar);
4712: #endif /* LINMINORIGINAL*/
4713: #endif /* POWELL */
1.126 brouard 4714:
1.162 brouard 4715: #ifdef NLOPT
4716: #ifdef NEWUOA
4717: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4718: #else
4719: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4720: #endif
4721: lb=vector(0,npar-1);
4722: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4723: nlopt_set_lower_bounds(opt, lb);
4724: nlopt_set_initial_step1(opt, 0.1);
4725:
4726: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4727: d->function = func;
4728: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4729: nlopt_set_min_objective(opt, myfunc, d);
4730: nlopt_set_xtol_rel(opt, ftol);
4731: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4732: printf("nlopt failed! %d\n",creturn);
4733: }
4734: else {
4735: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4736: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4737: iter=1; /* not equal */
4738: }
4739: nlopt_destroy(opt);
4740: #endif
1.319 brouard 4741: #ifdef FLATSUP
4742: /* npared = npar -flatd/ncovmodel; */
4743: /* xired= matrix(1,npared,1,npared); */
4744: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4745: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4746: /* free_matrix(xire,1,npared,1,npared); */
4747: #else /* FLATSUP */
4748: #endif /* FLATSUP */
1.126 brouard 4749: free_matrix(xi,1,npar,1,npar);
4750: fclose(ficrespow);
1.203 brouard 4751: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4752: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4753: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4754:
4755: }
4756:
4757: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4758: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4759: {
4760: double **a,**y,*x,pd;
1.203 brouard 4761: /* double **hess; */
1.164 brouard 4762: int i, j;
1.126 brouard 4763: int *indx;
4764:
4765: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4766: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4767: void lubksb(double **a, int npar, int *indx, double b[]) ;
4768: void ludcmp(double **a, int npar, int *indx, double *d) ;
4769: double gompertz(double p[]);
1.203 brouard 4770: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4771:
4772: printf("\nCalculation of the hessian matrix. Wait...\n");
4773: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4774: for (i=1;i<=npar;i++){
1.203 brouard 4775: printf("%d-",i);fflush(stdout);
4776: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4777:
4778: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4779:
4780: /* printf(" %f ",p[i]);
4781: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4782: }
4783:
4784: for (i=1;i<=npar;i++) {
4785: for (j=1;j<=npar;j++) {
4786: if (j>i) {
1.203 brouard 4787: printf(".%d-%d",i,j);fflush(stdout);
4788: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4789: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4790:
4791: hess[j][i]=hess[i][j];
4792: /*printf(" %lf ",hess[i][j]);*/
4793: }
4794: }
4795: }
4796: printf("\n");
4797: fprintf(ficlog,"\n");
4798:
4799: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4800: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4801:
4802: a=matrix(1,npar,1,npar);
4803: y=matrix(1,npar,1,npar);
4804: x=vector(1,npar);
4805: indx=ivector(1,npar);
4806: for (i=1;i<=npar;i++)
4807: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4808: ludcmp(a,npar,indx,&pd);
4809:
4810: for (j=1;j<=npar;j++) {
4811: for (i=1;i<=npar;i++) x[i]=0;
4812: x[j]=1;
4813: lubksb(a,npar,indx,x);
4814: for (i=1;i<=npar;i++){
4815: matcov[i][j]=x[i];
4816: }
4817: }
4818:
4819: printf("\n#Hessian matrix#\n");
4820: fprintf(ficlog,"\n#Hessian matrix#\n");
4821: for (i=1;i<=npar;i++) {
4822: for (j=1;j<=npar;j++) {
1.203 brouard 4823: printf("%.6e ",hess[i][j]);
4824: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4825: }
4826: printf("\n");
4827: fprintf(ficlog,"\n");
4828: }
4829:
1.203 brouard 4830: /* printf("\n#Covariance matrix#\n"); */
4831: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4832: /* for (i=1;i<=npar;i++) { */
4833: /* for (j=1;j<=npar;j++) { */
4834: /* printf("%.6e ",matcov[i][j]); */
4835: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4836: /* } */
4837: /* printf("\n"); */
4838: /* fprintf(ficlog,"\n"); */
4839: /* } */
4840:
1.126 brouard 4841: /* Recompute Inverse */
1.203 brouard 4842: /* for (i=1;i<=npar;i++) */
4843: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4844: /* ludcmp(a,npar,indx,&pd); */
4845:
4846: /* printf("\n#Hessian matrix recomputed#\n"); */
4847:
4848: /* for (j=1;j<=npar;j++) { */
4849: /* for (i=1;i<=npar;i++) x[i]=0; */
4850: /* x[j]=1; */
4851: /* lubksb(a,npar,indx,x); */
4852: /* for (i=1;i<=npar;i++){ */
4853: /* y[i][j]=x[i]; */
4854: /* printf("%.3e ",y[i][j]); */
4855: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4856: /* } */
4857: /* printf("\n"); */
4858: /* fprintf(ficlog,"\n"); */
4859: /* } */
4860:
4861: /* Verifying the inverse matrix */
4862: #ifdef DEBUGHESS
4863: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4864:
1.203 brouard 4865: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4866: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4867:
4868: for (j=1;j<=npar;j++) {
4869: for (i=1;i<=npar;i++){
1.203 brouard 4870: printf("%.2f ",y[i][j]);
4871: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4872: }
4873: printf("\n");
4874: fprintf(ficlog,"\n");
4875: }
1.203 brouard 4876: #endif
1.126 brouard 4877:
4878: free_matrix(a,1,npar,1,npar);
4879: free_matrix(y,1,npar,1,npar);
4880: free_vector(x,1,npar);
4881: free_ivector(indx,1,npar);
1.203 brouard 4882: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4883:
4884:
4885: }
4886:
4887: /*************** hessian matrix ****************/
4888: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4889: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4890: int i;
4891: int l=1, lmax=20;
1.203 brouard 4892: double k1,k2, res, fx;
1.132 brouard 4893: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4894: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4895: int k=0,kmax=10;
4896: double l1;
4897:
4898: fx=func(x);
4899: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4900: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4901: l1=pow(10,l);
4902: delts=delt;
4903: for(k=1 ; k <kmax; k=k+1){
4904: delt = delta*(l1*k);
4905: p2[theta]=x[theta] +delt;
1.145 brouard 4906: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4907: p2[theta]=x[theta]-delt;
4908: k2=func(p2)-fx;
4909: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4910: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4911:
1.203 brouard 4912: #ifdef DEBUGHESSII
1.126 brouard 4913: 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);
4914: 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);
4915: #endif
4916: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4917: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4918: k=kmax;
4919: }
4920: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4921: k=kmax; l=lmax*10;
1.126 brouard 4922: }
4923: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4924: delts=delt;
4925: }
1.203 brouard 4926: } /* End loop k */
1.126 brouard 4927: }
4928: delti[theta]=delts;
4929: return res;
4930:
4931: }
4932:
1.203 brouard 4933: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4934: {
4935: int i;
1.164 brouard 4936: int l=1, lmax=20;
1.126 brouard 4937: double k1,k2,k3,k4,res,fx;
1.132 brouard 4938: double p2[MAXPARM+1];
1.203 brouard 4939: int k, kmax=1;
4940: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4941:
4942: int firstime=0;
1.203 brouard 4943:
1.126 brouard 4944: fx=func(x);
1.203 brouard 4945: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4946: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4947: p2[thetai]=x[thetai]+delti[thetai]*k;
4948: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4949: k1=func(p2)-fx;
4950:
1.203 brouard 4951: p2[thetai]=x[thetai]+delti[thetai]*k;
4952: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4953: k2=func(p2)-fx;
4954:
1.203 brouard 4955: p2[thetai]=x[thetai]-delti[thetai]*k;
4956: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4957: k3=func(p2)-fx;
4958:
1.203 brouard 4959: p2[thetai]=x[thetai]-delti[thetai]*k;
4960: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4961: k4=func(p2)-fx;
1.203 brouard 4962: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4963: if(k1*k2*k3*k4 <0.){
1.208 brouard 4964: firstime=1;
1.203 brouard 4965: kmax=kmax+10;
1.208 brouard 4966: }
4967: if(kmax >=10 || firstime ==1){
1.246 brouard 4968: 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);
4969: 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 4970: 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);
4971: 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);
4972: }
4973: #ifdef DEBUGHESSIJ
4974: v1=hess[thetai][thetai];
4975: v2=hess[thetaj][thetaj];
4976: cv12=res;
4977: /* Computing eigen value of Hessian matrix */
4978: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4979: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4980: if ((lc2 <0) || (lc1 <0) ){
4981: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4982: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4983: 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);
4984: 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);
4985: }
1.126 brouard 4986: #endif
4987: }
4988: return res;
4989: }
4990:
1.203 brouard 4991: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4992: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4993: /* { */
4994: /* int i; */
4995: /* int l=1, lmax=20; */
4996: /* double k1,k2,k3,k4,res,fx; */
4997: /* double p2[MAXPARM+1]; */
4998: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4999: /* int k=0,kmax=10; */
5000: /* double l1; */
5001:
5002: /* fx=func(x); */
5003: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
5004: /* l1=pow(10,l); */
5005: /* delts=delt; */
5006: /* for(k=1 ; k <kmax; k=k+1){ */
5007: /* delt = delti*(l1*k); */
5008: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
5009: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
5010: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
5011: /* k1=func(p2)-fx; */
5012:
5013: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
5014: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
5015: /* k2=func(p2)-fx; */
5016:
5017: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
5018: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
5019: /* k3=func(p2)-fx; */
5020:
5021: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
5022: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
5023: /* k4=func(p2)-fx; */
5024: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
5025: /* #ifdef DEBUGHESSIJ */
5026: /* 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); */
5027: /* 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); */
5028: /* #endif */
5029: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
5030: /* k=kmax; */
5031: /* } */
5032: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
5033: /* k=kmax; l=lmax*10; */
5034: /* } */
5035: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
5036: /* delts=delt; */
5037: /* } */
5038: /* } /\* End loop k *\/ */
5039: /* } */
5040: /* delti[theta]=delts; */
5041: /* return res; */
5042: /* } */
5043:
5044:
1.126 brouard 5045: /************** Inverse of matrix **************/
5046: void ludcmp(double **a, int n, int *indx, double *d)
5047: {
5048: int i,imax,j,k;
5049: double big,dum,sum,temp;
5050: double *vv;
5051:
5052: vv=vector(1,n);
5053: *d=1.0;
5054: for (i=1;i<=n;i++) {
5055: big=0.0;
5056: for (j=1;j<=n;j++)
5057: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 5058: if (big == 0.0){
5059: printf(" Singular Hessian matrix at row %d:\n",i);
5060: for (j=1;j<=n;j++) {
5061: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
5062: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
5063: }
5064: fflush(ficlog);
5065: fclose(ficlog);
5066: nrerror("Singular matrix in routine ludcmp");
5067: }
1.126 brouard 5068: vv[i]=1.0/big;
5069: }
5070: for (j=1;j<=n;j++) {
5071: for (i=1;i<j;i++) {
5072: sum=a[i][j];
5073: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
5074: a[i][j]=sum;
5075: }
5076: big=0.0;
5077: for (i=j;i<=n;i++) {
5078: sum=a[i][j];
5079: for (k=1;k<j;k++)
5080: sum -= a[i][k]*a[k][j];
5081: a[i][j]=sum;
5082: if ( (dum=vv[i]*fabs(sum)) >= big) {
5083: big=dum;
5084: imax=i;
5085: }
5086: }
5087: if (j != imax) {
5088: for (k=1;k<=n;k++) {
5089: dum=a[imax][k];
5090: a[imax][k]=a[j][k];
5091: a[j][k]=dum;
5092: }
5093: *d = -(*d);
5094: vv[imax]=vv[j];
5095: }
5096: indx[j]=imax;
5097: if (a[j][j] == 0.0) a[j][j]=TINY;
5098: if (j != n) {
5099: dum=1.0/(a[j][j]);
5100: for (i=j+1;i<=n;i++) a[i][j] *= dum;
5101: }
5102: }
5103: free_vector(vv,1,n); /* Doesn't work */
5104: ;
5105: }
5106:
5107: void lubksb(double **a, int n, int *indx, double b[])
5108: {
5109: int i,ii=0,ip,j;
5110: double sum;
5111:
5112: for (i=1;i<=n;i++) {
5113: ip=indx[i];
5114: sum=b[ip];
5115: b[ip]=b[i];
5116: if (ii)
5117: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
5118: else if (sum) ii=i;
5119: b[i]=sum;
5120: }
5121: for (i=n;i>=1;i--) {
5122: sum=b[i];
5123: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
5124: b[i]=sum/a[i][i];
5125: }
5126: }
5127:
5128: void pstamp(FILE *fichier)
5129: {
1.196 brouard 5130: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 5131: }
5132:
1.297 brouard 5133: void date2dmy(double date,double *day, double *month, double *year){
5134: double yp=0., yp1=0., yp2=0.;
5135:
5136: yp1=modf(date,&yp);/* extracts integral of date in yp and
5137: fractional in yp1 */
5138: *year=yp;
5139: yp2=modf((yp1*12),&yp);
5140: *month=yp;
5141: yp1=modf((yp2*30.5),&yp);
5142: *day=yp;
5143: if(*day==0) *day=1;
5144: if(*month==0) *month=1;
5145: }
5146:
1.253 brouard 5147:
5148:
1.126 brouard 5149: /************ Frequencies ********************/
1.251 brouard 5150: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 5151: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
5152: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 5153: { /* Some frequencies as well as proposing some starting values */
1.332 brouard 5154: /* Frequencies of any combination of dummy covariate used in the model equation */
1.265 brouard 5155: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 5156: int iind=0, iage=0;
5157: int mi; /* Effective wave */
5158: int first;
5159: double ***freq; /* Frequencies */
1.268 brouard 5160: 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 */
5161: 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 5162: double *meanq, *stdq, *idq;
1.226 brouard 5163: double **meanqt;
5164: double *pp, **prop, *posprop, *pospropt;
5165: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
5166: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
5167: double agebegin, ageend;
5168:
5169: pp=vector(1,nlstate);
1.251 brouard 5170: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5171: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
5172: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
5173: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
5174: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 5175: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 5176: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 5177: meanqt=matrix(1,lastpass,1,nqtveff);
5178: strcpy(fileresp,"P_");
5179: strcat(fileresp,fileresu);
5180: /*strcat(fileresphtm,fileresu);*/
5181: if((ficresp=fopen(fileresp,"w"))==NULL) {
5182: printf("Problem with prevalence resultfile: %s\n", fileresp);
5183: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
5184: exit(0);
5185: }
1.240 brouard 5186:
1.226 brouard 5187: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
5188: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
5189: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5190: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5191: fflush(ficlog);
5192: exit(70);
5193: }
5194: else{
5195: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 5196: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5197: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5198: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5199: }
1.319 brouard 5200: 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 5201:
1.226 brouard 5202: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
5203: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
5204: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5205: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5206: fflush(ficlog);
5207: exit(70);
1.240 brouard 5208: } else{
1.226 brouard 5209: 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 5210: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5211: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5212: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5213: }
1.319 brouard 5214: 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 5215:
1.253 brouard 5216: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
5217: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 5218: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5219: j1=0;
1.126 brouard 5220:
1.227 brouard 5221: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
1.335 brouard 5222: j=cptcoveff; /* Only simple dummy covariates used in the model */
1.330 brouard 5223: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 5224: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 5225:
5226:
1.226 brouard 5227: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
5228: reference=low_education V1=0,V2=0
5229: med_educ V1=1 V2=0,
5230: high_educ V1=0 V2=1
1.330 brouard 5231: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 5232: */
1.249 brouard 5233: dateintsum=0;
5234: k2cpt=0;
5235:
1.253 brouard 5236: if(cptcoveff == 0 )
1.265 brouard 5237: nl=1; /* Constant and age model only */
1.253 brouard 5238: else
5239: nl=2;
1.265 brouard 5240:
5241: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
5242: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.335 brouard 5243: * Loop on j1(1 to 2**cptcoveff) covariate combination
1.265 brouard 5244: * freq[s1][s2][iage] =0.
5245: * Loop on iind
5246: * ++freq[s1][s2][iage] weighted
5247: * end iind
5248: * if covariate and j!0
5249: * headers Variable on one line
5250: * endif cov j!=0
5251: * header of frequency table by age
5252: * Loop on age
5253: * pp[s1]+=freq[s1][s2][iage] weighted
5254: * pos+=freq[s1][s2][iage] weighted
5255: * Loop on s1 initial state
5256: * fprintf(ficresp
5257: * end s1
5258: * end age
5259: * if j!=0 computes starting values
5260: * end compute starting values
5261: * end j1
5262: * end nl
5263: */
1.253 brouard 5264: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
5265: if(nj==1)
5266: j=0; /* First pass for the constant */
1.265 brouard 5267: else{
1.335 brouard 5268: 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 5269: }
1.251 brouard 5270: first=1;
1.332 brouard 5271: 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 5272: posproptt=0.;
1.330 brouard 5273: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 5274: scanf("%d", i);*/
5275: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 5276: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 5277: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 5278: freq[i][s2][m]=0;
1.251 brouard 5279:
5280: for (i=1; i<=nlstate; i++) {
1.240 brouard 5281: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 5282: prop[i][m]=0;
5283: posprop[i]=0;
5284: pospropt[i]=0;
5285: }
1.283 brouard 5286: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 5287: idq[z1]=0.;
5288: meanq[z1]=0.;
5289: stdq[z1]=0.;
1.283 brouard 5290: }
5291: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 5292: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 5293: /* meanqt[m][z1]=0.; */
5294: /* } */
5295: /* } */
1.251 brouard 5296: /* dateintsum=0; */
5297: /* k2cpt=0; */
5298:
1.265 brouard 5299: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 5300: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
5301: bool=1;
5302: if(j !=0){
5303: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.335 brouard 5304: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
5305: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
1.251 brouard 5306: /* if(Tvaraff[z1] ==-20){ */
5307: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
5308: /* }else if(Tvaraff[z1] ==-10){ */
5309: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 brouard 5310: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
1.335 brouard 5311: /* 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); */
5312: if(Tvaraff[z1]<1 || Tvaraff[z1]>=NCOVMAX)
1.338 brouard 5313: printf("Error Tvaraff[z1]=%d<1 or >=%d, cptcoveff=%d model=1+age+%s\n",Tvaraff[z1],NCOVMAX, cptcoveff, model);
1.332 brouard 5314: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]){ /* for combination j1 of covariates */
1.265 brouard 5315: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 5316: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.332 brouard 5317: /* 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", */
5318: /* bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),*/
5319: /* j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
1.251 brouard 5320: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
5321: } /* Onlyf fixed */
5322: } /* end z1 */
1.335 brouard 5323: } /* cptcoveff > 0 */
1.251 brouard 5324: } /* end any */
5325: }/* end j==0 */
1.265 brouard 5326: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 5327: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 5328: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 5329: m=mw[mi][iind];
5330: if(j!=0){
5331: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.335 brouard 5332: for (z1=1; z1<=cptcoveff; z1++) {
1.251 brouard 5333: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 5334: /* iv= Tvar[Tmodelind[z1]]-ncovcol-nqv; /\* Good *\/ */
5335: iv= Tvar[Tmodelind[z1]]; /* Good *//* because cotvar starts now at first at ncovcol+nqv+ntv */
1.332 brouard 5336: 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 5337: value is -1, we don't select. It differs from the
5338: constant and age model which counts them. */
5339: bool=0; /* not selected */
5340: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.334 brouard 5341: /* i1=Tvaraff[z1]; */
5342: /* i2=TnsdVar[i1]; */
5343: /* i3=nbcode[i1][i2]; */
5344: /* i4=covar[i1][iind]; */
5345: /* if(i4 != i3){ */
5346: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) { /* Bug valgrind */
1.251 brouard 5347: bool=0;
5348: }
5349: }
5350: }
5351: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
5352: } /* end j==0 */
5353: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 5354: if(bool==1){ /*Selected */
1.251 brouard 5355: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
5356: and mw[mi+1][iind]. dh depends on stepm. */
5357: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
5358: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
5359: if(m >=firstpass && m <=lastpass){
5360: k2=anint[m][iind]+(mint[m][iind]/12.);
5361: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
5362: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
5363: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
5364: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
5365: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
5366: if (m<lastpass) {
5367: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
5368: /* 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]); */
5369: if(s[m][iind]==-1)
5370: 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.));
5371: 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 5372: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
5373: if(!isnan(covar[ncovcol+z1][iind])){
1.332 brouard 5374: idq[z1]=idq[z1]+weight[iind];
5375: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
5376: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
5377: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
1.311 brouard 5378: }
1.284 brouard 5379: }
1.251 brouard 5380: /* if((int)agev[m][iind] == 55) */
5381: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
5382: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
5383: 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 5384: }
1.251 brouard 5385: } /* end if between passes */
5386: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
5387: dateintsum=dateintsum+k2; /* on all covariates ?*/
5388: k2cpt++;
5389: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 5390: }
1.251 brouard 5391: }else{
5392: bool=1;
5393: }/* end bool 2 */
5394: } /* end m */
1.284 brouard 5395: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
5396: /* idq[z1]=idq[z1]+weight[iind]; */
5397: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
5398: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
5399: /* } */
1.251 brouard 5400: } /* end bool */
5401: } /* end iind = 1 to imx */
1.319 brouard 5402: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 5403: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
5404:
5405:
5406: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.335 brouard 5407: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
1.265 brouard 5408: pstamp(ficresp);
1.335 brouard 5409: if (cptcoveff>0 && j!=0){
1.265 brouard 5410: pstamp(ficresp);
1.251 brouard 5411: printf( "\n#********** Variable ");
5412: fprintf(ficresp, "\n#********** Variable ");
5413: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
5414: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
5415: fprintf(ficlog, "\n#********** Variable ");
1.340 brouard 5416: for (z1=1; z1<=cptcoveff; z1++){
1.251 brouard 5417: if(!FixedV[Tvaraff[z1]]){
1.330 brouard 5418: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5419: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5420: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5421: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5422: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 5423: }else{
1.330 brouard 5424: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5425: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5426: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5427: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5428: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 5429: }
5430: }
5431: printf( "**********\n#");
5432: fprintf(ficresp, "**********\n#");
5433: fprintf(ficresphtm, "**********</h3>\n");
5434: fprintf(ficresphtmfr, "**********</h3>\n");
5435: fprintf(ficlog, "**********\n");
5436: }
1.284 brouard 5437: /*
5438: Printing means of quantitative variables if any
5439: */
5440: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5441: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5442: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5443: if(weightopt==1){
5444: printf(" Weighted mean and standard deviation of");
5445: fprintf(ficlog," Weighted mean and standard deviation of");
5446: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5447: }
1.311 brouard 5448: /* mu = \frac{w x}{\sum w}
5449: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5450: */
5451: 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]));
5452: 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]));
5453: 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 5454: }
5455: /* for (z1=1; z1<= nqtveff; z1++) { */
5456: /* for(m=1;m<=lastpass;m++){ */
5457: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5458: /* } */
5459: /* } */
1.283 brouard 5460:
1.251 brouard 5461: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.335 brouard 5462: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 5463: fprintf(ficresp, " Age");
1.335 brouard 5464: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) {
5465: 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]]);
5466: fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5467: }
1.251 brouard 5468: for(i=1; i<=nlstate;i++) {
1.335 brouard 5469: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5470: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5471: }
1.335 brouard 5472: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5473: fprintf(ficresphtm, "\n");
5474:
5475: /* Header of frequency table by age */
5476: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5477: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5478: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5479: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5480: if(s2!=0 && m!=0)
5481: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5482: }
1.226 brouard 5483: }
1.251 brouard 5484: fprintf(ficresphtmfr, "\n");
5485:
5486: /* For each age */
5487: for(iage=iagemin; iage <= iagemax+3; iage++){
5488: fprintf(ficresphtm,"<tr>");
5489: if(iage==iagemax+1){
5490: fprintf(ficlog,"1");
5491: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5492: }else if(iage==iagemax+2){
5493: fprintf(ficlog,"0");
5494: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5495: }else if(iage==iagemax+3){
5496: fprintf(ficlog,"Total");
5497: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5498: }else{
1.240 brouard 5499: if(first==1){
1.251 brouard 5500: first=0;
5501: printf("See log file for details...\n");
5502: }
5503: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5504: fprintf(ficlog,"Age %d", iage);
5505: }
1.265 brouard 5506: for(s1=1; s1 <=nlstate ; s1++){
5507: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5508: pp[s1] += freq[s1][m][iage];
1.251 brouard 5509: }
1.265 brouard 5510: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5511: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5512: pos += freq[s1][m][iage];
5513: if(pp[s1]>=1.e-10){
1.251 brouard 5514: if(first==1){
1.265 brouard 5515: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5516: }
1.265 brouard 5517: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5518: }else{
5519: if(first==1)
1.265 brouard 5520: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5521: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5522: }
5523: }
5524:
1.265 brouard 5525: for(s1=1; s1 <=nlstate ; s1++){
5526: /* posprop[s1]=0; */
5527: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5528: pp[s1] += freq[s1][m][iage];
5529: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5530:
5531: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5532: pos += pp[s1]; /* pos is the total number of transitions until this age */
5533: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5534: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5535: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5536: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5537: }
5538:
5539: /* Writing ficresp */
1.335 brouard 5540: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5541: if( iage <= iagemax){
5542: fprintf(ficresp," %d",iage);
5543: }
5544: }else if( nj==2){
5545: if( iage <= iagemax){
5546: fprintf(ficresp," %d",iage);
1.335 brouard 5547: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.265 brouard 5548: }
1.240 brouard 5549: }
1.265 brouard 5550: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5551: if(pos>=1.e-5){
1.251 brouard 5552: if(first==1)
1.265 brouard 5553: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5554: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5555: }else{
5556: if(first==1)
1.265 brouard 5557: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5558: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5559: }
5560: if( iage <= iagemax){
5561: if(pos>=1.e-5){
1.335 brouard 5562: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5563: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5564: }else if( nj==2){
5565: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5566: }
5567: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5568: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5569: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5570: } else{
1.335 brouard 5571: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 5572: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5573: }
1.240 brouard 5574: }
1.265 brouard 5575: pospropt[s1] +=posprop[s1];
5576: } /* end loop s1 */
1.251 brouard 5577: /* pospropt=0.; */
1.265 brouard 5578: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5579: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5580: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5581: if(first==1){
1.265 brouard 5582: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5583: }
1.265 brouard 5584: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5585: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5586: }
1.265 brouard 5587: if(s1!=0 && m!=0)
5588: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5589: }
1.265 brouard 5590: } /* end loop s1 */
1.251 brouard 5591: posproptt=0.;
1.265 brouard 5592: for(s1=1; s1 <=nlstate; s1++){
5593: posproptt += pospropt[s1];
1.251 brouard 5594: }
5595: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5596: fprintf(ficresphtm,"</tr>\n");
1.335 brouard 5597: if((cptcoveff==0 && nj==1)|| nj==2 ) {
1.265 brouard 5598: if(iage <= iagemax)
5599: fprintf(ficresp,"\n");
1.240 brouard 5600: }
1.251 brouard 5601: if(first==1)
5602: printf("Others in log...\n");
5603: fprintf(ficlog,"\n");
5604: } /* end loop age iage */
1.265 brouard 5605:
1.251 brouard 5606: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5607: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5608: if(posproptt < 1.e-5){
1.265 brouard 5609: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5610: }else{
1.265 brouard 5611: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5612: }
1.226 brouard 5613: }
1.251 brouard 5614: fprintf(ficresphtm,"</tr>\n");
5615: fprintf(ficresphtm,"</table>\n");
5616: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5617: if(posproptt < 1.e-5){
1.251 brouard 5618: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5619: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5620: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5621: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5622: invalidvarcomb[j1]=1;
1.226 brouard 5623: }else{
1.338 brouard 5624: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced (or no resultline).</p>",j1);
1.251 brouard 5625: invalidvarcomb[j1]=0;
1.226 brouard 5626: }
1.251 brouard 5627: fprintf(ficresphtmfr,"</table>\n");
5628: fprintf(ficlog,"\n");
5629: if(j!=0){
5630: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5631: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5632: for(k=1; k <=(nlstate+ndeath); k++){
5633: if (k != i) {
1.265 brouard 5634: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5635: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5636: if(j1==1){ /* All dummy covariates to zero */
5637: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5638: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5639: printf("%d%d ",i,k);
5640: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5641: 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]));
5642: 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]));
5643: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5644: }
1.253 brouard 5645: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5646: for(iage=iagemin; iage <= iagemax+3; iage++){
5647: x[iage]= (double)iage;
5648: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5649: /* 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 5650: }
1.268 brouard 5651: /* Some are not finite, but linreg will ignore these ages */
5652: no=0;
1.253 brouard 5653: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5654: pstart[s1]=b;
5655: pstart[s1-1]=a;
1.252 brouard 5656: }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 */
5657: 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]);
5658: 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 5659: 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 5660: printf("%d%d ",i,k);
5661: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5662: 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 5663: }else{ /* Other cases, like quantitative fixed or varying covariates */
5664: ;
5665: }
5666: /* printf("%12.7f )", param[i][jj][k]); */
5667: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5668: s1++;
1.251 brouard 5669: } /* end jj */
5670: } /* end k!= i */
5671: } /* end k */
1.265 brouard 5672: } /* end i, s1 */
1.251 brouard 5673: } /* end j !=0 */
5674: } /* end selected combination of covariate j1 */
5675: if(j==0){ /* We can estimate starting values from the occurences in each case */
5676: printf("#Freqsummary: Starting values for the constants:\n");
5677: fprintf(ficlog,"\n");
1.265 brouard 5678: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5679: for(k=1; k <=(nlstate+ndeath); k++){
5680: if (k != i) {
5681: printf("%d%d ",i,k);
5682: fprintf(ficlog,"%d%d ",i,k);
5683: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5684: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5685: if(jj==1){ /* Age has to be done */
1.265 brouard 5686: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5687: 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]));
5688: 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 5689: }
5690: /* printf("%12.7f )", param[i][jj][k]); */
5691: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5692: s1++;
1.250 brouard 5693: }
1.251 brouard 5694: printf("\n");
5695: fprintf(ficlog,"\n");
1.250 brouard 5696: }
5697: }
1.284 brouard 5698: } /* end of state i */
1.251 brouard 5699: printf("#Freqsummary\n");
5700: fprintf(ficlog,"\n");
1.265 brouard 5701: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5702: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5703: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5704: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5705: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5706: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5707: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5708: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5709: /* } */
5710: }
1.265 brouard 5711: } /* end loop s1 */
1.251 brouard 5712:
5713: printf("\n");
5714: fprintf(ficlog,"\n");
5715: } /* end j=0 */
1.249 brouard 5716: } /* end j */
1.252 brouard 5717:
1.253 brouard 5718: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5719: for(i=1, jk=1; i <=nlstate; i++){
5720: for(j=1; j <=nlstate+ndeath; j++){
5721: if(j!=i){
5722: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5723: printf("%1d%1d",i,j);
5724: fprintf(ficparo,"%1d%1d",i,j);
5725: for(k=1; k<=ncovmodel;k++){
5726: /* printf(" %lf",param[i][j][k]); */
5727: /* fprintf(ficparo," %lf",param[i][j][k]); */
5728: p[jk]=pstart[jk];
5729: printf(" %f ",pstart[jk]);
5730: fprintf(ficparo," %f ",pstart[jk]);
5731: jk++;
5732: }
5733: printf("\n");
5734: fprintf(ficparo,"\n");
5735: }
5736: }
5737: }
5738: } /* end mle=-2 */
1.226 brouard 5739: dateintmean=dateintsum/k2cpt;
1.296 brouard 5740: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5741:
1.226 brouard 5742: fclose(ficresp);
5743: fclose(ficresphtm);
5744: fclose(ficresphtmfr);
1.283 brouard 5745: free_vector(idq,1,nqfveff);
1.226 brouard 5746: free_vector(meanq,1,nqfveff);
1.284 brouard 5747: free_vector(stdq,1,nqfveff);
1.226 brouard 5748: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5749: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5750: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5751: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5752: free_vector(pospropt,1,nlstate);
5753: free_vector(posprop,1,nlstate);
1.251 brouard 5754: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5755: free_vector(pp,1,nlstate);
5756: /* End of freqsummary */
5757: }
1.126 brouard 5758:
1.268 brouard 5759: /* Simple linear regression */
5760: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5761:
5762: /* y=a+bx regression */
5763: double sumx = 0.0; /* sum of x */
5764: double sumx2 = 0.0; /* sum of x**2 */
5765: double sumxy = 0.0; /* sum of x * y */
5766: double sumy = 0.0; /* sum of y */
5767: double sumy2 = 0.0; /* sum of y**2 */
5768: double sume2 = 0.0; /* sum of square or residuals */
5769: double yhat;
5770:
5771: double denom=0;
5772: int i;
5773: int ne=*no;
5774:
5775: for ( i=ifi, ne=0;i<=ila;i++) {
5776: if(!isfinite(x[i]) || !isfinite(y[i])){
5777: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5778: continue;
5779: }
5780: ne=ne+1;
5781: sumx += x[i];
5782: sumx2 += x[i]*x[i];
5783: sumxy += x[i] * y[i];
5784: sumy += y[i];
5785: sumy2 += y[i]*y[i];
5786: denom = (ne * sumx2 - sumx*sumx);
5787: /* 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); */
5788: }
5789:
5790: denom = (ne * sumx2 - sumx*sumx);
5791: if (denom == 0) {
5792: // vertical, slope m is infinity
5793: *b = INFINITY;
5794: *a = 0;
5795: if (r) *r = 0;
5796: return 1;
5797: }
5798:
5799: *b = (ne * sumxy - sumx * sumy) / denom;
5800: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5801: if (r!=NULL) {
5802: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5803: sqrt((sumx2 - sumx*sumx/ne) *
5804: (sumy2 - sumy*sumy/ne));
5805: }
5806: *no=ne;
5807: for ( i=ifi, ne=0;i<=ila;i++) {
5808: if(!isfinite(x[i]) || !isfinite(y[i])){
5809: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5810: continue;
5811: }
5812: ne=ne+1;
5813: yhat = y[i] - *a -*b* x[i];
5814: sume2 += yhat * yhat ;
5815:
5816: denom = (ne * sumx2 - sumx*sumx);
5817: /* 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); */
5818: }
5819: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5820: *sa= *sb * sqrt(sumx2/ne);
5821:
5822: return 0;
5823: }
5824:
1.126 brouard 5825: /************ Prevalence ********************/
1.227 brouard 5826: 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)
5827: {
5828: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5829: in each health status at the date of interview (if between dateprev1 and dateprev2).
5830: We still use firstpass and lastpass as another selection.
5831: */
1.126 brouard 5832:
1.227 brouard 5833: int i, m, jk, j1, bool, z1,j, iv;
5834: int mi; /* Effective wave */
5835: int iage;
5836: double agebegin, ageend;
5837:
5838: double **prop;
5839: double posprop;
5840: double y2; /* in fractional years */
5841: int iagemin, iagemax;
5842: int first; /** to stop verbosity which is redirected to log file */
5843:
5844: iagemin= (int) agemin;
5845: iagemax= (int) agemax;
5846: /*pp=vector(1,nlstate);*/
1.251 brouard 5847: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5848: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5849: j1=0;
1.222 brouard 5850:
1.227 brouard 5851: /*j=cptcoveff;*/
5852: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5853:
1.288 brouard 5854: first=0;
1.335 brouard 5855: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of simple dummy covariates */
1.227 brouard 5856: for (i=1; i<=nlstate; i++)
1.251 brouard 5857: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5858: prop[i][iage]=0.0;
5859: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5860: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5861: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5862:
5863: for (i=1; i<=imx; i++) { /* Each individual */
5864: bool=1;
5865: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5866: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5867: m=mw[mi][i];
5868: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5869: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5870: for (z1=1; z1<=cptcoveff; z1++){
5871: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 5872: iv= Tvar[Tmodelind[z1]];/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.332 brouard 5873: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality */
1.227 brouard 5874: bool=0;
5875: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.332 brouard 5876: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.227 brouard 5877: bool=0;
5878: }
5879: }
5880: if(bool==1){ /* Otherwise we skip that wave/person */
5881: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5882: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5883: if(m >=firstpass && m <=lastpass){
5884: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5885: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5886: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5887: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5888: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5889: 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);
5890: exit(1);
5891: }
5892: if (s[m][i]>0 && s[m][i]<=nlstate) {
5893: /*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]]);*/
5894: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5895: prop[s[m][i]][iagemax+3] += weight[i];
5896: } /* end valid statuses */
5897: } /* end selection of dates */
5898: } /* end selection of waves */
5899: } /* end bool */
5900: } /* end wave */
5901: } /* end individual */
5902: for(i=iagemin; i <= iagemax+3; i++){
5903: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5904: posprop += prop[jk][i];
5905: }
5906:
5907: for(jk=1; jk <=nlstate ; jk++){
5908: if( i <= iagemax){
5909: if(posprop>=1.e-5){
5910: probs[i][jk][j1]= prop[jk][i]/posprop;
5911: } else{
1.288 brouard 5912: if(!first){
5913: first=1;
1.266 brouard 5914: 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]);
5915: }else{
1.288 brouard 5916: 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 5917: }
5918: }
5919: }
5920: }/* end jk */
5921: }/* end i */
1.222 brouard 5922: /*} *//* end i1 */
1.227 brouard 5923: } /* end j1 */
1.222 brouard 5924:
1.227 brouard 5925: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5926: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5927: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5928: } /* End of prevalence */
1.126 brouard 5929:
5930: /************* Waves Concatenation ***************/
5931:
5932: 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)
5933: {
1.298 brouard 5934: /* 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 5935: Death is a valid wave (if date is known).
5936: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5937: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5938: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5939: */
1.126 brouard 5940:
1.224 brouard 5941: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5942: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5943: double sum=0., jmean=0.;*/
1.224 brouard 5944: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5945: int j, k=0,jk, ju, jl;
5946: double sum=0.;
5947: first=0;
1.214 brouard 5948: firstwo=0;
1.217 brouard 5949: firsthree=0;
1.218 brouard 5950: firstfour=0;
1.164 brouard 5951: jmin=100000;
1.126 brouard 5952: jmax=-1;
5953: jmean=0.;
1.224 brouard 5954:
5955: /* Treating live states */
1.214 brouard 5956: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5957: mi=0; /* First valid wave */
1.227 brouard 5958: mli=0; /* Last valid wave */
1.309 brouard 5959: m=firstpass; /* Loop on waves */
5960: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5961: 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 */
5962: mli=m-1;/* mw[++mi][i]=m-1; */
5963: }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 5964: 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 5965: mli=m;
1.224 brouard 5966: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5967: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5968: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5969: }
1.309 brouard 5970: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5971: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5972: break;
1.224 brouard 5973: #else
1.317 brouard 5974: 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 5975: if(firsthree == 0){
1.302 brouard 5976: 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 5977: firsthree=1;
1.317 brouard 5978: }else if(firsthree >=1 && firsthree < 10){
5979: 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);
5980: firsthree++;
5981: }else if(firsthree == 10){
5982: printf("Information, too many Information flags: no more reported to log either\n");
5983: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
5984: firsthree++;
5985: }else{
5986: firsthree++;
1.227 brouard 5987: }
1.309 brouard 5988: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5989: mli=m;
5990: }
5991: if(s[m][i]==-2){ /* Vital status is really unknown */
5992: nbwarn++;
1.309 brouard 5993: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5994: 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);
5995: 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);
5996: }
5997: break;
5998: }
5999: break;
1.224 brouard 6000: #endif
1.227 brouard 6001: }/* End m >= lastpass */
1.126 brouard 6002: }/* end while */
1.224 brouard 6003:
1.227 brouard 6004: /* 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 6005: /* After last pass */
1.224 brouard 6006: /* Treating death states */
1.214 brouard 6007: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 6008: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
6009: /* } */
1.126 brouard 6010: mi++; /* Death is another wave */
6011: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 6012: /* Only death is a correct wave */
1.126 brouard 6013: mw[mi][i]=m;
1.257 brouard 6014: } /* else not in a death state */
1.224 brouard 6015: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 6016: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 6017: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 6018: 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 6019: nbwarn++;
6020: if(firstfiv==0){
1.309 brouard 6021: 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 6022: firstfiv=1;
6023: }else{
1.309 brouard 6024: 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 6025: }
1.309 brouard 6026: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
6027: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 6028: nberr++;
6029: if(firstwo==0){
1.309 brouard 6030: 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 6031: firstwo=1;
6032: }
1.309 brouard 6033: 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 6034: }
1.257 brouard 6035: }else{ /* if date of interview is unknown */
1.227 brouard 6036: /* death is known but not confirmed by death status at any wave */
6037: if(firstfour==0){
1.309 brouard 6038: 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 6039: firstfour=1;
6040: }
1.309 brouard 6041: 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 6042: }
1.224 brouard 6043: } /* end if date of death is known */
6044: #endif
1.309 brouard 6045: wav[i]=mi; /* mi should be the last effective wave (or mli), */
6046: /* wav[i]=mw[mi][i]; */
1.126 brouard 6047: if(mi==0){
6048: nbwarn++;
6049: if(first==0){
1.227 brouard 6050: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
6051: first=1;
1.126 brouard 6052: }
6053: if(first==1){
1.227 brouard 6054: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 6055: }
6056: } /* end mi==0 */
6057: } /* End individuals */
1.214 brouard 6058: /* wav and mw are no more changed */
1.223 brouard 6059:
1.317 brouard 6060: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
6061: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
6062:
6063:
1.126 brouard 6064: for(i=1; i<=imx; i++){
6065: for(mi=1; mi<wav[i];mi++){
6066: if (stepm <=0)
1.227 brouard 6067: dh[mi][i]=1;
1.126 brouard 6068: else{
1.260 brouard 6069: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 6070: if (agedc[i] < 2*AGESUP) {
6071: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
6072: if(j==0) j=1; /* Survives at least one month after exam */
6073: else if(j<0){
6074: nberr++;
6075: 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]);
6076: j=1; /* Temporary Dangerous patch */
6077: 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);
6078: 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]);
6079: 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);
6080: }
6081: k=k+1;
6082: if (j >= jmax){
6083: jmax=j;
6084: ijmax=i;
6085: }
6086: if (j <= jmin){
6087: jmin=j;
6088: ijmin=i;
6089: }
6090: sum=sum+j;
6091: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
6092: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
6093: }
6094: }
6095: else{
6096: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 6097: /* 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 6098:
1.227 brouard 6099: k=k+1;
6100: if (j >= jmax) {
6101: jmax=j;
6102: ijmax=i;
6103: }
6104: else if (j <= jmin){
6105: jmin=j;
6106: ijmin=i;
6107: }
6108: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
6109: /*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]);*/
6110: if(j<0){
6111: nberr++;
6112: 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]);
6113: 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]);
6114: }
6115: sum=sum+j;
6116: }
6117: jk= j/stepm;
6118: jl= j -jk*stepm;
6119: ju= j -(jk+1)*stepm;
6120: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
6121: if(jl==0){
6122: dh[mi][i]=jk;
6123: bh[mi][i]=0;
6124: }else{ /* We want a negative bias in order to only have interpolation ie
6125: * to avoid the price of an extra matrix product in likelihood */
6126: dh[mi][i]=jk+1;
6127: bh[mi][i]=ju;
6128: }
6129: }else{
6130: if(jl <= -ju){
6131: dh[mi][i]=jk;
6132: bh[mi][i]=jl; /* bias is positive if real duration
6133: * is higher than the multiple of stepm and negative otherwise.
6134: */
6135: }
6136: else{
6137: dh[mi][i]=jk+1;
6138: bh[mi][i]=ju;
6139: }
6140: if(dh[mi][i]==0){
6141: dh[mi][i]=1; /* At least one step */
6142: bh[mi][i]=ju; /* At least one step */
6143: /* 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);*/
6144: }
6145: } /* end if mle */
1.126 brouard 6146: }
6147: } /* end wave */
6148: }
6149: jmean=sum/k;
6150: 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 6151: 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 6152: }
1.126 brouard 6153:
6154: /*********** Tricode ****************************/
1.220 brouard 6155: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 6156: {
6157: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
6158: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
6159: * Boring subroutine which should only output nbcode[Tvar[j]][k]
6160: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
6161: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
6162: */
1.130 brouard 6163:
1.242 brouard 6164: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
6165: int modmaxcovj=0; /* Modality max of covariates j */
6166: int cptcode=0; /* Modality max of covariates j */
6167: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 6168:
6169:
1.242 brouard 6170: /* cptcoveff=0; */
6171: /* *cptcov=0; */
1.126 brouard 6172:
1.242 brouard 6173: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 6174: for (k=1; k <= maxncov; k++)
6175: for(j=1; j<=2; j++)
6176: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 6177:
1.242 brouard 6178: /* Loop on covariates without age and products and no quantitative variable */
1.335 brouard 6179: 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 6180: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
1.339 brouard 6181: printf("Testing k=%d, cptcovt=%d\n",k, cptcovt);
6182: if(Dummy[k]==0 && Typevar[k] !=1 && Typevar[k] != 2){ /* Dummy covariate and not age product nor fixed product */
1.242 brouard 6183: switch(Fixed[k]) {
6184: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 6185: modmaxcovj=0;
6186: modmincovj=0;
1.242 brouard 6187: 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 6188: /* 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 6189: ij=(int)(covar[Tvar[k]][i]);
6190: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
6191: * If product of Vn*Vm, still boolean *:
6192: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
6193: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
6194: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
6195: modality of the nth covariate of individual i. */
6196: if (ij > modmaxcovj)
6197: modmaxcovj=ij;
6198: else if (ij < modmincovj)
6199: modmincovj=ij;
1.287 brouard 6200: if (ij <0 || ij >1 ){
1.311 brouard 6201: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6202: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6203: fflush(ficlog);
6204: exit(1);
1.287 brouard 6205: }
6206: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 6207: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
6208: exit(1);
6209: }else
6210: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
6211: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
6212: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
6213: /* getting the maximum value of the modality of the covariate
6214: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
6215: female ies 1, then modmaxcovj=1.
6216: */
6217: } /* end for loop on individuals i */
6218: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6219: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6220: cptcode=modmaxcovj;
6221: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
6222: /*for (i=0; i<=cptcode; i++) {*/
6223: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
6224: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6225: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6226: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
6227: if( j != -1){
6228: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
6229: covariate for which somebody answered excluding
6230: undefined. Usually 2: 0 and 1. */
6231: }
6232: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
6233: covariate for which somebody answered including
6234: undefined. Usually 3: -1, 0 and 1. */
6235: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
6236: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
6237: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 6238:
1.242 brouard 6239: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
6240: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
6241: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
6242: /* modmincovj=3; modmaxcovj = 7; */
6243: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
6244: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
6245: /* defining two dummy variables: variables V1_1 and V1_2.*/
6246: /* nbcode[Tvar[j]][ij]=k; */
6247: /* nbcode[Tvar[j]][1]=0; */
6248: /* nbcode[Tvar[j]][2]=1; */
6249: /* nbcode[Tvar[j]][3]=2; */
6250: /* To be continued (not working yet). */
6251: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 6252:
6253: /* 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*/
6254: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
6255: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
6256: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
6257: /*, could be restored in the future */
6258: 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 6259: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
6260: break;
6261: }
6262: ij++;
1.287 brouard 6263: 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 6264: cptcode = ij; /* New max modality for covar j */
6265: } /* end of loop on modality i=-1 to 1 or more */
6266: break;
6267: case 1: /* Testing on varying covariate, could be simple and
6268: * should look at waves or product of fixed *
6269: * varying. No time to test -1, assuming 0 and 1 only */
6270: ij=0;
6271: for(i=0; i<=1;i++){
6272: nbcode[Tvar[k]][++ij]=i;
6273: }
6274: break;
6275: default:
6276: break;
6277: } /* end switch */
6278: } /* end dummy test */
1.342 ! brouard 6279: if(Dummy[k]==1 && Typevar[k] !=1 && Fixed ==0){ /* Fixed Quantitative covariate and not age product */
1.311 brouard 6280: 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 6281: if(Tvar[k]<=0 || Tvar[k]>=NCOVMAX){
6282: printf("Error k=%d \n",k);
6283: exit(1);
6284: }
1.311 brouard 6285: if(isnan(covar[Tvar[k]][i])){
6286: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6287: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6288: fflush(ficlog);
6289: exit(1);
6290: }
6291: }
1.335 brouard 6292: } /* end Quanti */
1.287 brouard 6293: } /* 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 6294:
6295: for (k=-1; k< maxncov; k++) Ndum[k]=0;
6296: /* Look at fixed dummy (single or product) covariates to check empty modalities */
6297: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
6298: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
6299: 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 */
6300: 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 */
6301: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
6302: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
6303:
6304: ij=0;
6305: /* 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 6306: 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 */
6307: /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
1.242 brouard 6308: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
6309: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
1.335 brouard 6310: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy simple and non empty in the model */
6311: /* Typevar[k] =0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
6312: /* 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 6313: /* If product not in single variable we don't print results */
6314: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
1.335 brouard 6315: ++ij;/* V5 + V4 + V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V1, *//* V5 quanti, V2 quanti, V4, V3, V1 dummies */
6316: /* k= 1 2 3 4 5 6 7 8 9 */
6317: /* Tvar[k]= 5 4 3 6 5 2 7 1 1 */
6318: /* ij 1 2 3 */
6319: /* Tvaraff[ij]= 4 3 1 */
6320: /* Tmodelind[ij]=2 3 9 */
6321: /* TmodelInvind[ij]=2 1 1 */
1.242 brouard 6322: 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*/
6323: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
6324: 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 */
6325: if(Fixed[k]!=0)
6326: anyvaryingduminmodel=1;
6327: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
6328: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
6329: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
6330: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
6331: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
6332: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
6333: }
6334: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
6335: /* ij--; */
6336: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.335 brouard 6337: *cptcov=ij; /* cptcov= Number of total real effective simple dummies (fixed or time arying) effective (used as cptcoveff in other functions)
1.242 brouard 6338: * because they can be excluded from the model and real
6339: * if in the model but excluded because missing values, but how to get k from ij?*/
6340: for(j=ij+1; j<= cptcovt; j++){
6341: Tvaraff[j]=0;
6342: Tmodelind[j]=0;
6343: }
6344: for(j=ntveff+1; j<= cptcovt; j++){
6345: TmodelInvind[j]=0;
6346: }
6347: /* To be sorted */
6348: ;
6349: }
1.126 brouard 6350:
1.145 brouard 6351:
1.126 brouard 6352: /*********** Health Expectancies ****************/
6353:
1.235 brouard 6354: 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 6355:
6356: {
6357: /* Health expectancies, no variances */
1.329 brouard 6358: /* cij is the combination in the list of combination of dummy covariates */
6359: /* strstart is a string of time at start of computing */
1.164 brouard 6360: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 6361: int nhstepma, nstepma; /* Decreasing with age */
6362: double age, agelim, hf;
6363: double ***p3mat;
6364: double eip;
6365:
1.238 brouard 6366: /* pstamp(ficreseij); */
1.126 brouard 6367: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
6368: fprintf(ficreseij,"# Age");
6369: for(i=1; i<=nlstate;i++){
6370: for(j=1; j<=nlstate;j++){
6371: fprintf(ficreseij," e%1d%1d ",i,j);
6372: }
6373: fprintf(ficreseij," e%1d. ",i);
6374: }
6375: fprintf(ficreseij,"\n");
6376:
6377:
6378: if(estepm < stepm){
6379: printf ("Problem %d lower than %d\n",estepm, stepm);
6380: }
6381: else hstepm=estepm;
6382: /* We compute the life expectancy from trapezoids spaced every estepm months
6383: * This is mainly to measure the difference between two models: for example
6384: * if stepm=24 months pijx are given only every 2 years and by summing them
6385: * we are calculating an estimate of the Life Expectancy assuming a linear
6386: * progression in between and thus overestimating or underestimating according
6387: * to the curvature of the survival function. If, for the same date, we
6388: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6389: * to compare the new estimate of Life expectancy with the same linear
6390: * hypothesis. A more precise result, taking into account a more precise
6391: * curvature will be obtained if estepm is as small as stepm. */
6392:
6393: /* For example we decided to compute the life expectancy with the smallest unit */
6394: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6395: nhstepm is the number of hstepm from age to agelim
6396: nstepm is the number of stepm from age to agelin.
1.270 brouard 6397: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 6398: and note for a fixed period like estepm months */
6399: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6400: survival function given by stepm (the optimization length). Unfortunately it
6401: means that if the survival funtion is printed only each two years of age and if
6402: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6403: results. So we changed our mind and took the option of the best precision.
6404: */
6405: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6406:
6407: agelim=AGESUP;
6408: /* If stepm=6 months */
6409: /* Computed by stepm unit matrices, product of hstepm matrices, stored
6410: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
6411:
6412: /* nhstepm age range expressed in number of stepm */
6413: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6414: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6415: /* if (stepm >= YEARM) hstepm=1;*/
6416: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6417: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6418:
6419: for (age=bage; age<=fage; age ++){
6420: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6421: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6422: /* if (stepm >= YEARM) hstepm=1;*/
6423: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
6424:
6425: /* If stepm=6 months */
6426: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6427: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 brouard 6428: /* 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 6429: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 6430:
6431: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6432:
6433: printf("%d|",(int)age);fflush(stdout);
6434: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6435:
6436: /* Computing expectancies */
6437: for(i=1; i<=nlstate;i++)
6438: for(j=1; j<=nlstate;j++)
6439: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6440: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
6441:
6442: /* 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]);*/
6443:
6444: }
6445:
6446: fprintf(ficreseij,"%3.0f",age );
6447: for(i=1; i<=nlstate;i++){
6448: eip=0;
6449: for(j=1; j<=nlstate;j++){
6450: eip +=eij[i][j][(int)age];
6451: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
6452: }
6453: fprintf(ficreseij,"%9.4f", eip );
6454: }
6455: fprintf(ficreseij,"\n");
6456:
6457: }
6458: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6459: printf("\n");
6460: fprintf(ficlog,"\n");
6461:
6462: }
6463:
1.235 brouard 6464: 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 6465:
6466: {
6467: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6468: to initial status i, ei. .
1.126 brouard 6469: */
1.336 brouard 6470: /* Very time consuming function, but already optimized with precov */
1.126 brouard 6471: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6472: int nhstepma, nstepma; /* Decreasing with age */
6473: double age, agelim, hf;
6474: double ***p3matp, ***p3matm, ***varhe;
6475: double **dnewm,**doldm;
6476: double *xp, *xm;
6477: double **gp, **gm;
6478: double ***gradg, ***trgradg;
6479: int theta;
6480:
6481: double eip, vip;
6482:
6483: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6484: xp=vector(1,npar);
6485: xm=vector(1,npar);
6486: dnewm=matrix(1,nlstate*nlstate,1,npar);
6487: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6488:
6489: pstamp(ficresstdeij);
6490: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6491: fprintf(ficresstdeij,"# Age");
6492: for(i=1; i<=nlstate;i++){
6493: for(j=1; j<=nlstate;j++)
6494: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6495: fprintf(ficresstdeij," e%1d. ",i);
6496: }
6497: fprintf(ficresstdeij,"\n");
6498:
6499: pstamp(ficrescveij);
6500: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6501: fprintf(ficrescveij,"# Age");
6502: for(i=1; i<=nlstate;i++)
6503: for(j=1; j<=nlstate;j++){
6504: cptj= (j-1)*nlstate+i;
6505: for(i2=1; i2<=nlstate;i2++)
6506: for(j2=1; j2<=nlstate;j2++){
6507: cptj2= (j2-1)*nlstate+i2;
6508: if(cptj2 <= cptj)
6509: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6510: }
6511: }
6512: fprintf(ficrescveij,"\n");
6513:
6514: if(estepm < stepm){
6515: printf ("Problem %d lower than %d\n",estepm, stepm);
6516: }
6517: else hstepm=estepm;
6518: /* We compute the life expectancy from trapezoids spaced every estepm months
6519: * This is mainly to measure the difference between two models: for example
6520: * if stepm=24 months pijx are given only every 2 years and by summing them
6521: * we are calculating an estimate of the Life Expectancy assuming a linear
6522: * progression in between and thus overestimating or underestimating according
6523: * to the curvature of the survival function. If, for the same date, we
6524: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6525: * to compare the new estimate of Life expectancy with the same linear
6526: * hypothesis. A more precise result, taking into account a more precise
6527: * curvature will be obtained if estepm is as small as stepm. */
6528:
6529: /* For example we decided to compute the life expectancy with the smallest unit */
6530: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6531: nhstepm is the number of hstepm from age to agelim
6532: nstepm is the number of stepm from age to agelin.
6533: Look at hpijx to understand the reason of that which relies in memory size
6534: and note for a fixed period like estepm months */
6535: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6536: survival function given by stepm (the optimization length). Unfortunately it
6537: means that if the survival funtion is printed only each two years of age and if
6538: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6539: results. So we changed our mind and took the option of the best precision.
6540: */
6541: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6542:
6543: /* If stepm=6 months */
6544: /* nhstepm age range expressed in number of stepm */
6545: agelim=AGESUP;
6546: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6547: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6548: /* if (stepm >= YEARM) hstepm=1;*/
6549: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6550:
6551: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6552: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6553: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6554: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6555: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6556: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6557:
6558: for (age=bage; age<=fage; age ++){
6559: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6560: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6561: /* if (stepm >= YEARM) hstepm=1;*/
6562: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6563:
1.126 brouard 6564: /* If stepm=6 months */
6565: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6566: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6567:
6568: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6569:
1.126 brouard 6570: /* Computing Variances of health expectancies */
6571: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6572: decrease memory allocation */
6573: for(theta=1; theta <=npar; theta++){
6574: for(i=1; i<=npar; i++){
1.222 brouard 6575: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6576: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6577: }
1.235 brouard 6578: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6579: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6580:
1.126 brouard 6581: for(j=1; j<= nlstate; j++){
1.222 brouard 6582: for(i=1; i<=nlstate; i++){
6583: for(h=0; h<=nhstepm-1; h++){
6584: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6585: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6586: }
6587: }
1.126 brouard 6588: }
1.218 brouard 6589:
1.126 brouard 6590: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6591: for(h=0; h<=nhstepm-1; h++){
6592: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6593: }
1.126 brouard 6594: }/* End theta */
6595:
6596:
6597: for(h=0; h<=nhstepm-1; h++)
6598: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6599: for(theta=1; theta <=npar; theta++)
6600: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6601:
1.218 brouard 6602:
1.222 brouard 6603: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6604: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6605: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6606:
1.222 brouard 6607: printf("%d|",(int)age);fflush(stdout);
6608: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6609: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6610: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6611: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6612: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6613: for(ij=1;ij<=nlstate*nlstate;ij++)
6614: for(ji=1;ji<=nlstate*nlstate;ji++)
6615: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6616: }
6617: }
1.320 brouard 6618: /* if((int)age ==50){ */
6619: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6620: /* } */
1.126 brouard 6621: /* Computing expectancies */
1.235 brouard 6622: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6623: for(i=1; i<=nlstate;i++)
6624: for(j=1; j<=nlstate;j++)
1.222 brouard 6625: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6626: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6627:
1.222 brouard 6628: /* 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 6629:
1.222 brouard 6630: }
1.269 brouard 6631:
6632: /* Standard deviation of expectancies ij */
1.126 brouard 6633: fprintf(ficresstdeij,"%3.0f",age );
6634: for(i=1; i<=nlstate;i++){
6635: eip=0.;
6636: vip=0.;
6637: for(j=1; j<=nlstate;j++){
1.222 brouard 6638: eip += eij[i][j][(int)age];
6639: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6640: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6641: 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 6642: }
6643: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6644: }
6645: fprintf(ficresstdeij,"\n");
1.218 brouard 6646:
1.269 brouard 6647: /* Variance of expectancies ij */
1.126 brouard 6648: fprintf(ficrescveij,"%3.0f",age );
6649: for(i=1; i<=nlstate;i++)
6650: for(j=1; j<=nlstate;j++){
1.222 brouard 6651: cptj= (j-1)*nlstate+i;
6652: for(i2=1; i2<=nlstate;i2++)
6653: for(j2=1; j2<=nlstate;j2++){
6654: cptj2= (j2-1)*nlstate+i2;
6655: if(cptj2 <= cptj)
6656: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6657: }
1.126 brouard 6658: }
6659: fprintf(ficrescveij,"\n");
1.218 brouard 6660:
1.126 brouard 6661: }
6662: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6663: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6664: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6665: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6666: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6667: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6668: printf("\n");
6669: fprintf(ficlog,"\n");
1.218 brouard 6670:
1.126 brouard 6671: free_vector(xm,1,npar);
6672: free_vector(xp,1,npar);
6673: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6674: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6675: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6676: }
1.218 brouard 6677:
1.126 brouard 6678: /************ Variance ******************/
1.235 brouard 6679: 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 6680: {
1.279 brouard 6681: /** Variance of health expectancies
6682: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6683: * double **newm;
6684: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6685: */
1.218 brouard 6686:
6687: /* int movingaverage(); */
6688: double **dnewm,**doldm;
6689: double **dnewmp,**doldmp;
6690: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6691: int first=0;
1.218 brouard 6692: int k;
6693: double *xp;
1.279 brouard 6694: double **gp, **gm; /**< for var eij */
6695: double ***gradg, ***trgradg; /**< for var eij */
6696: double **gradgp, **trgradgp; /**< for var p point j */
6697: double *gpp, *gmp; /**< for var p point j */
6698: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6699: double ***p3mat;
6700: double age,agelim, hf;
6701: /* double ***mobaverage; */
6702: int theta;
6703: char digit[4];
6704: char digitp[25];
6705:
6706: char fileresprobmorprev[FILENAMELENGTH];
6707:
6708: if(popbased==1){
6709: if(mobilav!=0)
6710: strcpy(digitp,"-POPULBASED-MOBILAV_");
6711: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6712: }
6713: else
6714: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6715:
1.218 brouard 6716: /* if (mobilav!=0) { */
6717: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6718: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6719: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6720: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6721: /* } */
6722: /* } */
6723:
6724: strcpy(fileresprobmorprev,"PRMORPREV-");
6725: sprintf(digit,"%-d",ij);
6726: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6727: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6728: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6729: strcat(fileresprobmorprev,fileresu);
6730: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6731: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6732: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6733: }
6734: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6735: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6736: pstamp(ficresprobmorprev);
6737: 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 6738: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
1.337 brouard 6739:
6740: /* We use TinvDoQresult[nres][resultmodel[nres][j] we sort according to the equation model and the resultline: it is a choice */
6741: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ /\* To be done*\/ */
6742: /* fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
6743: /* } */
6744: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */ /* To be done*/
6745: fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 6746: }
1.337 brouard 6747: /* for(j=1;j<=cptcoveff;j++) */
6748: /* fprintf(ficresprobmorprev," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,TnsdVar[Tvaraff[j]])]); */
1.238 brouard 6749: fprintf(ficresprobmorprev,"\n");
6750:
1.218 brouard 6751: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6752: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6753: fprintf(ficresprobmorprev," p.%-d SE",j);
6754: for(i=1; i<=nlstate;i++)
6755: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6756: }
6757: fprintf(ficresprobmorprev,"\n");
6758:
6759: fprintf(ficgp,"\n# Routine varevsij");
6760: fprintf(ficgp,"\nunset title \n");
6761: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6762: 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");
6763: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6764:
1.218 brouard 6765: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6766: pstamp(ficresvij);
6767: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6768: if(popbased==1)
6769: 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);
6770: else
6771: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6772: fprintf(ficresvij,"# Age");
6773: for(i=1; i<=nlstate;i++)
6774: for(j=1; j<=nlstate;j++)
6775: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6776: fprintf(ficresvij,"\n");
6777:
6778: xp=vector(1,npar);
6779: dnewm=matrix(1,nlstate,1,npar);
6780: doldm=matrix(1,nlstate,1,nlstate);
6781: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6782: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6783:
6784: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6785: gpp=vector(nlstate+1,nlstate+ndeath);
6786: gmp=vector(nlstate+1,nlstate+ndeath);
6787: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6788:
1.218 brouard 6789: if(estepm < stepm){
6790: printf ("Problem %d lower than %d\n",estepm, stepm);
6791: }
6792: else hstepm=estepm;
6793: /* For example we decided to compute the life expectancy with the smallest unit */
6794: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6795: nhstepm is the number of hstepm from age to agelim
6796: nstepm is the number of stepm from age to agelim.
6797: Look at function hpijx to understand why because of memory size limitations,
6798: we decided (b) to get a life expectancy respecting the most precise curvature of the
6799: survival function given by stepm (the optimization length). Unfortunately it
6800: means that if the survival funtion is printed every two years of age and if
6801: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6802: results. So we changed our mind and took the option of the best precision.
6803: */
6804: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6805: agelim = AGESUP;
6806: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6807: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6808: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6809: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6810: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6811: gp=matrix(0,nhstepm,1,nlstate);
6812: gm=matrix(0,nhstepm,1,nlstate);
6813:
6814:
6815: for(theta=1; theta <=npar; theta++){
6816: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6817: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6818: }
1.279 brouard 6819: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6820: * returns into prlim .
1.288 brouard 6821: */
1.242 brouard 6822: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6823:
6824: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6825: if (popbased==1) {
6826: if(mobilav ==0){
6827: for(i=1; i<=nlstate;i++)
6828: prlim[i][i]=probs[(int)age][i][ij];
6829: }else{ /* mobilav */
6830: for(i=1; i<=nlstate;i++)
6831: prlim[i][i]=mobaverage[(int)age][i][ij];
6832: }
6833: }
1.295 brouard 6834: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6835: */
6836: 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 6837: /**< 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 6838: * at horizon h in state j including mortality.
6839: */
1.218 brouard 6840: for(j=1; j<= nlstate; j++){
6841: for(h=0; h<=nhstepm; h++){
6842: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6843: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6844: }
6845: }
1.279 brouard 6846: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6847: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6848: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6849: */
6850: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6851: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6852: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6853: }
6854:
6855: /* Again with minus shift */
1.218 brouard 6856:
6857: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6858: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6859:
1.242 brouard 6860: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6861:
6862: if (popbased==1) {
6863: if(mobilav ==0){
6864: for(i=1; i<=nlstate;i++)
6865: prlim[i][i]=probs[(int)age][i][ij];
6866: }else{ /* mobilav */
6867: for(i=1; i<=nlstate;i++)
6868: prlim[i][i]=mobaverage[(int)age][i][ij];
6869: }
6870: }
6871:
1.235 brouard 6872: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6873:
6874: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6875: for(h=0; h<=nhstepm; h++){
6876: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6877: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6878: }
6879: }
6880: /* This for computing probability of death (h=1 means
6881: computed over hstepm matrices product = hstepm*stepm months)
6882: as a weighted average of prlim.
6883: */
6884: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6885: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6886: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6887: }
1.279 brouard 6888: /* end shifting computations */
6889:
6890: /**< Computing gradient matrix at horizon h
6891: */
1.218 brouard 6892: for(j=1; j<= nlstate; j++) /* vareij */
6893: for(h=0; h<=nhstepm; h++){
6894: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6895: }
1.279 brouard 6896: /**< Gradient of overall mortality p.3 (or p.j)
6897: */
6898: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6899: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6900: }
6901:
6902: } /* End theta */
1.279 brouard 6903:
6904: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6905: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6906:
6907: for(h=0; h<=nhstepm; h++) /* veij */
6908: for(j=1; j<=nlstate;j++)
6909: for(theta=1; theta <=npar; theta++)
6910: trgradg[h][j][theta]=gradg[h][theta][j];
6911:
6912: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6913: for(theta=1; theta <=npar; theta++)
6914: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6915: /**< as well as its transposed matrix
6916: */
1.218 brouard 6917:
6918: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6919: for(i=1;i<=nlstate;i++)
6920: for(j=1;j<=nlstate;j++)
6921: vareij[i][j][(int)age] =0.;
1.279 brouard 6922:
6923: /* Computing trgradg by matcov by gradg at age and summing over h
6924: * and k (nhstepm) formula 15 of article
6925: * Lievre-Brouard-Heathcote
6926: */
6927:
1.218 brouard 6928: for(h=0;h<=nhstepm;h++){
6929: for(k=0;k<=nhstepm;k++){
6930: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6931: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6932: for(i=1;i<=nlstate;i++)
6933: for(j=1;j<=nlstate;j++)
6934: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6935: }
6936: }
6937:
1.279 brouard 6938: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6939: * p.j overall mortality formula 49 but computed directly because
6940: * we compute the grad (wix pijx) instead of grad (pijx),even if
6941: * wix is independent of theta.
6942: */
1.218 brouard 6943: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6944: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6945: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6946: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6947: varppt[j][i]=doldmp[j][i];
6948: /* end ppptj */
6949: /* x centered again */
6950:
1.242 brouard 6951: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6952:
6953: if (popbased==1) {
6954: if(mobilav ==0){
6955: for(i=1; i<=nlstate;i++)
6956: prlim[i][i]=probs[(int)age][i][ij];
6957: }else{ /* mobilav */
6958: for(i=1; i<=nlstate;i++)
6959: prlim[i][i]=mobaverage[(int)age][i][ij];
6960: }
6961: }
6962:
6963: /* This for computing probability of death (h=1 means
6964: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6965: as a weighted average of prlim.
6966: */
1.235 brouard 6967: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6968: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6969: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6970: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6971: }
6972: /* end probability of death */
6973:
6974: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6975: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6976: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6977: for(i=1; i<=nlstate;i++){
6978: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6979: }
6980: }
6981: fprintf(ficresprobmorprev,"\n");
6982:
6983: fprintf(ficresvij,"%.0f ",age );
6984: for(i=1; i<=nlstate;i++)
6985: for(j=1; j<=nlstate;j++){
6986: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6987: }
6988: fprintf(ficresvij,"\n");
6989: free_matrix(gp,0,nhstepm,1,nlstate);
6990: free_matrix(gm,0,nhstepm,1,nlstate);
6991: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6992: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6993: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6994: } /* End age */
6995: free_vector(gpp,nlstate+1,nlstate+ndeath);
6996: free_vector(gmp,nlstate+1,nlstate+ndeath);
6997: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6998: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6999: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
7000: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
7001: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
7002: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
7003: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
7004: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
7005: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
7006: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
7007: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
7008: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
7009: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
7010: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
7011: 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);
7012: /* 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 7013: */
1.218 brouard 7014: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
7015: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 7016:
1.218 brouard 7017: free_vector(xp,1,npar);
7018: free_matrix(doldm,1,nlstate,1,nlstate);
7019: free_matrix(dnewm,1,nlstate,1,npar);
7020: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
7021: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
7022: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
7023: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7024: fclose(ficresprobmorprev);
7025: fflush(ficgp);
7026: fflush(fichtm);
7027: } /* end varevsij */
1.126 brouard 7028:
7029: /************ Variance of prevlim ******************/
1.269 brouard 7030: 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 7031: {
1.205 brouard 7032: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 7033: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 7034:
1.268 brouard 7035: double **dnewmpar,**doldm;
1.126 brouard 7036: int i, j, nhstepm, hstepm;
7037: double *xp;
7038: double *gp, *gm;
7039: double **gradg, **trgradg;
1.208 brouard 7040: double **mgm, **mgp;
1.126 brouard 7041: double age,agelim;
7042: int theta;
7043:
7044: pstamp(ficresvpl);
1.288 brouard 7045: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 7046: fprintf(ficresvpl,"# Age ");
7047: if(nresult >=1)
7048: fprintf(ficresvpl," Result# ");
1.126 brouard 7049: for(i=1; i<=nlstate;i++)
7050: fprintf(ficresvpl," %1d-%1d",i,i);
7051: fprintf(ficresvpl,"\n");
7052:
7053: xp=vector(1,npar);
1.268 brouard 7054: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 7055: doldm=matrix(1,nlstate,1,nlstate);
7056:
7057: hstepm=1*YEARM; /* Every year of age */
7058: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
7059: agelim = AGESUP;
7060: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
7061: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
7062: if (stepm >= YEARM) hstepm=1;
7063: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
7064: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 7065: mgp=matrix(1,npar,1,nlstate);
7066: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 7067: gp=vector(1,nlstate);
7068: gm=vector(1,nlstate);
7069:
7070: for(theta=1; theta <=npar; theta++){
7071: for(i=1; i<=npar; i++){ /* Computes gradient */
7072: xp[i] = x[i] + (i==theta ?delti[theta]:0);
7073: }
1.288 brouard 7074: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
7075: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
7076: /* else */
7077: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 7078: for(i=1;i<=nlstate;i++){
1.126 brouard 7079: gp[i] = prlim[i][i];
1.208 brouard 7080: mgp[theta][i] = prlim[i][i];
7081: }
1.126 brouard 7082: for(i=1; i<=npar; i++) /* Computes gradient */
7083: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 7084: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
7085: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
7086: /* else */
7087: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 7088: for(i=1;i<=nlstate;i++){
1.126 brouard 7089: gm[i] = prlim[i][i];
1.208 brouard 7090: mgm[theta][i] = prlim[i][i];
7091: }
1.126 brouard 7092: for(i=1;i<=nlstate;i++)
7093: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 7094: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 7095: } /* End theta */
7096:
7097: trgradg =matrix(1,nlstate,1,npar);
7098:
7099: for(j=1; j<=nlstate;j++)
7100: for(theta=1; theta <=npar; theta++)
7101: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 7102: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7103: /* printf("\nmgm mgp %d ",(int)age); */
7104: /* for(j=1; j<=nlstate;j++){ */
7105: /* printf(" %d ",j); */
7106: /* for(theta=1; theta <=npar; theta++) */
7107: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7108: /* printf("\n "); */
7109: /* } */
7110: /* } */
7111: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7112: /* printf("\n gradg %d ",(int)age); */
7113: /* for(j=1; j<=nlstate;j++){ */
7114: /* printf("%d ",j); */
7115: /* for(theta=1; theta <=npar; theta++) */
7116: /* printf("%d %lf ",theta,gradg[theta][j]); */
7117: /* printf("\n "); */
7118: /* } */
7119: /* } */
1.126 brouard 7120:
7121: for(i=1;i<=nlstate;i++)
7122: varpl[i][(int)age] =0.;
1.209 brouard 7123: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 7124: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7125: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 7126: }else{
1.268 brouard 7127: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7128: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 7129: }
1.126 brouard 7130: for(i=1;i<=nlstate;i++)
7131: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7132:
7133: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 7134: if(nresult >=1)
7135: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 7136: for(i=1; i<=nlstate;i++){
1.126 brouard 7137: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 7138: /* for(j=1;j<=nlstate;j++) */
7139: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
7140: }
1.126 brouard 7141: fprintf(ficresvpl,"\n");
7142: free_vector(gp,1,nlstate);
7143: free_vector(gm,1,nlstate);
1.208 brouard 7144: free_matrix(mgm,1,npar,1,nlstate);
7145: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 7146: free_matrix(gradg,1,npar,1,nlstate);
7147: free_matrix(trgradg,1,nlstate,1,npar);
7148: } /* End age */
7149:
7150: free_vector(xp,1,npar);
7151: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 7152: free_matrix(dnewmpar,1,nlstate,1,nlstate);
7153:
7154: }
7155:
7156:
7157: /************ Variance of backprevalence limit ******************/
1.269 brouard 7158: 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 7159: {
7160: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
7161: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
7162:
7163: double **dnewmpar,**doldm;
7164: int i, j, nhstepm, hstepm;
7165: double *xp;
7166: double *gp, *gm;
7167: double **gradg, **trgradg;
7168: double **mgm, **mgp;
7169: double age,agelim;
7170: int theta;
7171:
7172: pstamp(ficresvbl);
7173: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
7174: fprintf(ficresvbl,"# Age ");
7175: if(nresult >=1)
7176: fprintf(ficresvbl," Result# ");
7177: for(i=1; i<=nlstate;i++)
7178: fprintf(ficresvbl," %1d-%1d",i,i);
7179: fprintf(ficresvbl,"\n");
7180:
7181: xp=vector(1,npar);
7182: dnewmpar=matrix(1,nlstate,1,npar);
7183: doldm=matrix(1,nlstate,1,nlstate);
7184:
7185: hstepm=1*YEARM; /* Every year of age */
7186: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
7187: agelim = AGEINF;
7188: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
7189: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
7190: if (stepm >= YEARM) hstepm=1;
7191: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
7192: gradg=matrix(1,npar,1,nlstate);
7193: mgp=matrix(1,npar,1,nlstate);
7194: mgm=matrix(1,npar,1,nlstate);
7195: gp=vector(1,nlstate);
7196: gm=vector(1,nlstate);
7197:
7198: for(theta=1; theta <=npar; theta++){
7199: for(i=1; i<=npar; i++){ /* Computes gradient */
7200: xp[i] = x[i] + (i==theta ?delti[theta]:0);
7201: }
7202: if(mobilavproj > 0 )
7203: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7204: else
7205: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7206: for(i=1;i<=nlstate;i++){
7207: gp[i] = bprlim[i][i];
7208: mgp[theta][i] = bprlim[i][i];
7209: }
7210: for(i=1; i<=npar; i++) /* Computes gradient */
7211: xp[i] = x[i] - (i==theta ?delti[theta]:0);
7212: if(mobilavproj > 0 )
7213: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7214: else
7215: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7216: for(i=1;i<=nlstate;i++){
7217: gm[i] = bprlim[i][i];
7218: mgm[theta][i] = bprlim[i][i];
7219: }
7220: for(i=1;i<=nlstate;i++)
7221: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
7222: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
7223: } /* End theta */
7224:
7225: trgradg =matrix(1,nlstate,1,npar);
7226:
7227: for(j=1; j<=nlstate;j++)
7228: for(theta=1; theta <=npar; theta++)
7229: trgradg[j][theta]=gradg[theta][j];
7230: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7231: /* printf("\nmgm mgp %d ",(int)age); */
7232: /* for(j=1; j<=nlstate;j++){ */
7233: /* printf(" %d ",j); */
7234: /* for(theta=1; theta <=npar; theta++) */
7235: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7236: /* printf("\n "); */
7237: /* } */
7238: /* } */
7239: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7240: /* printf("\n gradg %d ",(int)age); */
7241: /* for(j=1; j<=nlstate;j++){ */
7242: /* printf("%d ",j); */
7243: /* for(theta=1; theta <=npar; theta++) */
7244: /* printf("%d %lf ",theta,gradg[theta][j]); */
7245: /* printf("\n "); */
7246: /* } */
7247: /* } */
7248:
7249: for(i=1;i<=nlstate;i++)
7250: varbpl[i][(int)age] =0.;
7251: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
7252: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7253: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7254: }else{
7255: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7256: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7257: }
7258: for(i=1;i<=nlstate;i++)
7259: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7260:
7261: fprintf(ficresvbl,"%.0f ",age );
7262: if(nresult >=1)
7263: fprintf(ficresvbl,"%d ",nres );
7264: for(i=1; i<=nlstate;i++)
7265: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
7266: fprintf(ficresvbl,"\n");
7267: free_vector(gp,1,nlstate);
7268: free_vector(gm,1,nlstate);
7269: free_matrix(mgm,1,npar,1,nlstate);
7270: free_matrix(mgp,1,npar,1,nlstate);
7271: free_matrix(gradg,1,npar,1,nlstate);
7272: free_matrix(trgradg,1,nlstate,1,npar);
7273: } /* End age */
7274:
7275: free_vector(xp,1,npar);
7276: free_matrix(doldm,1,nlstate,1,npar);
7277: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 7278:
7279: }
7280:
7281: /************ Variance of one-step probabilities ******************/
7282: 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 7283: {
7284: int i, j=0, k1, l1, tj;
7285: int k2, l2, j1, z1;
7286: int k=0, l;
7287: int first=1, first1, first2;
1.326 brouard 7288: int nres=0; /* New */
1.222 brouard 7289: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
7290: double **dnewm,**doldm;
7291: double *xp;
7292: double *gp, *gm;
7293: double **gradg, **trgradg;
7294: double **mu;
7295: double age, cov[NCOVMAX+1];
7296: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
7297: int theta;
7298: char fileresprob[FILENAMELENGTH];
7299: char fileresprobcov[FILENAMELENGTH];
7300: char fileresprobcor[FILENAMELENGTH];
7301: double ***varpij;
7302:
7303: strcpy(fileresprob,"PROB_");
7304: strcat(fileresprob,fileres);
7305: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
7306: printf("Problem with resultfile: %s\n", fileresprob);
7307: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
7308: }
7309: strcpy(fileresprobcov,"PROBCOV_");
7310: strcat(fileresprobcov,fileresu);
7311: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
7312: printf("Problem with resultfile: %s\n", fileresprobcov);
7313: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
7314: }
7315: strcpy(fileresprobcor,"PROBCOR_");
7316: strcat(fileresprobcor,fileresu);
7317: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
7318: printf("Problem with resultfile: %s\n", fileresprobcor);
7319: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
7320: }
7321: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7322: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7323: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7324: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7325: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7326: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7327: pstamp(ficresprob);
7328: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
7329: fprintf(ficresprob,"# Age");
7330: pstamp(ficresprobcov);
7331: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
7332: fprintf(ficresprobcov,"# Age");
7333: pstamp(ficresprobcor);
7334: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
7335: fprintf(ficresprobcor,"# Age");
1.126 brouard 7336:
7337:
1.222 brouard 7338: for(i=1; i<=nlstate;i++)
7339: for(j=1; j<=(nlstate+ndeath);j++){
7340: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
7341: fprintf(ficresprobcov," p%1d-%1d ",i,j);
7342: fprintf(ficresprobcor," p%1d-%1d ",i,j);
7343: }
7344: /* fprintf(ficresprob,"\n");
7345: fprintf(ficresprobcov,"\n");
7346: fprintf(ficresprobcor,"\n");
7347: */
7348: xp=vector(1,npar);
7349: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7350: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7351: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
7352: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
7353: first=1;
7354: fprintf(ficgp,"\n# Routine varprob");
7355: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
7356: fprintf(fichtm,"\n");
7357:
1.288 brouard 7358: 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 7359: 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);
7360: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 7361: and drawn. It helps understanding how is the covariance between two incidences.\
7362: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 7363: 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 7364: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
7365: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
7366: standard deviations wide on each axis. <br>\
7367: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
7368: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
7369: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
7370:
1.222 brouard 7371: cov[1]=1;
7372: /* tj=cptcoveff; */
1.225 brouard 7373: tj = (int) pow(2,cptcoveff);
1.222 brouard 7374: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
7375: j1=0;
1.332 brouard 7376:
7377: for(nres=1;nres <=nresult; nres++){ /* For each resultline */
7378: for(j1=1; j1<=tj;j1++){ /* For any combination of dummy covariates, fixed and varying */
1.342 ! brouard 7379: /* 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 7380: if(tj != 1 && TKresult[nres]!= j1)
7381: continue;
7382:
7383: /* for(j1=1; j1<=tj;j1++){ /\* For each valid combination of covariates or only once*\/ */
7384: /* for(nres=1;nres <=1; nres++){ /\* For each resultline *\/ */
7385: /* /\* for(nres=1;nres <=nresult; nres++){ /\\* For each resultline *\\/ *\/ */
1.222 brouard 7386: if (cptcovn>0) {
1.334 brouard 7387: fprintf(ficresprob, "\n#********** Variable ");
7388: fprintf(ficresprobcov, "\n#********** Variable ");
7389: fprintf(ficgp, "\n#********** Variable ");
7390: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
7391: fprintf(ficresprobcor, "\n#********** Variable ");
7392:
7393: /* Including quantitative variables of the resultline to be done */
7394: for (z1=1; z1<=cptcovs; z1++){ /* Loop on each variable of this resultline */
1.338 brouard 7395: printf("Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model);
7396: fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model);
7397: /* 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 7398: if(Dummy[modelresult[nres][z1]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to z1 in resultline */
7399: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed referenced to model equation */
7400: 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 */
7401: 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 */
7402: 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 */
7403: 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 */
7404: 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 */
7405: fprintf(ficresprob,"fixed ");
7406: fprintf(ficresprobcov,"fixed ");
7407: fprintf(ficgp,"fixed ");
7408: fprintf(fichtmcov,"fixed ");
7409: fprintf(ficresprobcor,"fixed ");
7410: }else{
7411: fprintf(ficresprob,"varyi ");
7412: fprintf(ficresprobcov,"varyi ");
7413: fprintf(ficgp,"varyi ");
7414: fprintf(fichtmcov,"varyi ");
7415: fprintf(ficresprobcor,"varyi ");
7416: }
7417: }else if(Dummy[modelresult[nres][z1]]==1){ /* Quanti variable */
7418: /* For each selected (single) quantitative value */
1.337 brouard 7419: fprintf(ficresprob," V%d=%lg ",Tvqresult[nres][z1],Tqresult[nres][z1]);
1.334 brouard 7420: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed */
7421: fprintf(ficresprob,"fixed ");
7422: fprintf(ficresprobcov,"fixed ");
7423: fprintf(ficgp,"fixed ");
7424: fprintf(fichtmcov,"fixed ");
7425: fprintf(ficresprobcor,"fixed ");
7426: }else{
7427: fprintf(ficresprob,"varyi ");
7428: fprintf(ficresprobcov,"varyi ");
7429: fprintf(ficgp,"varyi ");
7430: fprintf(fichtmcov,"varyi ");
7431: fprintf(ficresprobcor,"varyi ");
7432: }
7433: }else{
7434: 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 */
7435: 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 */
7436: exit(1);
7437: }
7438: } /* End loop on variable of this resultline */
7439: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]); */
1.222 brouard 7440: fprintf(ficresprob, "**********\n#\n");
7441: fprintf(ficresprobcov, "**********\n#\n");
7442: fprintf(ficgp, "**********\n#\n");
7443: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
7444: fprintf(ficresprobcor, "**********\n#");
7445: if(invalidvarcomb[j1]){
7446: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
7447: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
7448: continue;
7449: }
7450: }
7451: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
7452: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7453: gp=vector(1,(nlstate)*(nlstate+ndeath));
7454: gm=vector(1,(nlstate)*(nlstate+ndeath));
1.334 brouard 7455: for (age=bage; age<=fage; age ++){ /* Fo each age we feed the model equation with covariates, using precov as in hpxij() ? */
1.222 brouard 7456: cov[2]=age;
7457: if(nagesqr==1)
7458: cov[3]= age*age;
1.334 brouard 7459: /* New code end of combination but for each resultline */
7460: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
7461: if(Typevar[k1]==1){ /* A product with age */
7462: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.326 brouard 7463: }else{
1.334 brouard 7464: cov[2+nagesqr+k1]=precov[nres][k1];
1.326 brouard 7465: }
1.334 brouard 7466: }/* End of loop on model equation */
7467: /* Old code */
7468: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
7469: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; *\/ */
7470: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
7471: /* /\* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates *\/ */
7472: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TnsdVar[TvarsD[k]])]; */
7473: /* /\*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*\//\* j1 1 2 3 4 */
7474: /* * 1 1 1 1 1 */
7475: /* * 2 2 1 1 1 */
7476: /* * 3 1 2 1 1 */
7477: /* *\/ */
7478: /* /\* nbcode[1][1]=0 nbcode[1][2]=1;*\/ */
7479: /* } */
7480: /* /\* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
7481: /* /\* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] *\/ */
7482: /* /\*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; *\/ */
7483: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
7484: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
7485: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,TnsdVar[Tvar[Tage[k]]])]*cov[2]; */
7486: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7487: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
7488: /* 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]); */
7489: /* /\* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\\* Using the mean of quantitative variable Tvar[Tage[k]] /\\* Tqresult[nres][k]; *\\/ *\/ */
7490: /* /\* exit(1); *\/ */
7491: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
7492: /* } */
7493: /* /\* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7494: /* } */
7495: /* for (k=1; k<=cptcovprod;k++){/\* For product without age *\/ */
7496: /* if(Dummy[Tvard[k][1]]==0){ */
7497: /* if(Dummy[Tvard[k][2]]==0){ */
7498: /* 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]])]; */
7499: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7500: /* }else{ /\* Should we use the mean of the quantitative variables? *\/ */
7501: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * Tqresult[nres][resultmodel[nres][k]]; */
7502: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
7503: /* } */
7504: /* }else{ */
7505: /* if(Dummy[Tvard[k][2]]==0){ */
7506: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][TnsdVar[Tvard[k][1]]]; */
7507: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
7508: /* }else{ */
7509: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][TnsdVar[Tvard[k][1]]]* Tqinvresult[nres][TnsdVar[Tvard[k][2]]]; */
7510: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
7511: /* } */
7512: /* } */
7513: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7514: /* } */
1.326 brouard 7515: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 7516: for(theta=1; theta <=npar; theta++){
7517: for(i=1; i<=npar; i++)
7518: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7519:
1.222 brouard 7520: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7521:
1.222 brouard 7522: k=0;
7523: for(i=1; i<= (nlstate); i++){
7524: for(j=1; j<=(nlstate+ndeath);j++){
7525: k=k+1;
7526: gp[k]=pmmij[i][j];
7527: }
7528: }
1.220 brouard 7529:
1.222 brouard 7530: for(i=1; i<=npar; i++)
7531: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7532:
1.222 brouard 7533: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7534: k=0;
7535: for(i=1; i<=(nlstate); i++){
7536: for(j=1; j<=(nlstate+ndeath);j++){
7537: k=k+1;
7538: gm[k]=pmmij[i][j];
7539: }
7540: }
1.220 brouard 7541:
1.222 brouard 7542: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7543: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7544: }
1.126 brouard 7545:
1.222 brouard 7546: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7547: for(theta=1; theta <=npar; theta++)
7548: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7549:
1.222 brouard 7550: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7551: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7552:
1.222 brouard 7553: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7554:
1.222 brouard 7555: k=0;
7556: for(i=1; i<=(nlstate); i++){
7557: for(j=1; j<=(nlstate+ndeath);j++){
7558: k=k+1;
7559: mu[k][(int) age]=pmmij[i][j];
7560: }
7561: }
7562: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7563: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7564: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7565:
1.222 brouard 7566: /*printf("\n%d ",(int)age);
7567: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7568: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7569: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7570: }*/
1.220 brouard 7571:
1.222 brouard 7572: fprintf(ficresprob,"\n%d ",(int)age);
7573: fprintf(ficresprobcov,"\n%d ",(int)age);
7574: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7575:
1.222 brouard 7576: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7577: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7578: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7579: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7580: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7581: }
7582: i=0;
7583: for (k=1; k<=(nlstate);k++){
7584: for (l=1; l<=(nlstate+ndeath);l++){
7585: i++;
7586: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7587: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7588: for (j=1; j<=i;j++){
7589: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7590: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7591: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7592: }
7593: }
7594: }/* end of loop for state */
7595: } /* end of loop for age */
7596: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7597: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7598: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7599: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7600:
7601: /* Confidence intervalle of pij */
7602: /*
7603: fprintf(ficgp,"\nunset parametric;unset label");
7604: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7605: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7606: 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);
7607: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7608: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7609: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7610: */
7611:
7612: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7613: first1=1;first2=2;
7614: for (k2=1; k2<=(nlstate);k2++){
7615: for (l2=1; l2<=(nlstate+ndeath);l2++){
7616: if(l2==k2) continue;
7617: j=(k2-1)*(nlstate+ndeath)+l2;
7618: for (k1=1; k1<=(nlstate);k1++){
7619: for (l1=1; l1<=(nlstate+ndeath);l1++){
7620: if(l1==k1) continue;
7621: i=(k1-1)*(nlstate+ndeath)+l1;
7622: if(i<=j) continue;
7623: for (age=bage; age<=fage; age ++){
7624: if ((int)age %5==0){
7625: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7626: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7627: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7628: mu1=mu[i][(int) age]/stepm*YEARM ;
7629: mu2=mu[j][(int) age]/stepm*YEARM;
7630: c12=cv12/sqrt(v1*v2);
7631: /* Computing eigen value of matrix of covariance */
7632: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7633: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7634: if ((lc2 <0) || (lc1 <0) ){
7635: if(first2==1){
7636: first1=0;
7637: 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);
7638: }
7639: 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);
7640: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7641: /* lc2=fabs(lc2); */
7642: }
1.220 brouard 7643:
1.222 brouard 7644: /* Eigen vectors */
1.280 brouard 7645: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7646: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7647: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7648: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7649: }else
7650: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7651: /*v21=sqrt(1.-v11*v11); *//* error */
7652: v21=(lc1-v1)/cv12*v11;
7653: v12=-v21;
7654: v22=v11;
7655: tnalp=v21/v11;
7656: if(first1==1){
7657: first1=0;
7658: 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);
7659: }
7660: 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);
7661: /*printf(fignu*/
7662: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7663: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7664: if(first==1){
7665: first=0;
7666: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7667: fprintf(ficgp,"\nset parametric;unset label");
7668: 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);
7669: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7670: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7671: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7672: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7673: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7674: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7675: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7676: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7677: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7678: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7679: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7680: 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 7681: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7682: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7683: }else{
7684: first=0;
7685: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7686: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7687: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7688: 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 7689: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7690: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7691: }/* if first */
7692: } /* age mod 5 */
7693: } /* end loop age */
7694: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7695: first=1;
7696: } /*l12 */
7697: } /* k12 */
7698: } /*l1 */
7699: }/* k1 */
1.332 brouard 7700: } /* loop on combination of covariates j1 */
1.326 brouard 7701: } /* loop on nres */
1.222 brouard 7702: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7703: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7704: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7705: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7706: free_vector(xp,1,npar);
7707: fclose(ficresprob);
7708: fclose(ficresprobcov);
7709: fclose(ficresprobcor);
7710: fflush(ficgp);
7711: fflush(fichtmcov);
7712: }
1.126 brouard 7713:
7714:
7715: /******************* Printing html file ***********/
1.201 brouard 7716: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7717: int lastpass, int stepm, int weightopt, char model[],\
7718: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7719: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7720: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7721: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7722: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7723: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7724: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7725: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7726: </ul>");
1.319 brouard 7727: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7728: /* </ul>", model); */
1.214 brouard 7729: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7730: 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",
7731: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
1.332 brouard 7732: 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 7733: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7734: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7735: fprintf(fichtm,"\
7736: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7737: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7738: fprintf(fichtm,"\
1.217 brouard 7739: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7740: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7741: fprintf(fichtm,"\
1.288 brouard 7742: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7743: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7744: fprintf(fichtm,"\
1.288 brouard 7745: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7746: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7747: fprintf(fichtm,"\
1.211 brouard 7748: - (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 7749: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7750: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7751: if(prevfcast==1){
7752: fprintf(fichtm,"\
7753: - Prevalence projections by age and states: \
1.201 brouard 7754: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7755: }
1.126 brouard 7756:
7757:
1.225 brouard 7758: m=pow(2,cptcoveff);
1.222 brouard 7759: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7760:
1.317 brouard 7761: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7762:
7763: jj1=0;
7764:
7765: fprintf(fichtm," \n<ul>");
1.337 brouard 7766: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7767: /* k1=nres; */
1.338 brouard 7768: k1=TKresult[nres];
7769: if(TKresult[nres]==0)k1=1; /* To be checked for no result */
1.337 brouard 7770: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
7771: /* if(m != 1 && TKresult[nres]!= k1) */
7772: /* continue; */
1.264 brouard 7773: jj1++;
7774: if (cptcovn > 0) {
7775: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
1.337 brouard 7776: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
7777: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7778: }
1.337 brouard 7779: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
7780: /* fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
7781: /* } */
7782: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7783: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7784: /* } */
1.264 brouard 7785: fprintf(fichtm,"\">");
7786:
7787: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7788: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 7789: for (cpt=1; cpt<=cptcovs;cpt++){
7790: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7791: }
1.337 brouard 7792: /* fprintf(fichtm,"************ Results for covariates"); */
7793: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
7794: /* fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
7795: /* } */
7796: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7797: /* fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7798: /* } */
1.264 brouard 7799: if(invalidvarcomb[k1]){
7800: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7801: continue;
7802: }
7803: fprintf(fichtm,"</a></li>");
7804: } /* cptcovn >0 */
7805: }
1.317 brouard 7806: fprintf(fichtm," \n</ul>");
1.264 brouard 7807:
1.222 brouard 7808: jj1=0;
1.237 brouard 7809:
1.337 brouard 7810: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7811: /* k1=nres; */
1.338 brouard 7812: k1=TKresult[nres];
7813: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 7814: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
7815: /* if(m != 1 && TKresult[nres]!= k1) */
7816: /* continue; */
1.220 brouard 7817:
1.222 brouard 7818: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7819: jj1++;
7820: if (cptcovn > 0) {
1.264 brouard 7821: fprintf(fichtm,"\n<p><a name=\"rescov");
1.337 brouard 7822: for (cpt=1; cpt<=cptcovs;cpt++){
7823: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7824: }
1.337 brouard 7825: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7826: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7827: /* } */
1.264 brouard 7828: fprintf(fichtm,"\"</a>");
7829:
1.222 brouard 7830: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 7831: for (cpt=1; cpt<=cptcovs;cpt++){
7832: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
7833: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 7834: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7835: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7836: }
1.230 brouard 7837: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.338 brouard 7838: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7839: if(invalidvarcomb[k1]){
7840: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7841: printf("\nCombination (%d) ignored because no cases \n",k1);
7842: continue;
7843: }
7844: }
7845: /* aij, bij */
1.259 brouard 7846: 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 7847: <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 7848: /* Pij */
1.241 brouard 7849: 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> \
7850: <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 7851: /* Quasi-incidences */
7852: 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 7853: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7854: 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 7855: 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> \
7856: <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 7857: /* Survival functions (period) in state j */
7858: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7859: 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);
7860: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7861: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7862: }
7863: /* State specific survival functions (period) */
7864: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7865: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7866: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.329 brouard 7867: <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);
7868: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7869: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7870: }
1.288 brouard 7871: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7872: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7873: 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 7874: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
1.329 brouard 7875: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7876: }
1.296 brouard 7877: if(prevbcast==1){
1.288 brouard 7878: /* Backward prevalence in each health state */
1.222 brouard 7879: for(cpt=1; cpt<=nlstate;cpt++){
1.338 brouard 7880: 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);
7881: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJB_"),subdirf2(optionfilefiname,"PIJB_"));
7882: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.222 brouard 7883: }
1.217 brouard 7884: }
1.222 brouard 7885: if(prevfcast==1){
1.288 brouard 7886: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7887: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7888: 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);
7889: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7890: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7891: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7892: }
7893: }
1.296 brouard 7894: if(prevbcast==1){
1.268 brouard 7895: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7896: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7897: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7898: 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 \
7899: 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 7900: 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);
7901: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7902: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7903: }
7904: }
1.220 brouard 7905:
1.222 brouard 7906: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7907: 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);
7908: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7909: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7910: }
7911: /* } /\* end i1 *\/ */
1.337 brouard 7912: }/* End k1=nres */
1.222 brouard 7913: fprintf(fichtm,"</ul>");
1.126 brouard 7914:
1.222 brouard 7915: fprintf(fichtm,"\
1.126 brouard 7916: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7917: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7918: - 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 7919: But because parameters are usually highly correlated (a higher incidence of disability \
7920: and a higher incidence of recovery can give very close observed transition) it might \
7921: be very useful to look not only at linear confidence intervals estimated from the \
7922: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7923: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7924: covariance matrix of the one-step probabilities. \
7925: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7926:
1.222 brouard 7927: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7928: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7929: fprintf(fichtm,"\
1.126 brouard 7930: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7931: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7932:
1.222 brouard 7933: fprintf(fichtm,"\
1.126 brouard 7934: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7935: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7936: fprintf(fichtm,"\
1.126 brouard 7937: - 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): \
7938: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7939: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7940: fprintf(fichtm,"\
1.126 brouard 7941: - (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): \
7942: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7943: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7944: fprintf(fichtm,"\
1.288 brouard 7945: - 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 7946: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7947: fprintf(fichtm,"\
1.128 brouard 7948: - 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 7949: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7950: fprintf(fichtm,"\
1.288 brouard 7951: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7952: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7953:
7954: /* if(popforecast==1) fprintf(fichtm,"\n */
7955: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7956: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7957: /* <br>",fileres,fileres,fileres,fileres); */
7958: /* else */
1.338 brouard 7959: /* 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 7960: fflush(fichtm);
1.126 brouard 7961:
1.225 brouard 7962: m=pow(2,cptcoveff);
1.222 brouard 7963: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7964:
1.317 brouard 7965: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
7966:
7967: jj1=0;
7968:
7969: fprintf(fichtm," \n<ul>");
1.337 brouard 7970: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7971: /* k1=nres; */
1.338 brouard 7972: k1=TKresult[nres];
1.337 brouard 7973: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
7974: /* if(m != 1 && TKresult[nres]!= k1) */
7975: /* continue; */
1.317 brouard 7976: jj1++;
7977: if (cptcovn > 0) {
7978: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
1.337 brouard 7979: for (cpt=1; cpt<=cptcovs;cpt++){
7980: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 7981: }
7982: fprintf(fichtm,"\">");
7983:
7984: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7985: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 7986: for (cpt=1; cpt<=cptcovs;cpt++){
7987: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 7988: }
7989: if(invalidvarcomb[k1]){
7990: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7991: continue;
7992: }
7993: fprintf(fichtm,"</a></li>");
7994: } /* cptcovn >0 */
1.337 brouard 7995: } /* End nres */
1.317 brouard 7996: fprintf(fichtm," \n</ul>");
7997:
1.222 brouard 7998: jj1=0;
1.237 brouard 7999:
1.241 brouard 8000: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8001: /* k1=nres; */
1.338 brouard 8002: k1=TKresult[nres];
8003: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8004: /* for(k1=1; k1<=m;k1++){ */
8005: /* if(m != 1 && TKresult[nres]!= k1) */
8006: /* continue; */
1.222 brouard 8007: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
8008: jj1++;
1.126 brouard 8009: if (cptcovn > 0) {
1.317 brouard 8010: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
1.337 brouard 8011: for (cpt=1; cpt<=cptcovs;cpt++){
8012: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8013: }
8014: fprintf(fichtm,"\"</a>");
8015:
1.126 brouard 8016: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 8017: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcoveff number of variables */
8018: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
8019: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 8020: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 8021: }
1.237 brouard 8022:
1.338 brouard 8023: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 8024:
1.222 brouard 8025: if(invalidvarcomb[k1]){
8026: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
8027: continue;
8028: }
1.337 brouard 8029: } /* If cptcovn >0 */
1.126 brouard 8030: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 8031: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 8032: 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);
8033: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
8034: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 8035: }
8036: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 8037: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 8038: true period expectancies (those weighted with period prevalences are also\
8039: drawn in addition to the population based expectancies computed using\
1.314 brouard 8040: 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);
8041: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
8042: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 8043: /* } /\* end i1 *\/ */
1.241 brouard 8044: }/* End nres */
1.222 brouard 8045: fprintf(fichtm,"</ul>");
8046: fflush(fichtm);
1.126 brouard 8047: }
8048:
8049: /******************* Gnuplot file **************/
1.296 brouard 8050: 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 8051:
8052: char dirfileres[132],optfileres[132];
1.264 brouard 8053: char gplotcondition[132], gplotlabel[132];
1.237 brouard 8054: int cpt=0,k1=0,i=0,k=0,j=0,jk=0,k2=0,k3=0,k4=0,ij=0, ijp=0, l=0;
1.211 brouard 8055: int lv=0, vlv=0, kl=0;
1.130 brouard 8056: int ng=0;
1.201 brouard 8057: int vpopbased;
1.223 brouard 8058: int ioffset; /* variable offset for columns */
1.270 brouard 8059: int iyearc=1; /* variable column for year of projection */
8060: int iagec=1; /* variable column for age of projection */
1.235 brouard 8061: int nres=0; /* Index of resultline */
1.266 brouard 8062: int istart=1; /* For starting graphs in projections */
1.219 brouard 8063:
1.126 brouard 8064: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
8065: /* printf("Problem with file %s",optionfilegnuplot); */
8066: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
8067: /* } */
8068:
8069: /*#ifdef windows */
8070: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 8071: /*#endif */
1.225 brouard 8072: m=pow(2,cptcoveff);
1.126 brouard 8073:
1.274 brouard 8074: /* diagram of the model */
8075: fprintf(ficgp,"\n#Diagram of the model \n");
8076: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
8077: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
8078: 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);
8079:
8080: fprintf(ficgp,"\n#Centripete arrows (turning in other direction (1-i) instead of (i-1)) \nset for [i=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);
8081: fprintf(ficgp,"\n#show arrow\nunset label\n");
8082: 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);
8083: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
8084: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
8085: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
8086: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
8087:
1.202 brouard 8088: /* Contribution to likelihood */
8089: /* Plot the probability implied in the likelihood */
1.223 brouard 8090: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
8091: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
8092: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
8093: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 8094: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 8095: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
8096: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 8097: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
8098: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
8099: 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));
8100: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
8101: 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));
8102: for (i=1; i<= nlstate ; i ++) {
8103: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
8104: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
8105: 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);
8106: for (j=2; j<= nlstate+ndeath ; j ++) {
8107: 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);
8108: }
8109: fprintf(ficgp,";\nset out; unset ylabel;\n");
8110: }
8111: /* 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 */
8112: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
8113: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
8114: fprintf(ficgp,"\nset out;unset log\n");
8115: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 8116:
1.126 brouard 8117: strcpy(dirfileres,optionfilefiname);
8118: strcpy(optfileres,"vpl");
1.223 brouard 8119: /* 1eme*/
1.238 brouard 8120: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
1.337 brouard 8121: /* for (k1=1; k1<= m ; k1 ++){ /\* For each valid combination of covariate *\/ */
1.236 brouard 8122: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8123: k1=TKresult[nres];
1.338 brouard 8124: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.238 brouard 8125: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.337 brouard 8126: /* if(m != 1 && TKresult[nres]!= k1) */
8127: /* continue; */
1.238 brouard 8128: /* We are interested in selected combination by the resultline */
1.246 brouard 8129: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 8130: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 8131: strcpy(gplotlabel,"(");
1.337 brouard 8132: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8133: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8134: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8135:
8136: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate k get corresponding value lv for combination k1 *\/ */
8137: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the value of the covariate corresponding to k1 combination *\\/ *\/ */
8138: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8139: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8140: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8141: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8142: /* vlv= nbcode[Tvaraff[k]][lv]; /\* vlv is the value of the covariate lv, 0 or 1 *\/ */
8143: /* /\* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv *\/ */
8144: /* /\* printf(" V%d=%d ",Tvaraff[k],vlv); *\/ */
8145: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8146: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8147: /* } */
8148: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8149: /* /\* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); *\/ */
8150: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8151: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.264 brouard 8152: }
8153: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 8154: /* printf("\n#\n"); */
1.238 brouard 8155: fprintf(ficgp,"\n#\n");
8156: if(invalidvarcomb[k1]){
1.260 brouard 8157: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 8158: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8159: continue;
8160: }
1.235 brouard 8161:
1.241 brouard 8162: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
8163: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 8164: /* 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 8165: fprintf(ficgp,"set title \"Alive state %d %s model=1+age+%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 8166: 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);
8167: /* 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); */
8168: /* k1-1 error should be nres-1*/
1.238 brouard 8169: for (i=1; i<= nlstate ; i ++) {
8170: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8171: else fprintf(ficgp," %%*lf (%%*lf)");
8172: }
1.288 brouard 8173: 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 8174: for (i=1; i<= nlstate ; i ++) {
8175: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8176: else fprintf(ficgp," %%*lf (%%*lf)");
8177: }
1.260 brouard 8178: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 8179: for (i=1; i<= nlstate ; i ++) {
8180: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8181: else fprintf(ficgp," %%*lf (%%*lf)");
8182: }
1.265 brouard 8183: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
8184:
8185: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
8186: if(cptcoveff ==0){
1.271 brouard 8187: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 8188: }else{
8189: kl=0;
8190: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8191: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8192: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.265 brouard 8193: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8194: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8195: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8196: vlv= nbcode[Tvaraff[k]][lv];
8197: kl++;
8198: /* 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 *\/ */
8199: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8200: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8201: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8202: if(k==cptcoveff){
8203: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
8204: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
8205: }else{
8206: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
8207: kl++;
8208: }
8209: } /* end covariate */
8210: } /* end if no covariate */
8211:
1.296 brouard 8212: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 8213: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 8214: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 8215: if(cptcoveff ==0){
1.245 brouard 8216: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 8217: }else{
8218: kl=0;
8219: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8220: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8221: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 8222: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8223: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8224: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8225: /* vlv= nbcode[Tvaraff[k]][lv]; */
8226: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 8227: kl++;
1.238 brouard 8228: /* 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 *\/ */
8229: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8230: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8231: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8232: if(k==cptcoveff){
1.245 brouard 8233: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 8234: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 8235: }else{
1.332 brouard 8236: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]);
1.238 brouard 8237: kl++;
8238: }
8239: } /* end covariate */
8240: } /* end if no covariate */
1.296 brouard 8241: if(prevbcast == 1){
1.268 brouard 8242: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
8243: /* k1-1 error should be nres-1*/
8244: for (i=1; i<= nlstate ; i ++) {
8245: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8246: else fprintf(ficgp," %%*lf (%%*lf)");
8247: }
1.271 brouard 8248: 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 8249: for (i=1; i<= nlstate ; i ++) {
8250: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8251: else fprintf(ficgp," %%*lf (%%*lf)");
8252: }
1.276 brouard 8253: 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 8254: for (i=1; i<= nlstate ; i ++) {
8255: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8256: else fprintf(ficgp," %%*lf (%%*lf)");
8257: }
1.274 brouard 8258: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 8259: } /* end if backprojcast */
1.296 brouard 8260: } /* end if prevbcast */
1.276 brouard 8261: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
8262: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 8263: } /* nres */
1.337 brouard 8264: /* } /\* k1 *\/ */
1.201 brouard 8265: } /* cpt */
1.235 brouard 8266:
8267:
1.126 brouard 8268: /*2 eme*/
1.337 brouard 8269: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8270: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8271: k1=TKresult[nres];
1.338 brouard 8272: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8273: /* if(m != 1 && TKresult[nres]!= k1) */
8274: /* continue; */
1.238 brouard 8275: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 8276: strcpy(gplotlabel,"(");
1.337 brouard 8277: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8278: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8279: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8280: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8281: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8282: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8283: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8284: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8285: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8286: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8287: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8288: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8289: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8290: /* } */
8291: /* /\* for(k=1; k <= ncovds; k++){ *\/ */
8292: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8293: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8294: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8295: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 8296: }
1.264 brouard 8297: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8298: fprintf(ficgp,"\n#\n");
1.223 brouard 8299: if(invalidvarcomb[k1]){
8300: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8301: continue;
8302: }
1.219 brouard 8303:
1.241 brouard 8304: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 8305: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 8306: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
8307: if(vpopbased==0){
1.238 brouard 8308: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 8309: }else
1.238 brouard 8310: fprintf(ficgp,"\nreplot ");
8311: for (i=1; i<= nlstate+1 ; i ++) {
8312: k=2*i;
1.261 brouard 8313: 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 8314: for (j=1; j<= nlstate+1 ; j ++) {
8315: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8316: else fprintf(ficgp," %%*lf (%%*lf)");
8317: }
8318: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
8319: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 8320: 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 8321: for (j=1; j<= nlstate+1 ; j ++) {
8322: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8323: else fprintf(ficgp," %%*lf (%%*lf)");
8324: }
8325: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 8326: 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 8327: for (j=1; j<= nlstate+1 ; j ++) {
8328: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8329: else fprintf(ficgp," %%*lf (%%*lf)");
8330: }
8331: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
8332: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
8333: } /* state */
8334: } /* vpopbased */
1.264 brouard 8335: 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 8336: } /* end nres */
1.337 brouard 8337: /* } /\* k1 end 2 eme*\/ */
1.238 brouard 8338:
8339:
8340: /*3eme*/
1.337 brouard 8341: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8342: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8343: k1=TKresult[nres];
1.338 brouard 8344: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8345: /* if(m != 1 && TKresult[nres]!= k1) */
8346: /* continue; */
1.238 brouard 8347:
1.332 brouard 8348: for (cpt=1; cpt<= nlstate ; cpt ++) { /* Fragile no verification of covariate values */
1.261 brouard 8349: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 8350: strcpy(gplotlabel,"(");
1.337 brouard 8351: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8352: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8353: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8354: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8355: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8356: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8357: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8358: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8359: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8360: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8361: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8362: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8363: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8364: /* } */
8365: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8366: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
8367: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
8368: }
1.264 brouard 8369: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8370: fprintf(ficgp,"\n#\n");
8371: if(invalidvarcomb[k1]){
8372: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8373: continue;
8374: }
8375:
8376: /* k=2+nlstate*(2*cpt-2); */
8377: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 8378: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 8379: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 8380: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 8381: 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 8382: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8383: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8384: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
8385: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8386: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8387: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 8388:
1.238 brouard 8389: */
8390: for (i=1; i< nlstate ; i ++) {
1.261 brouard 8391: 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 8392: /* 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 8393:
1.238 brouard 8394: }
1.261 brouard 8395: 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 8396: }
1.264 brouard 8397: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 8398: } /* end nres */
1.337 brouard 8399: /* } /\* end kl 3eme *\/ */
1.126 brouard 8400:
1.223 brouard 8401: /* 4eme */
1.201 brouard 8402: /* Survival functions (period) from state i in state j by initial state i */
1.337 brouard 8403: /* for (k1=1; k1<=m; k1++){ /\* For each covariate and each value *\/ */
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: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 8410: strcpy(gplotlabel,"(");
1.337 brouard 8411: fprintf(ficgp,"\n#\n#\n# Survival functions in state %d : 'LIJ_' files, cov=%d state=%d", cpt, k1, cpt);
8412: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8413: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8414: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8415: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8416: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8417: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8418: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8419: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8420: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8421: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8422: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8423: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8424: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8425: /* } */
8426: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8427: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8428: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8429: }
1.264 brouard 8430: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8431: fprintf(ficgp,"\n#\n");
8432: if(invalidvarcomb[k1]){
8433: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8434: continue;
1.223 brouard 8435: }
1.238 brouard 8436:
1.241 brouard 8437: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 8438: 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 8439: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8440: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8441: k=3;
8442: for (i=1; i<= nlstate ; i ++){
8443: if(i==1){
8444: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8445: }else{
8446: fprintf(ficgp,", '' ");
8447: }
8448: l=(nlstate+ndeath)*(i-1)+1;
8449: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8450: for (j=2; j<= nlstate+ndeath ; j ++)
8451: fprintf(ficgp,"+$%d",k+l+j-1);
8452: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
8453: } /* nlstate */
1.264 brouard 8454: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8455: } /* end cpt state*/
8456: } /* end nres */
1.337 brouard 8457: /* } /\* end covariate k1 *\/ */
1.238 brouard 8458:
1.220 brouard 8459: /* 5eme */
1.201 brouard 8460: /* Survival functions (period) from state i in state j by final state j */
1.337 brouard 8461: /* for (k1=1; k1<= m ; k1++){ /\* For each covariate combination if any *\/ */
1.238 brouard 8462: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8463: k1=TKresult[nres];
1.338 brouard 8464: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8465: /* if(m != 1 && TKresult[nres]!= k1) */
8466: /* continue; */
1.238 brouard 8467: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 8468: strcpy(gplotlabel,"(");
1.238 brouard 8469: 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 8470: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8471: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8472: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8473: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8474: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8475: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8476: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8477: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8478: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8479: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8480: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8481: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8482: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8483: /* } */
8484: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8485: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8486: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8487: }
1.264 brouard 8488: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8489: fprintf(ficgp,"\n#\n");
8490: if(invalidvarcomb[k1]){
8491: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8492: continue;
8493: }
1.227 brouard 8494:
1.241 brouard 8495: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 8496: 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 8497: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8498: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8499: k=3;
8500: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8501: if(j==1)
8502: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8503: else
8504: fprintf(ficgp,", '' ");
8505: l=(nlstate+ndeath)*(cpt-1) +j;
8506: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
8507: /* for (i=2; i<= nlstate+ndeath ; i ++) */
8508: /* fprintf(ficgp,"+$%d",k+l+i-1); */
8509: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
8510: } /* nlstate */
8511: fprintf(ficgp,", '' ");
8512: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
8513: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8514: l=(nlstate+ndeath)*(cpt-1) +j;
8515: if(j < nlstate)
8516: fprintf(ficgp,"$%d +",k+l);
8517: else
8518: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
8519: }
1.264 brouard 8520: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8521: } /* end cpt state*/
1.337 brouard 8522: /* } /\* end covariate *\/ */
1.238 brouard 8523: } /* end nres */
1.227 brouard 8524:
1.220 brouard 8525: /* 6eme */
1.202 brouard 8526: /* CV preval stable (period) for each covariate */
1.337 brouard 8527: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8528: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8529: k1=TKresult[nres];
1.338 brouard 8530: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8531: /* if(m != 1 && TKresult[nres]!= k1) */
8532: /* continue; */
1.255 brouard 8533: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 8534: strcpy(gplotlabel,"(");
1.288 brouard 8535: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8536: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8537: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8538: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8539: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8540: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8541: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8542: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8543: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8544: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8545: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8546: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8547: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8548: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8549: /* } */
8550: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8551: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8552: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8553: }
1.264 brouard 8554: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8555: fprintf(ficgp,"\n#\n");
1.223 brouard 8556: if(invalidvarcomb[k1]){
1.227 brouard 8557: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8558: continue;
1.223 brouard 8559: }
1.227 brouard 8560:
1.241 brouard 8561: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 8562: 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 8563: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8564: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 8565: k=3; /* Offset */
1.255 brouard 8566: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8567: if(i==1)
8568: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8569: else
8570: fprintf(ficgp,", '' ");
1.255 brouard 8571: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8572: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8573: for (j=2; j<= nlstate ; j ++)
8574: fprintf(ficgp,"+$%d",k+l+j-1);
8575: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8576: } /* nlstate */
1.264 brouard 8577: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8578: } /* end cpt state*/
8579: } /* end covariate */
1.227 brouard 8580:
8581:
1.220 brouard 8582: /* 7eme */
1.296 brouard 8583: if(prevbcast == 1){
1.288 brouard 8584: /* CV backward prevalence for each covariate */
1.337 brouard 8585: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8586: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8587: k1=TKresult[nres];
1.338 brouard 8588: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8589: /* if(m != 1 && TKresult[nres]!= k1) */
8590: /* continue; */
1.268 brouard 8591: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8592: strcpy(gplotlabel,"(");
1.288 brouard 8593: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8594: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8595: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8596: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8597: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8598: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8599: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8600: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8601: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8602: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8603: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8604: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8605: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8606: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8607: /* } */
8608: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8609: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8610: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8611: }
1.264 brouard 8612: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8613: fprintf(ficgp,"\n#\n");
8614: if(invalidvarcomb[k1]){
8615: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8616: continue;
8617: }
8618:
1.241 brouard 8619: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8620: 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 8621: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8622: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8623: k=3; /* Offset */
1.268 brouard 8624: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8625: if(i==1)
8626: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8627: else
8628: fprintf(ficgp,", '' ");
8629: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8630: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8631: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8632: /* 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 8633: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8634: /* for (j=2; j<= nlstate ; j ++) */
8635: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8636: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8637: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8638: } /* nlstate */
1.264 brouard 8639: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8640: } /* end cpt state*/
8641: } /* end covariate */
1.296 brouard 8642: } /* End if prevbcast */
1.218 brouard 8643:
1.223 brouard 8644: /* 8eme */
1.218 brouard 8645: if(prevfcast==1){
1.288 brouard 8646: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8647:
1.337 brouard 8648: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8649: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8650: k1=TKresult[nres];
1.338 brouard 8651: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8652: /* if(m != 1 && TKresult[nres]!= k1) */
8653: /* continue; */
1.211 brouard 8654: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8655: strcpy(gplotlabel,"(");
1.288 brouard 8656: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8657: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8658: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8659: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8660: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
8661: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
8662: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8663: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8664: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8665: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8666: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8667: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8668: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8669: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8670: /* } */
8671: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8672: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8673: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8674: }
1.264 brouard 8675: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8676: fprintf(ficgp,"\n#\n");
8677: if(invalidvarcomb[k1]){
8678: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8679: continue;
8680: }
8681:
8682: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8683: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8684: 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 8685: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8686: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8687:
8688: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8689: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8690: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8691: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8692: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8693: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8694: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8695: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8696: if(i==istart){
1.227 brouard 8697: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8698: }else{
8699: fprintf(ficgp,",\\\n '' ");
8700: }
8701: if(cptcoveff ==0){ /* No covariate */
8702: ioffset=2; /* Age is in 2 */
8703: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8704: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8705: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8706: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8707: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8708: if(i==nlstate+1){
1.270 brouard 8709: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8710: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8711: fprintf(ficgp,",\\\n '' ");
8712: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8713: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8714: offyear, \
1.268 brouard 8715: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8716: }else
1.227 brouard 8717: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8718: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8719: }else{ /* more than 2 covariates */
1.270 brouard 8720: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8721: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8722: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8723: iyearc=ioffset-1;
8724: iagec=ioffset;
1.227 brouard 8725: fprintf(ficgp," u %d:(",ioffset);
8726: kl=0;
8727: strcpy(gplotcondition,"(");
8728: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8729: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8730: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8731: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8732: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8733: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8734: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8735: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8736: kl++;
8737: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8738: kl++;
8739: if(k <cptcoveff && cptcoveff>1)
8740: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8741: }
8742: strcpy(gplotcondition+strlen(gplotcondition),")");
8743: /* 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 *\/ */
8744: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8745: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8746: /* '' 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*/
8747: if(i==nlstate+1){
1.270 brouard 8748: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8749: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8750: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8751: fprintf(ficgp," u %d:(",iagec);
8752: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8753: iyearc, iagec, offyear, \
8754: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8755: /* '' 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 8756: }else{
8757: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8758: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8759: }
8760: } /* end if covariate */
8761: } /* nlstate */
1.264 brouard 8762: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8763: } /* end cpt state*/
8764: } /* end covariate */
8765: } /* End if prevfcast */
1.227 brouard 8766:
1.296 brouard 8767: if(prevbcast==1){
1.268 brouard 8768: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8769:
1.337 brouard 8770: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.268 brouard 8771: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8772: k1=TKresult[nres];
1.338 brouard 8773: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8774: /* if(m != 1 && TKresult[nres]!= k1) */
8775: /* continue; */
1.268 brouard 8776: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8777: strcpy(gplotlabel,"(");
8778: 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 8779: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8780: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8781: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8782: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
8783: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
8784: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8785: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8786: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8787: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8788: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8789: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8790: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8791: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8792: /* } */
8793: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8794: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8795: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.268 brouard 8796: }
8797: strcpy(gplotlabel+strlen(gplotlabel),")");
8798: fprintf(ficgp,"\n#\n");
8799: if(invalidvarcomb[k1]){
8800: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8801: continue;
8802: }
8803:
8804: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8805: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8806: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8807: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8808: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8809:
8810: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8811: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8812: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8813: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8814: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8815: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8816: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8817: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8818: if(i==istart){
8819: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8820: }else{
8821: fprintf(ficgp,",\\\n '' ");
8822: }
8823: if(cptcoveff ==0){ /* No covariate */
8824: ioffset=2; /* Age is in 2 */
8825: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8826: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8827: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8828: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8829: fprintf(ficgp," u %d:(", ioffset);
8830: if(i==nlstate+1){
1.270 brouard 8831: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8832: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8833: fprintf(ficgp,",\\\n '' ");
8834: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8835: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8836: offbyear, \
8837: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8838: }else
8839: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8840: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8841: }else{ /* more than 2 covariates */
1.270 brouard 8842: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8843: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8844: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8845: iyearc=ioffset-1;
8846: iagec=ioffset;
1.268 brouard 8847: fprintf(ficgp," u %d:(",ioffset);
8848: kl=0;
8849: strcpy(gplotcondition,"(");
1.337 brouard 8850: for (k=1; k<=cptcovs; k++){ /* For each covariate k of the resultline, get corresponding value lv for combination k1 */
1.338 brouard 8851: if(Dummy[modelresult[nres][k]]==0){ /* To be verified */
1.337 brouard 8852: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate writing the chain of conditions *\/ */
8853: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8854: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8855: lv=Tvresult[nres][k];
8856: vlv=TinvDoQresult[nres][Tvresult[nres][k]];
8857: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8858: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8859: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8860: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8861: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8862: kl++;
8863: /* sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]); */
8864: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%lg " ,kl,Tvresult[nres][k], kl+1,TinvDoQresult[nres][Tvresult[nres][k]]);
8865: kl++;
1.338 brouard 8866: if(k <cptcovs && cptcovs>1)
1.337 brouard 8867: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8868: }
1.268 brouard 8869: }
8870: strcpy(gplotcondition+strlen(gplotcondition),")");
8871: /* 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 *\/ */
8872: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8873: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8874: /* '' 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*/
8875: if(i==nlstate+1){
1.270 brouard 8876: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
8877: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 8878: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8879: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 8880: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 8881: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
8882: iyearc,iagec,offbyear, \
8883: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 8884: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
8885: }else{
8886: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
8887: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
8888: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
8889: }
8890: } /* end if covariate */
8891: } /* nlstate */
8892: fprintf(ficgp,"\nset out; unset label;\n");
8893: } /* end cpt state*/
8894: } /* end covariate */
1.296 brouard 8895: } /* End if prevbcast */
1.268 brouard 8896:
1.227 brouard 8897:
1.238 brouard 8898: /* 9eme writing MLE parameters */
8899: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 8900: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 8901: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 8902: for(k=1; k <=(nlstate+ndeath); k++){
8903: if (k != i) {
1.227 brouard 8904: fprintf(ficgp,"# current state %d\n",k);
8905: for(j=1; j <=ncovmodel; j++){
8906: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
8907: jk++;
8908: }
8909: fprintf(ficgp,"\n");
1.126 brouard 8910: }
8911: }
1.223 brouard 8912: }
1.187 brouard 8913: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8914:
1.145 brouard 8915: /*goto avoid;*/
1.238 brouard 8916: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8917: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8918: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8919: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8920: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8921: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8922: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8923: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8924: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8925: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8926: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8927: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8928: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8929: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8930: fprintf(ficgp,"#\n");
1.223 brouard 8931: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8932: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.338 brouard 8933: fprintf(ficgp,"#model=1+age+%s \n",model);
1.238 brouard 8934: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8935: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
1.337 brouard 8936: /* for(k1=1; k1 <=m; k1++) /\* For each combination of covariate *\/ */
1.237 brouard 8937: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8938: /* k1=nres; */
1.338 brouard 8939: k1=TKresult[nres];
8940: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8941: fprintf(ficgp,"\n\n# Resultline k1=%d ",k1);
1.264 brouard 8942: strcpy(gplotlabel,"(");
1.276 brouard 8943: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.337 brouard 8944: for (k=1; k<=cptcovs; k++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
8945: /* for each resultline nres, and position k, Tvresult[nres][k] gives the name of the variable and
8946: TinvDoQresult[nres][Tvresult[nres][k]] gives its value double or integer) */
8947: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8948: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8949: }
8950: /* if(m != 1 && TKresult[nres]!= k1) */
8951: /* continue; */
8952: /* fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1); */
8953: /* strcpy(gplotlabel,"("); */
8954: /* /\*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*\/ */
8955: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
8956: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
8957: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8958: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8959: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8960: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8961: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8962: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8963: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8964: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8965: /* } */
8966: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8967: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8968: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8969: /* } */
1.264 brouard 8970: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8971: fprintf(ficgp,"\n#\n");
1.264 brouard 8972: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8973: fprintf(ficgp,"\nset key outside ");
8974: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8975: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8976: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8977: if (ng==1){
8978: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8979: fprintf(ficgp,"\nunset log y");
8980: }else if (ng==2){
8981: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8982: fprintf(ficgp,"\nset log y");
8983: }else if (ng==3){
8984: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8985: fprintf(ficgp,"\nset log y");
8986: }else
8987: fprintf(ficgp,"\nunset title ");
8988: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8989: i=1;
8990: for(k2=1; k2<=nlstate; k2++) {
8991: k3=i;
8992: for(k=1; k<=(nlstate+ndeath); k++) {
8993: if (k != k2){
8994: switch( ng) {
8995: case 1:
8996: if(nagesqr==0)
8997: fprintf(ficgp," p%d+p%d*x",i,i+1);
8998: else /* nagesqr =1 */
8999: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
9000: break;
9001: case 2: /* ng=2 */
9002: if(nagesqr==0)
9003: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
9004: else /* nagesqr =1 */
9005: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
9006: break;
9007: case 3:
9008: if(nagesqr==0)
9009: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
9010: else /* nagesqr =1 */
9011: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
9012: break;
9013: }
9014: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 9015: ijp=1; /* product no age */
9016: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
9017: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 9018: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 9019: switch(Typevar[j]){
9020: case 1:
9021: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
9022: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
9023: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
9024: if(DummyV[j]==0){/* Bug valgrind */
9025: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
9026: }else{ /* quantitative */
9027: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
9028: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9029: }
9030: ij++;
1.268 brouard 9031: }
1.237 brouard 9032: }
1.329 brouard 9033: }
9034: break;
9035: case 2:
9036: if(cptcovprod >0){
9037: if(j==Tprod[ijp]) { /* */
9038: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
9039: if(ijp <=cptcovprod) { /* Product */
9040: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
9041: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
9042: /* 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)]); */
9043: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
9044: }else{ /* Vn is dummy and Vm is quanti */
9045: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
9046: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9047: }
9048: }else{ /* Vn*Vm Vn is quanti */
9049: if(DummyV[Tvard[ijp][2]]==0){
9050: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
9051: }else{ /* Both quanti */
9052: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9053: }
1.268 brouard 9054: }
1.329 brouard 9055: ijp++;
1.237 brouard 9056: }
1.329 brouard 9057: } /* end Tprod */
9058: }
9059: break;
9060: case 0:
9061: /* simple covariate */
1.264 brouard 9062: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 9063: if(Dummy[j]==0){
9064: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
9065: }else{ /* quantitative */
9066: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 9067: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 9068: }
1.329 brouard 9069: /* end simple */
9070: break;
9071: default:
9072: break;
9073: } /* end switch */
1.237 brouard 9074: } /* end j */
1.329 brouard 9075: }else{ /* k=k2 */
9076: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
9077: fprintf(ficgp," (1.");i=i-ncovmodel;
9078: }else
9079: i=i-ncovmodel;
1.223 brouard 9080: }
1.227 brouard 9081:
1.223 brouard 9082: if(ng != 1){
9083: fprintf(ficgp,")/(1");
1.227 brouard 9084:
1.264 brouard 9085: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 9086: if(nagesqr==0)
1.264 brouard 9087: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 9088: else /* nagesqr =1 */
1.264 brouard 9089: 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 9090:
1.223 brouard 9091: ij=1;
1.329 brouard 9092: ijp=1;
9093: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
9094: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
9095: switch(Typevar[j]){
9096: case 1:
9097: if(cptcovage >0){
9098: if(j==Tage[ij]) { /* Bug valgrind */
9099: if(ij <=cptcovage) { /* Bug valgrind */
9100: if(DummyV[j]==0){/* Bug valgrind */
9101: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
9102: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
9103: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
9104: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
9105: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9106: }else{ /* quantitative */
9107: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
9108: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
9109: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
9110: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9111: }
9112: ij++;
9113: }
9114: }
9115: }
9116: break;
9117: case 2:
9118: if(cptcovprod >0){
9119: if(j==Tprod[ijp]) { /* */
9120: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
9121: if(ijp <=cptcovprod) { /* Product */
9122: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
9123: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
9124: /* 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)]); */
9125: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
9126: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
9127: }else{ /* Vn is dummy and Vm is quanti */
9128: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
9129: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9130: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
9131: }
9132: }else{ /* Vn*Vm Vn is quanti */
9133: if(DummyV[Tvard[ijp][2]]==0){
9134: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
9135: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
9136: }else{ /* Both quanti */
9137: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9138: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
9139: }
9140: }
9141: ijp++;
9142: }
9143: } /* end Tprod */
9144: } /* end if */
9145: break;
9146: case 0:
9147: /* simple covariate */
9148: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
9149: if(Dummy[j]==0){
9150: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
9151: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
9152: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
9153: }else{ /* quantitative */
9154: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
9155: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
9156: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9157: }
9158: /* end simple */
9159: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
9160: break;
9161: default:
9162: break;
9163: } /* end switch */
1.223 brouard 9164: }
9165: fprintf(ficgp,")");
9166: }
9167: fprintf(ficgp,")");
9168: if(ng ==2)
1.276 brouard 9169: 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 9170: else /* ng= 3 */
1.276 brouard 9171: 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 9172: }else{ /* end ng <> 1 */
1.223 brouard 9173: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 9174: 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 9175: }
9176: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
9177: fprintf(ficgp,",");
9178: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
9179: fprintf(ficgp,",");
9180: i=i+ncovmodel;
9181: } /* end k */
9182: } /* end k2 */
1.276 brouard 9183: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
9184: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.337 brouard 9185: } /* end resultline */
1.223 brouard 9186: } /* end ng */
9187: /* avoid: */
9188: fflush(ficgp);
1.126 brouard 9189: } /* end gnuplot */
9190:
9191:
9192: /*************** Moving average **************/
1.219 brouard 9193: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 9194: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 9195:
1.222 brouard 9196: int i, cpt, cptcod;
9197: int modcovmax =1;
9198: int mobilavrange, mob;
9199: int iage=0;
1.288 brouard 9200: int firstA1=0, firstA2=0;
1.222 brouard 9201:
1.266 brouard 9202: double sum=0., sumr=0.;
1.222 brouard 9203: double age;
1.266 brouard 9204: double *sumnewp, *sumnewm, *sumnewmr;
9205: double *agemingood, *agemaxgood;
9206: double *agemingoodr, *agemaxgoodr;
1.222 brouard 9207:
9208:
1.278 brouard 9209: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
9210: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 9211:
9212: sumnewp = vector(1,ncovcombmax);
9213: sumnewm = vector(1,ncovcombmax);
1.266 brouard 9214: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 9215: agemingood = vector(1,ncovcombmax);
1.266 brouard 9216: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 9217: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 9218: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 9219:
9220: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 9221: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 9222: sumnewp[cptcod]=0.;
1.266 brouard 9223: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
9224: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 9225: }
9226: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
9227:
1.266 brouard 9228: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
9229: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 9230: else mobilavrange=mobilav;
9231: for (age=bage; age<=fage; age++)
9232: for (i=1; i<=nlstate;i++)
9233: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
9234: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9235: /* We keep the original values on the extreme ages bage, fage and for
9236: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
9237: we use a 5 terms etc. until the borders are no more concerned.
9238: */
9239: for (mob=3;mob <=mobilavrange;mob=mob+2){
9240: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 9241: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
9242: sumnewm[cptcod]=0.;
9243: for (i=1; i<=nlstate;i++){
1.222 brouard 9244: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
9245: for (cpt=1;cpt<=(mob-1)/2;cpt++){
9246: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
9247: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
9248: }
9249: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 9250: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9251: } /* end i */
9252: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
9253: } /* end cptcod */
1.222 brouard 9254: }/* end age */
9255: }/* end mob */
1.266 brouard 9256: }else{
9257: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 9258: return -1;
1.266 brouard 9259: }
9260:
9261: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 9262: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
9263: if(invalidvarcomb[cptcod]){
9264: printf("\nCombination (%d) ignored because no cases \n",cptcod);
9265: continue;
9266: }
1.219 brouard 9267:
1.266 brouard 9268: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
9269: sumnewm[cptcod]=0.;
9270: sumnewmr[cptcod]=0.;
9271: for (i=1; i<=nlstate;i++){
9272: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9273: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9274: }
9275: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9276: agemingoodr[cptcod]=age;
9277: }
9278: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9279: agemingood[cptcod]=age;
9280: }
9281: } /* age */
9282: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 9283: sumnewm[cptcod]=0.;
1.266 brouard 9284: sumnewmr[cptcod]=0.;
1.222 brouard 9285: for (i=1; i<=nlstate;i++){
9286: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9287: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9288: }
9289: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9290: agemaxgoodr[cptcod]=age;
1.222 brouard 9291: }
9292: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 9293: agemaxgood[cptcod]=age;
9294: }
9295: } /* age */
9296: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
9297: /* but they will change */
1.288 brouard 9298: firstA1=0;firstA2=0;
1.266 brouard 9299: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
9300: sumnewm[cptcod]=0.;
9301: sumnewmr[cptcod]=0.;
9302: for (i=1; i<=nlstate;i++){
9303: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9304: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9305: }
9306: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9307: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9308: agemaxgoodr[cptcod]=age; /* age min */
9309: for (i=1; i<=nlstate;i++)
9310: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9311: }else{ /* bad we change the value with the values of good ages */
9312: for (i=1; i<=nlstate;i++){
9313: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
9314: } /* i */
9315: } /* end bad */
9316: }else{
9317: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9318: agemaxgood[cptcod]=age;
9319: }else{ /* bad we change the value with the values of good ages */
9320: for (i=1; i<=nlstate;i++){
9321: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
9322: } /* i */
9323: } /* end bad */
9324: }/* end else */
9325: sum=0.;sumr=0.;
9326: for (i=1; i<=nlstate;i++){
9327: sum+=mobaverage[(int)age][i][cptcod];
9328: sumr+=probs[(int)age][i][cptcod];
9329: }
9330: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 9331: if(!firstA1){
9332: firstA1=1;
9333: 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);
9334: }
9335: 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 9336: } /* end bad */
9337: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9338: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 9339: if(!firstA2){
9340: firstA2=1;
9341: 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);
9342: }
9343: 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 9344: } /* end bad */
9345: }/* age */
1.266 brouard 9346:
9347: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 9348: sumnewm[cptcod]=0.;
1.266 brouard 9349: sumnewmr[cptcod]=0.;
1.222 brouard 9350: for (i=1; i<=nlstate;i++){
9351: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9352: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9353: }
9354: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9355: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
9356: agemingoodr[cptcod]=age;
9357: for (i=1; i<=nlstate;i++)
9358: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9359: }else{ /* bad we change the value with the values of good ages */
9360: for (i=1; i<=nlstate;i++){
9361: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
9362: } /* i */
9363: } /* end bad */
9364: }else{
9365: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9366: agemingood[cptcod]=age;
9367: }else{ /* bad */
9368: for (i=1; i<=nlstate;i++){
9369: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
9370: } /* i */
9371: } /* end bad */
9372: }/* end else */
9373: sum=0.;sumr=0.;
9374: for (i=1; i<=nlstate;i++){
9375: sum+=mobaverage[(int)age][i][cptcod];
9376: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 9377: }
1.266 brouard 9378: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 9379: 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 9380: } /* end bad */
9381: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9382: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 9383: 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 9384: } /* end bad */
9385: }/* age */
1.266 brouard 9386:
1.222 brouard 9387:
9388: for (age=bage; age<=fage; age++){
1.235 brouard 9389: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 9390: sumnewp[cptcod]=0.;
9391: sumnewm[cptcod]=0.;
9392: for (i=1; i<=nlstate;i++){
9393: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
9394: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9395: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
9396: }
9397: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
9398: }
9399: /* printf("\n"); */
9400: /* } */
1.266 brouard 9401:
1.222 brouard 9402: /* brutal averaging */
1.266 brouard 9403: /* for (i=1; i<=nlstate;i++){ */
9404: /* for (age=1; age<=bage; age++){ */
9405: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
9406: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9407: /* } */
9408: /* for (age=fage; age<=AGESUP; age++){ */
9409: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
9410: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9411: /* } */
9412: /* } /\* end i status *\/ */
9413: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
9414: /* for (age=1; age<=AGESUP; age++){ */
9415: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
9416: /* mobaverage[(int)age][i][cptcod]=0.; */
9417: /* } */
9418: /* } */
1.222 brouard 9419: }/* end cptcod */
1.266 brouard 9420: free_vector(agemaxgoodr,1, ncovcombmax);
9421: free_vector(agemaxgood,1, ncovcombmax);
9422: free_vector(agemingood,1, ncovcombmax);
9423: free_vector(agemingoodr,1, ncovcombmax);
9424: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 9425: free_vector(sumnewm,1, ncovcombmax);
9426: free_vector(sumnewp,1, ncovcombmax);
9427: return 0;
9428: }/* End movingaverage */
1.218 brouard 9429:
1.126 brouard 9430:
1.296 brouard 9431:
1.126 brouard 9432: /************** Forecasting ******************/
1.296 brouard 9433: /* 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)*/
9434: 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){
9435: /* dateintemean, mean date of interviews
9436: dateprojd, year, month, day of starting projection
9437: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 9438: agemin, agemax range of age
9439: dateprev1 dateprev2 range of dates during which prevalence is computed
9440: */
1.296 brouard 9441: /* double anprojd, mprojd, jprojd; */
9442: /* double anprojf, mprojf, jprojf; */
1.267 brouard 9443: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 9444: double agec; /* generic age */
1.296 brouard 9445: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 9446: double *popeffectif,*popcount;
9447: double ***p3mat;
1.218 brouard 9448: /* double ***mobaverage; */
1.126 brouard 9449: char fileresf[FILENAMELENGTH];
9450:
9451: agelim=AGESUP;
1.211 brouard 9452: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9453: in each health status at the date of interview (if between dateprev1 and dateprev2).
9454: We still use firstpass and lastpass as another selection.
9455: */
1.214 brouard 9456: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9457: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 9458:
1.201 brouard 9459: strcpy(fileresf,"F_");
9460: strcat(fileresf,fileresu);
1.126 brouard 9461: if((ficresf=fopen(fileresf,"w"))==NULL) {
9462: printf("Problem with forecast resultfile: %s\n", fileresf);
9463: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
9464: }
1.235 brouard 9465: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
9466: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 9467:
1.225 brouard 9468: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 9469:
9470:
9471: stepsize=(int) (stepm+YEARM-1)/YEARM;
9472: if (stepm<=12) stepsize=1;
9473: if(estepm < stepm){
9474: printf ("Problem %d lower than %d\n",estepm, stepm);
9475: }
1.270 brouard 9476: else{
9477: hstepm=estepm;
9478: }
9479: if(estepm > stepm){ /* Yes every two year */
9480: stepsize=2;
9481: }
1.296 brouard 9482: hstepm=hstepm/stepm;
1.126 brouard 9483:
1.296 brouard 9484:
9485: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9486: /* fractional in yp1 *\/ */
9487: /* aintmean=yp; */
9488: /* yp2=modf((yp1*12),&yp); */
9489: /* mintmean=yp; */
9490: /* yp1=modf((yp2*30.5),&yp); */
9491: /* jintmean=yp; */
9492: /* if(jintmean==0) jintmean=1; */
9493: /* if(mintmean==0) mintmean=1; */
1.126 brouard 9494:
1.296 brouard 9495:
9496: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
9497: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
9498: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 9499: i1=pow(2,cptcoveff);
1.126 brouard 9500: if (cptcovn < 1){i1=1;}
9501:
1.296 brouard 9502: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 9503:
9504: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 9505:
1.126 brouard 9506: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 9507: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9508: 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 9509: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9510: continue;
1.227 brouard 9511: if(invalidvarcomb[k]){
9512: printf("\nCombination (%d) projection ignored because no cases \n",k);
9513: continue;
9514: }
9515: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
9516: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9517: /* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); */
9518: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.227 brouard 9519: }
1.235 brouard 9520: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 9521: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 9522: }
1.227 brouard 9523: fprintf(ficresf," yearproj age");
9524: for(j=1; j<=nlstate+ndeath;j++){
9525: for(i=1; i<=nlstate;i++)
9526: fprintf(ficresf," p%d%d",i,j);
9527: fprintf(ficresf," wp.%d",j);
9528: }
1.296 brouard 9529: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 9530: fprintf(ficresf,"\n");
1.296 brouard 9531: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 9532: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
9533: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 9534: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
9535: nhstepm = nhstepm/hstepm;
9536: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9537: oldm=oldms;savm=savms;
1.268 brouard 9538: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 9539: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 9540: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 9541: for (h=0; h<=nhstepm; h++){
9542: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 9543: break;
9544: }
9545: }
9546: fprintf(ficresf,"\n");
9547: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9548: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Tvaraff not correct *\/ */
9549: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* TnsdVar[Tvaraff] correct */
1.296 brouard 9550: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 9551:
9552: for(j=1; j<=nlstate+ndeath;j++) {
9553: ppij=0.;
9554: for(i=1; i<=nlstate;i++) {
1.278 brouard 9555: if (mobilav>=1)
9556: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
9557: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
9558: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
9559: }
1.268 brouard 9560: fprintf(ficresf," %.3f", p3mat[i][j][h]);
9561: } /* end i */
9562: fprintf(ficresf," %.3f", ppij);
9563: }/* end j */
1.227 brouard 9564: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9565: } /* end agec */
1.266 brouard 9566: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
9567: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 9568: } /* end yearp */
9569: } /* end k */
1.219 brouard 9570:
1.126 brouard 9571: fclose(ficresf);
1.215 brouard 9572: printf("End of Computing forecasting \n");
9573: fprintf(ficlog,"End of Computing forecasting\n");
9574:
1.126 brouard 9575: }
9576:
1.269 brouard 9577: /************** Back Forecasting ******************/
1.296 brouard 9578: /* 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){ */
9579: 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){
9580: /* back1, year, month, day of starting backprojection
1.267 brouard 9581: agemin, agemax range of age
9582: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 9583: anback2 year of end of backprojection (same day and month as back1).
9584: prevacurrent and prev are prevalences.
1.267 brouard 9585: */
9586: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
9587: double agec; /* generic age */
1.302 brouard 9588: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 9589: double *popeffectif,*popcount;
9590: double ***p3mat;
9591: /* double ***mobaverage; */
9592: char fileresfb[FILENAMELENGTH];
9593:
1.268 brouard 9594: agelim=AGEINF;
1.267 brouard 9595: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9596: in each health status at the date of interview (if between dateprev1 and dateprev2).
9597: We still use firstpass and lastpass as another selection.
9598: */
9599: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9600: /* firstpass, lastpass, stepm, weightopt, model); */
9601:
9602: /*Do we need to compute prevalence again?*/
9603:
9604: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
9605:
9606: strcpy(fileresfb,"FB_");
9607: strcat(fileresfb,fileresu);
9608: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
9609: printf("Problem with back forecast resultfile: %s\n", fileresfb);
9610: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
9611: }
9612: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9613: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9614:
9615: if (cptcoveff==0) ncodemax[cptcoveff]=1;
9616:
9617:
9618: stepsize=(int) (stepm+YEARM-1)/YEARM;
9619: if (stepm<=12) stepsize=1;
9620: if(estepm < stepm){
9621: printf ("Problem %d lower than %d\n",estepm, stepm);
9622: }
1.270 brouard 9623: else{
9624: hstepm=estepm;
9625: }
9626: if(estepm >= stepm){ /* Yes every two year */
9627: stepsize=2;
9628: }
1.267 brouard 9629:
9630: hstepm=hstepm/stepm;
1.296 brouard 9631: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9632: /* fractional in yp1 *\/ */
9633: /* aintmean=yp; */
9634: /* yp2=modf((yp1*12),&yp); */
9635: /* mintmean=yp; */
9636: /* yp1=modf((yp2*30.5),&yp); */
9637: /* jintmean=yp; */
9638: /* if(jintmean==0) jintmean=1; */
9639: /* if(mintmean==0) jintmean=1; */
1.267 brouard 9640:
9641: i1=pow(2,cptcoveff);
9642: if (cptcovn < 1){i1=1;}
9643:
1.296 brouard 9644: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
9645: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 9646:
9647: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
9648:
9649: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9650: for(k=1; k<=i1;k++){
9651: if(i1 != 1 && TKresult[nres]!= k)
9652: continue;
9653: if(invalidvarcomb[k]){
9654: printf("\nCombination (%d) projection ignored because no cases \n",k);
9655: continue;
9656: }
1.268 brouard 9657: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 9658: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9659: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.267 brouard 9660: }
9661: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9662: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9663: }
9664: fprintf(ficresfb," yearbproj age");
9665: for(j=1; j<=nlstate+ndeath;j++){
9666: for(i=1; i<=nlstate;i++)
1.268 brouard 9667: fprintf(ficresfb," b%d%d",i,j);
9668: fprintf(ficresfb," b.%d",j);
1.267 brouard 9669: }
1.296 brouard 9670: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 9671: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
9672: fprintf(ficresfb,"\n");
1.296 brouard 9673: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 9674: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 9675: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
9676: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 9677: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 9678: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 9679: nhstepm = nhstepm/hstepm;
9680: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9681: oldm=oldms;savm=savms;
1.268 brouard 9682: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 9683: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 9684: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 9685: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
9686: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
9687: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9688: for (h=0; h<=nhstepm; h++){
1.268 brouard 9689: if (h*hstepm/YEARM*stepm ==-yearp) {
9690: break;
9691: }
9692: }
9693: fprintf(ficresfb,"\n");
9694: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9695: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.296 brouard 9696: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9697: for(i=1; i<=nlstate+ndeath;i++) {
9698: ppij=0.;ppi=0.;
9699: for(j=1; j<=nlstate;j++) {
9700: /* if (mobilav==1) */
1.269 brouard 9701: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9702: ppi=ppi+prevacurrent[(int)agec][j][k];
9703: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9704: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9705: /* else { */
9706: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9707: /* } */
1.268 brouard 9708: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9709: } /* end j */
9710: if(ppi <0.99){
9711: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9712: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9713: }
9714: fprintf(ficresfb," %.3f", ppij);
9715: }/* end j */
1.267 brouard 9716: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9717: } /* end agec */
9718: } /* end yearp */
9719: } /* end k */
1.217 brouard 9720:
1.267 brouard 9721: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9722:
1.267 brouard 9723: fclose(ficresfb);
9724: printf("End of Computing Back forecasting \n");
9725: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9726:
1.267 brouard 9727: }
1.217 brouard 9728:
1.269 brouard 9729: /* Variance of prevalence limit: varprlim */
9730: 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 9731: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9732:
9733: char fileresvpl[FILENAMELENGTH];
9734: FILE *ficresvpl;
9735: double **oldm, **savm;
9736: double **varpl; /* Variances of prevalence limits by age */
9737: int i1, k, nres, j ;
9738:
9739: strcpy(fileresvpl,"VPL_");
9740: strcat(fileresvpl,fileresu);
9741: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9742: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9743: exit(0);
9744: }
1.288 brouard 9745: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9746: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9747:
9748: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9749: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9750:
9751: i1=pow(2,cptcoveff);
9752: if (cptcovn < 1){i1=1;}
9753:
1.337 brouard 9754: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9755: k=TKresult[nres];
1.338 brouard 9756: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 9757: /* for(k=1; k<=i1;k++){ /\* We find the combination equivalent to result line values of dummies *\/ */
1.269 brouard 9758: if(i1 != 1 && TKresult[nres]!= k)
9759: continue;
9760: fprintf(ficresvpl,"\n#****** ");
9761: printf("\n#****** ");
9762: fprintf(ficlog,"\n#****** ");
1.337 brouard 9763: for(j=1;j<=cptcovs;j++) {
9764: fprintf(ficresvpl,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9765: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9766: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9767: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9768: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.269 brouard 9769: }
1.337 brouard 9770: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
9771: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9772: /* fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9773: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9774: /* } */
1.269 brouard 9775: fprintf(ficresvpl,"******\n");
9776: printf("******\n");
9777: fprintf(ficlog,"******\n");
9778:
9779: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9780: oldm=oldms;savm=savms;
9781: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9782: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9783: /*}*/
9784: }
9785:
9786: fclose(ficresvpl);
1.288 brouard 9787: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9788: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9789:
9790: }
9791: /* Variance of back prevalence: varbprlim */
9792: 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){
9793: /*------- Variance of back (stable) prevalence------*/
9794:
9795: char fileresvbl[FILENAMELENGTH];
9796: FILE *ficresvbl;
9797:
9798: double **oldm, **savm;
9799: double **varbpl; /* Variances of back prevalence limits by age */
9800: int i1, k, nres, j ;
9801:
9802: strcpy(fileresvbl,"VBL_");
9803: strcat(fileresvbl,fileresu);
9804: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9805: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9806: exit(0);
9807: }
9808: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9809: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9810:
9811:
9812: i1=pow(2,cptcoveff);
9813: if (cptcovn < 1){i1=1;}
9814:
1.337 brouard 9815: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9816: k=TKresult[nres];
1.338 brouard 9817: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 9818: /* for(k=1; k<=i1;k++){ */
9819: /* if(i1 != 1 && TKresult[nres]!= k) */
9820: /* continue; */
1.269 brouard 9821: fprintf(ficresvbl,"\n#****** ");
9822: printf("\n#****** ");
9823: fprintf(ficlog,"\n#****** ");
1.337 brouard 9824: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */
1.338 brouard 9825: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
9826: fprintf(ficresvbl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
9827: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
1.337 brouard 9828: /* for(j=1;j<=cptcoveff;j++) { */
9829: /* fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9830: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9831: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9832: /* } */
9833: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
9834: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9835: /* fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9836: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
1.269 brouard 9837: }
9838: fprintf(ficresvbl,"******\n");
9839: printf("******\n");
9840: fprintf(ficlog,"******\n");
9841:
9842: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9843: oldm=oldms;savm=savms;
9844:
9845: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9846: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9847: /*}*/
9848: }
9849:
9850: fclose(ficresvbl);
9851: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9852: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9853:
9854: } /* End of varbprlim */
9855:
1.126 brouard 9856: /************** Forecasting *****not tested NB*************/
1.227 brouard 9857: /* 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 9858:
1.227 brouard 9859: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
9860: /* int *popage; */
9861: /* double calagedatem, agelim, kk1, kk2; */
9862: /* double *popeffectif,*popcount; */
9863: /* double ***p3mat,***tabpop,***tabpopprev; */
9864: /* /\* double ***mobaverage; *\/ */
9865: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 9866:
1.227 brouard 9867: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9868: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9869: /* agelim=AGESUP; */
9870: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 9871:
1.227 brouard 9872: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 9873:
9874:
1.227 brouard 9875: /* strcpy(filerespop,"POP_"); */
9876: /* strcat(filerespop,fileresu); */
9877: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
9878: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
9879: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
9880: /* } */
9881: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
9882: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 9883:
1.227 brouard 9884: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 9885:
1.227 brouard 9886: /* /\* if (mobilav!=0) { *\/ */
9887: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
9888: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
9889: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9890: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9891: /* /\* } *\/ */
9892: /* /\* } *\/ */
1.126 brouard 9893:
1.227 brouard 9894: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
9895: /* if (stepm<=12) stepsize=1; */
1.126 brouard 9896:
1.227 brouard 9897: /* agelim=AGESUP; */
1.126 brouard 9898:
1.227 brouard 9899: /* hstepm=1; */
9900: /* hstepm=hstepm/stepm; */
1.218 brouard 9901:
1.227 brouard 9902: /* if (popforecast==1) { */
9903: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
9904: /* printf("Problem with population file : %s\n",popfile);exit(0); */
9905: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
9906: /* } */
9907: /* popage=ivector(0,AGESUP); */
9908: /* popeffectif=vector(0,AGESUP); */
9909: /* popcount=vector(0,AGESUP); */
1.126 brouard 9910:
1.227 brouard 9911: /* i=1; */
9912: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 9913:
1.227 brouard 9914: /* imx=i; */
9915: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
9916: /* } */
1.218 brouard 9917:
1.227 brouard 9918: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
9919: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
9920: /* k=k+1; */
9921: /* fprintf(ficrespop,"\n#******"); */
9922: /* for(j=1;j<=cptcoveff;j++) { */
9923: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
9924: /* } */
9925: /* fprintf(ficrespop,"******\n"); */
9926: /* fprintf(ficrespop,"# Age"); */
9927: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
9928: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 9929:
1.227 brouard 9930: /* for (cpt=0; cpt<=0;cpt++) { */
9931: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 9932:
1.227 brouard 9933: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9934: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9935: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9936:
1.227 brouard 9937: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9938: /* oldm=oldms;savm=savms; */
9939: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 9940:
1.227 brouard 9941: /* for (h=0; h<=nhstepm; h++){ */
9942: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9943: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9944: /* } */
9945: /* for(j=1; j<=nlstate+ndeath;j++) { */
9946: /* kk1=0.;kk2=0; */
9947: /* for(i=1; i<=nlstate;i++) { */
9948: /* if (mobilav==1) */
9949: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
9950: /* else { */
9951: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
9952: /* } */
9953: /* } */
9954: /* if (h==(int)(calagedatem+12*cpt)){ */
9955: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
9956: /* /\*fprintf(ficrespop," %.3f", kk1); */
9957: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
9958: /* } */
9959: /* } */
9960: /* for(i=1; i<=nlstate;i++){ */
9961: /* kk1=0.; */
9962: /* for(j=1; j<=nlstate;j++){ */
9963: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
9964: /* } */
9965: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
9966: /* } */
1.218 brouard 9967:
1.227 brouard 9968: /* if (h==(int)(calagedatem+12*cpt)) */
9969: /* for(j=1; j<=nlstate;j++) */
9970: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
9971: /* } */
9972: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9973: /* } */
9974: /* } */
1.218 brouard 9975:
1.227 brouard 9976: /* /\******\/ */
1.218 brouard 9977:
1.227 brouard 9978: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
9979: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
9980: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9981: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9982: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9983:
1.227 brouard 9984: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9985: /* oldm=oldms;savm=savms; */
9986: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9987: /* for (h=0; h<=nhstepm; h++){ */
9988: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9989: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9990: /* } */
9991: /* for(j=1; j<=nlstate+ndeath;j++) { */
9992: /* kk1=0.;kk2=0; */
9993: /* for(i=1; i<=nlstate;i++) { */
9994: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
9995: /* } */
9996: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
9997: /* } */
9998: /* } */
9999: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10000: /* } */
10001: /* } */
10002: /* } */
10003: /* } */
1.218 brouard 10004:
1.227 brouard 10005: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 10006:
1.227 brouard 10007: /* if (popforecast==1) { */
10008: /* free_ivector(popage,0,AGESUP); */
10009: /* free_vector(popeffectif,0,AGESUP); */
10010: /* free_vector(popcount,0,AGESUP); */
10011: /* } */
10012: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10013: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10014: /* fclose(ficrespop); */
10015: /* } /\* End of popforecast *\/ */
1.218 brouard 10016:
1.126 brouard 10017: int fileappend(FILE *fichier, char *optionfich)
10018: {
10019: if((fichier=fopen(optionfich,"a"))==NULL) {
10020: printf("Problem with file: %s\n", optionfich);
10021: fprintf(ficlog,"Problem with file: %s\n", optionfich);
10022: return (0);
10023: }
10024: fflush(fichier);
10025: return (1);
10026: }
10027:
10028:
10029: /**************** function prwizard **********************/
10030: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
10031: {
10032:
10033: /* Wizard to print covariance matrix template */
10034:
1.164 brouard 10035: char ca[32], cb[32];
10036: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 10037: int numlinepar;
10038:
10039: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10040: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10041: for(i=1; i <=nlstate; i++){
10042: jj=0;
10043: for(j=1; j <=nlstate+ndeath; j++){
10044: if(j==i) continue;
10045: jj++;
10046: /*ca[0]= k+'a'-1;ca[1]='\0';*/
10047: printf("%1d%1d",i,j);
10048: fprintf(ficparo,"%1d%1d",i,j);
10049: for(k=1; k<=ncovmodel;k++){
10050: /* printf(" %lf",param[i][j][k]); */
10051: /* fprintf(ficparo," %lf",param[i][j][k]); */
10052: printf(" 0.");
10053: fprintf(ficparo," 0.");
10054: }
10055: printf("\n");
10056: fprintf(ficparo,"\n");
10057: }
10058: }
10059: printf("# Scales (for hessian or gradient estimation)\n");
10060: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
10061: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
10062: for(i=1; i <=nlstate; i++){
10063: jj=0;
10064: for(j=1; j <=nlstate+ndeath; j++){
10065: if(j==i) continue;
10066: jj++;
10067: fprintf(ficparo,"%1d%1d",i,j);
10068: printf("%1d%1d",i,j);
10069: fflush(stdout);
10070: for(k=1; k<=ncovmodel;k++){
10071: /* printf(" %le",delti3[i][j][k]); */
10072: /* fprintf(ficparo," %le",delti3[i][j][k]); */
10073: printf(" 0.");
10074: fprintf(ficparo," 0.");
10075: }
10076: numlinepar++;
10077: printf("\n");
10078: fprintf(ficparo,"\n");
10079: }
10080: }
10081: printf("# Covariance matrix\n");
10082: /* # 121 Var(a12)\n\ */
10083: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10084: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10085: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10086: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10087: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10088: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10089: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10090: fflush(stdout);
10091: fprintf(ficparo,"# Covariance matrix\n");
10092: /* # 121 Var(a12)\n\ */
10093: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10094: /* # ...\n\ */
10095: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10096:
10097: for(itimes=1;itimes<=2;itimes++){
10098: jj=0;
10099: for(i=1; i <=nlstate; i++){
10100: for(j=1; j <=nlstate+ndeath; j++){
10101: if(j==i) continue;
10102: for(k=1; k<=ncovmodel;k++){
10103: jj++;
10104: ca[0]= k+'a'-1;ca[1]='\0';
10105: if(itimes==1){
10106: printf("#%1d%1d%d",i,j,k);
10107: fprintf(ficparo,"#%1d%1d%d",i,j,k);
10108: }else{
10109: printf("%1d%1d%d",i,j,k);
10110: fprintf(ficparo,"%1d%1d%d",i,j,k);
10111: /* printf(" %.5le",matcov[i][j]); */
10112: }
10113: ll=0;
10114: for(li=1;li <=nlstate; li++){
10115: for(lj=1;lj <=nlstate+ndeath; lj++){
10116: if(lj==li) continue;
10117: for(lk=1;lk<=ncovmodel;lk++){
10118: ll++;
10119: if(ll<=jj){
10120: cb[0]= lk +'a'-1;cb[1]='\0';
10121: if(ll<jj){
10122: if(itimes==1){
10123: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10124: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10125: }else{
10126: printf(" 0.");
10127: fprintf(ficparo," 0.");
10128: }
10129: }else{
10130: if(itimes==1){
10131: printf(" Var(%s%1d%1d)",ca,i,j);
10132: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
10133: }else{
10134: printf(" 0.");
10135: fprintf(ficparo," 0.");
10136: }
10137: }
10138: }
10139: } /* end lk */
10140: } /* end lj */
10141: } /* end li */
10142: printf("\n");
10143: fprintf(ficparo,"\n");
10144: numlinepar++;
10145: } /* end k*/
10146: } /*end j */
10147: } /* end i */
10148: } /* end itimes */
10149:
10150: } /* end of prwizard */
10151: /******************* Gompertz Likelihood ******************************/
10152: double gompertz(double x[])
10153: {
1.302 brouard 10154: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 10155: int i,n=0; /* n is the size of the sample */
10156:
1.220 brouard 10157: for (i=1;i<=imx ; i++) {
1.126 brouard 10158: sump=sump+weight[i];
10159: /* sump=sump+1;*/
10160: num=num+1;
10161: }
1.302 brouard 10162: L=0.0;
10163: /* agegomp=AGEGOMP; */
1.126 brouard 10164: /* for (i=0; i<=imx; i++)
10165: 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]);*/
10166:
1.302 brouard 10167: for (i=1;i<=imx ; i++) {
10168: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
10169: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
10170: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
10171: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
10172: * +
10173: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
10174: */
10175: if (wav[i] > 1 || agedc[i] < AGESUP) {
10176: if (cens[i] == 1){
10177: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
10178: } else if (cens[i] == 0){
1.126 brouard 10179: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 10180: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
10181: } else
10182: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 10183: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 10184: L=L+A*weight[i];
1.126 brouard 10185: /* 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 10186: }
10187: }
1.126 brouard 10188:
1.302 brouard 10189: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 10190:
10191: return -2*L*num/sump;
10192: }
10193:
1.136 brouard 10194: #ifdef GSL
10195: /******************* Gompertz_f Likelihood ******************************/
10196: double gompertz_f(const gsl_vector *v, void *params)
10197: {
1.302 brouard 10198: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 10199: double *x= (double *) v->data;
10200: int i,n=0; /* n is the size of the sample */
10201:
10202: for (i=0;i<=imx-1 ; i++) {
10203: sump=sump+weight[i];
10204: /* sump=sump+1;*/
10205: num=num+1;
10206: }
10207:
10208:
10209: /* for (i=0; i<=imx; i++)
10210: 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]);*/
10211: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
10212: for (i=1;i<=imx ; i++)
10213: {
10214: if (cens[i] == 1 && wav[i]>1)
10215: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
10216:
10217: if (cens[i] == 0 && wav[i]>1)
10218: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
10219: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
10220:
10221: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
10222: if (wav[i] > 1 ) { /* ??? */
10223: LL=LL+A*weight[i];
10224: /* 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]);*/
10225: }
10226: }
10227:
10228: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
10229: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
10230:
10231: return -2*LL*num/sump;
10232: }
10233: #endif
10234:
1.126 brouard 10235: /******************* Printing html file ***********/
1.201 brouard 10236: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 10237: int lastpass, int stepm, int weightopt, char model[],\
10238: int imx, double p[],double **matcov,double agemortsup){
10239: int i,k;
10240:
10241: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
10242: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
10243: for (i=1;i<=2;i++)
10244: 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 10245: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 10246: fprintf(fichtm,"</ul>");
10247:
10248: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
10249:
10250: 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>");
10251:
10252: for (k=agegomp;k<(agemortsup-2);k++)
10253: 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]);
10254:
10255:
10256: fflush(fichtm);
10257: }
10258:
10259: /******************* Gnuplot file **************/
1.201 brouard 10260: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 10261:
10262: char dirfileres[132],optfileres[132];
1.164 brouard 10263:
1.126 brouard 10264: int ng;
10265:
10266:
10267: /*#ifdef windows */
10268: fprintf(ficgp,"cd \"%s\" \n",pathc);
10269: /*#endif */
10270:
10271:
10272: strcpy(dirfileres,optionfilefiname);
10273: strcpy(optfileres,"vpl");
1.199 brouard 10274: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 10275: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 10276: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 10277: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 10278: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
10279:
10280: }
10281:
1.136 brouard 10282: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
10283: {
1.126 brouard 10284:
1.136 brouard 10285: /*-------- data file ----------*/
10286: FILE *fic;
10287: char dummy[]=" ";
1.240 brouard 10288: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 10289: int lstra;
1.136 brouard 10290: int linei, month, year,iout;
1.302 brouard 10291: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 10292: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 10293: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 10294: char *stratrunc;
1.223 brouard 10295:
1.240 brouard 10296: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
10297: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.328 brouard 10298: for(v=1;v<NCOVMAX;v++){
10299: DummyV[v]=0;
10300: FixedV[v]=0;
10301: }
1.126 brouard 10302:
1.240 brouard 10303: for(v=1; v <=ncovcol;v++){
10304: DummyV[v]=0;
10305: FixedV[v]=0;
10306: }
10307: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
10308: DummyV[v]=1;
10309: FixedV[v]=0;
10310: }
10311: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
10312: DummyV[v]=0;
10313: FixedV[v]=1;
10314: }
10315: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
10316: DummyV[v]=1;
10317: FixedV[v]=1;
10318: }
10319: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
10320: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
10321: 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]);
10322: }
1.339 brouard 10323:
10324: ncovcolt=ncovcol+nqv+ntv+nqtv; /* total of covariates in the data, not in the model equation */
10325:
1.136 brouard 10326: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 10327: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10328: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 10329: }
1.126 brouard 10330:
1.302 brouard 10331: /* Is it a BOM UTF-8 Windows file? */
10332: /* First data line */
10333: linei=0;
10334: while(fgets(line, MAXLINE, fic)) {
10335: noffset=0;
10336: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10337: {
10338: noffset=noffset+3;
10339: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
10340: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
10341: fflush(ficlog); return 1;
10342: }
10343: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
10344: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
10345: {
10346: noffset=noffset+2;
1.304 brouard 10347: 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);
10348: 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 10349: fflush(ficlog); return 1;
10350: }
10351: else if( line[0] == 0 && line[1] == 0)
10352: {
10353: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10354: noffset=noffset+4;
1.304 brouard 10355: 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);
10356: 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 10357: fflush(ficlog); return 1;
10358: }
10359: } else{
10360: ;/*printf(" Not a BOM file\n");*/
10361: }
10362: /* If line starts with a # it is a comment */
10363: if (line[noffset] == '#') {
10364: linei=linei+1;
10365: break;
10366: }else{
10367: break;
10368: }
10369: }
10370: fclose(fic);
10371: if((fic=fopen(datafile,"r"))==NULL) {
10372: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10373: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
10374: }
10375: /* Not a Bom file */
10376:
1.136 brouard 10377: i=1;
10378: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
10379: linei=linei+1;
10380: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
10381: if(line[j] == '\t')
10382: line[j] = ' ';
10383: }
10384: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
10385: ;
10386: };
10387: line[j+1]=0; /* Trims blanks at end of line */
10388: if(line[0]=='#'){
10389: fprintf(ficlog,"Comment line\n%s\n",line);
10390: printf("Comment line\n%s\n",line);
10391: continue;
10392: }
10393: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 10394: strcpy(line, linetmp);
1.223 brouard 10395:
10396: /* Loops on waves */
10397: for (j=maxwav;j>=1;j--){
10398: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 10399: cutv(stra, strb, line, ' ');
10400: if(strb[0]=='.') { /* Missing value */
10401: lval=-1;
10402: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
1.341 brouard 10403: cotvar[j][ncovcol+nqv+ntv+iv][i]=-1; /* For performance reasons */
1.238 brouard 10404: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
10405: 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);
10406: 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);
10407: return 1;
10408: }
10409: }else{
10410: errno=0;
10411: /* what_kind_of_number(strb); */
10412: dval=strtod(strb,&endptr);
10413: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
10414: /* if(strb != endptr && *endptr == '\0') */
10415: /* dval=dlval; */
10416: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10417: if( strb[0]=='\0' || (*endptr != '\0')){
10418: 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);
10419: 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);
10420: return 1;
10421: }
10422: cotqvar[j][iv][i]=dval;
1.341 brouard 10423: cotvar[j][ncovcol+nqv+ntv+iv][i]=dval; /* because cotvar starts now at first ntv */
1.238 brouard 10424: }
10425: strcpy(line,stra);
1.223 brouard 10426: }/* end loop ntqv */
1.225 brouard 10427:
1.223 brouard 10428: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 10429: cutv(stra, strb, line, ' ');
10430: if(strb[0]=='.') { /* Missing value */
10431: lval=-1;
10432: }else{
10433: errno=0;
10434: lval=strtol(strb,&endptr,10);
10435: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10436: if( strb[0]=='\0' || (*endptr != '\0')){
10437: 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);
10438: 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);
10439: return 1;
10440: }
10441: }
10442: if(lval <-1 || lval >1){
10443: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10444: 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 10445: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10446: For example, for multinomial values like 1, 2 and 3,\n \
10447: build V1=0 V2=0 for the reference value (1),\n \
10448: V1=1 V2=0 for (2) \n \
1.223 brouard 10449: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10450: output of IMaCh is often meaningless.\n \
1.319 brouard 10451: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 10452: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10453: 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 10454: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10455: For example, for multinomial values like 1, 2 and 3,\n \
10456: build V1=0 V2=0 for the reference value (1),\n \
10457: V1=1 V2=0 for (2) \n \
1.223 brouard 10458: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10459: output of IMaCh is often meaningless.\n \
1.319 brouard 10460: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 10461: return 1;
10462: }
1.341 brouard 10463: cotvar[j][ncovcol+nqv+iv][i]=(double)(lval);
1.238 brouard 10464: strcpy(line,stra);
1.223 brouard 10465: }/* end loop ntv */
1.225 brouard 10466:
1.223 brouard 10467: /* Statuses at wave */
1.137 brouard 10468: cutv(stra, strb, line, ' ');
1.223 brouard 10469: if(strb[0]=='.') { /* Missing value */
1.238 brouard 10470: lval=-1;
1.136 brouard 10471: }else{
1.238 brouard 10472: errno=0;
10473: lval=strtol(strb,&endptr,10);
10474: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10475: if( strb[0]=='\0' || (*endptr != '\0')){
10476: 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);
10477: 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);
10478: return 1;
10479: }
1.136 brouard 10480: }
1.225 brouard 10481:
1.136 brouard 10482: s[j][i]=lval;
1.225 brouard 10483:
1.223 brouard 10484: /* Date of Interview */
1.136 brouard 10485: strcpy(line,stra);
10486: cutv(stra, strb,line,' ');
1.169 brouard 10487: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10488: }
1.169 brouard 10489: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 10490: month=99;
10491: year=9999;
1.136 brouard 10492: }else{
1.225 brouard 10493: 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);
10494: 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);
10495: return 1;
1.136 brouard 10496: }
10497: anint[j][i]= (double) year;
1.302 brouard 10498: mint[j][i]= (double)month;
10499: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
10500: /* 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]); */
10501: /* 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]); */
10502: /* } */
1.136 brouard 10503: strcpy(line,stra);
1.223 brouard 10504: } /* End loop on waves */
1.225 brouard 10505:
1.223 brouard 10506: /* Date of death */
1.136 brouard 10507: cutv(stra, strb,line,' ');
1.169 brouard 10508: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10509: }
1.169 brouard 10510: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 10511: month=99;
10512: year=9999;
10513: }else{
1.141 brouard 10514: 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 10515: 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);
10516: return 1;
1.136 brouard 10517: }
10518: andc[i]=(double) year;
10519: moisdc[i]=(double) month;
10520: strcpy(line,stra);
10521:
1.223 brouard 10522: /* Date of birth */
1.136 brouard 10523: cutv(stra, strb,line,' ');
1.169 brouard 10524: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10525: }
1.169 brouard 10526: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 10527: month=99;
10528: year=9999;
10529: }else{
1.141 brouard 10530: 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);
10531: 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 10532: return 1;
1.136 brouard 10533: }
10534: if (year==9999) {
1.141 brouard 10535: 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);
10536: 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 10537: return 1;
10538:
1.136 brouard 10539: }
10540: annais[i]=(double)(year);
1.302 brouard 10541: moisnais[i]=(double)(month);
10542: for (j=1;j<=maxwav;j++){
10543: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
10544: 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]);
10545: 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]);
10546: }
10547: }
10548:
1.136 brouard 10549: strcpy(line,stra);
1.225 brouard 10550:
1.223 brouard 10551: /* Sample weight */
1.136 brouard 10552: cutv(stra, strb,line,' ');
10553: errno=0;
10554: dval=strtod(strb,&endptr);
10555: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 10556: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
10557: 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 10558: fflush(ficlog);
10559: return 1;
10560: }
10561: weight[i]=dval;
10562: strcpy(line,stra);
1.225 brouard 10563:
1.223 brouard 10564: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
10565: cutv(stra, strb, line, ' ');
10566: if(strb[0]=='.') { /* Missing value */
1.225 brouard 10567: lval=-1;
1.311 brouard 10568: coqvar[iv][i]=NAN;
10569: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 10570: }else{
1.225 brouard 10571: errno=0;
10572: /* what_kind_of_number(strb); */
10573: dval=strtod(strb,&endptr);
10574: /* if(strb != endptr && *endptr == '\0') */
10575: /* dval=dlval; */
10576: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10577: if( strb[0]=='\0' || (*endptr != '\0')){
10578: 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);
10579: 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);
10580: return 1;
10581: }
10582: coqvar[iv][i]=dval;
1.226 brouard 10583: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 10584: }
10585: strcpy(line,stra);
10586: }/* end loop nqv */
1.136 brouard 10587:
1.223 brouard 10588: /* Covariate values */
1.136 brouard 10589: for (j=ncovcol;j>=1;j--){
10590: cutv(stra, strb,line,' ');
1.223 brouard 10591: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 10592: lval=-1;
1.136 brouard 10593: }else{
1.225 brouard 10594: errno=0;
10595: lval=strtol(strb,&endptr,10);
10596: if( strb[0]=='\0' || (*endptr != '\0')){
10597: 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);
10598: 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);
10599: return 1;
10600: }
1.136 brouard 10601: }
10602: if(lval <-1 || lval >1){
1.225 brouard 10603: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10604: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10605: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10606: For example, for multinomial values like 1, 2 and 3,\n \
10607: build V1=0 V2=0 for the reference value (1),\n \
10608: V1=1 V2=0 for (2) \n \
1.136 brouard 10609: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10610: output of IMaCh is often meaningless.\n \
1.136 brouard 10611: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 10612: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10613: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10614: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10615: For example, for multinomial values like 1, 2 and 3,\n \
10616: build V1=0 V2=0 for the reference value (1),\n \
10617: V1=1 V2=0 for (2) \n \
1.136 brouard 10618: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10619: output of IMaCh is often meaningless.\n \
1.136 brouard 10620: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 10621: return 1;
1.136 brouard 10622: }
10623: covar[j][i]=(double)(lval);
10624: strcpy(line,stra);
10625: }
10626: lstra=strlen(stra);
1.225 brouard 10627:
1.136 brouard 10628: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
10629: stratrunc = &(stra[lstra-9]);
10630: num[i]=atol(stratrunc);
10631: }
10632: else
10633: num[i]=atol(stra);
10634: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
10635: 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;}*/
10636:
10637: i=i+1;
10638: } /* End loop reading data */
1.225 brouard 10639:
1.136 brouard 10640: *imax=i-1; /* Number of individuals */
10641: fclose(fic);
1.225 brouard 10642:
1.136 brouard 10643: return (0);
1.164 brouard 10644: /* endread: */
1.225 brouard 10645: printf("Exiting readdata: ");
10646: fclose(fic);
10647: return (1);
1.223 brouard 10648: }
1.126 brouard 10649:
1.234 brouard 10650: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 10651: char *p1 = *stri, *p2 = *stri;
1.235 brouard 10652: while (*p2 == ' ')
1.234 brouard 10653: p2++;
10654: /* while ((*p1++ = *p2++) !=0) */
10655: /* ; */
10656: /* do */
10657: /* while (*p2 == ' ') */
10658: /* p2++; */
10659: /* while (*p1++ == *p2++); */
10660: *stri=p2;
1.145 brouard 10661: }
10662:
1.330 brouard 10663: int decoderesult( char resultline[], int nres)
1.230 brouard 10664: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
10665: {
1.235 brouard 10666: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 10667: char resultsav[MAXLINE];
1.330 brouard 10668: /* int resultmodel[MAXLINE]; */
1.334 brouard 10669: /* int modelresult[MAXLINE]; */
1.230 brouard 10670: char stra[80], strb[80], strc[80], strd[80],stre[80];
10671:
1.234 brouard 10672: removefirstspace(&resultline);
1.332 brouard 10673: printf("decoderesult:%s\n",resultline);
1.230 brouard 10674:
1.332 brouard 10675: strcpy(resultsav,resultline);
1.342 ! brouard 10676: /* printf("Decoderesult resultsav=\"%s\" resultline=\"%s\"\n", resultsav, resultline); */
1.230 brouard 10677: if (strlen(resultsav) >1){
1.334 brouard 10678: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' in this resultline */
1.230 brouard 10679: }
1.253 brouard 10680: if(j == 0){ /* Resultline but no = */
10681: TKresult[nres]=0; /* Combination for the nresult and the model */
10682: return (0);
10683: }
1.234 brouard 10684: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.334 brouard 10685: 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);
10686: 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 10687: /* return 1;*/
1.234 brouard 10688: }
1.334 brouard 10689: for(k=1; k<=j;k++){ /* Loop on any covariate of the RESULT LINE */
1.234 brouard 10690: if(nbocc(resultsav,'=') >1){
1.318 brouard 10691: 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 10692: /* 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 10693: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.332 brouard 10694: /* If a blank, then strc="V4=" and strd='\0' */
10695: if(strc[0]=='\0'){
10696: printf("Error in resultline, probably a blank after the \"%s\", \"result:%s\", stra=\"%s\" resultsav=\"%s\"\n",strb,resultline, stra, resultsav);
10697: fprintf(ficlog,"Error in resultline, probably a blank after the \"V%s=\", resultline=%s\n",strb,resultline);
10698: return 1;
10699: }
1.234 brouard 10700: }else
10701: cutl(strc,strd,resultsav,'=');
1.318 brouard 10702: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 10703:
1.230 brouard 10704: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 10705: 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 10706: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
10707: /* cptcovsel++; */
10708: if (nbocc(stra,'=') >0)
10709: strcpy(resultsav,stra); /* and analyzes it */
10710: }
1.235 brouard 10711: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10712: /* 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 10713: 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 10714: if(Typevar[k1]==0){ /* Single covariate in model */
10715: /* 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 10716: match=0;
1.318 brouard 10717: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10718: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10719: modelresult[nres][k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10720: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10721: break;
10722: }
10723: }
10724: if(match == 0){
1.338 brouard 10725: 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]);
10726: 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 10727: return 1;
1.234 brouard 10728: }
1.332 brouard 10729: }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*/
10730: /* We feed resultmodel[k1]=k2; */
10731: match=0;
10732: 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 */
10733: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10734: 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 10735: resultmodel[nres][k1]=k2; /* Added here */
1.342 ! brouard 10736: /* printf("Decoderesult first modelresult[k2=%d]=%d (k1) V%d*AGE\n",k2,k1,Tvar[k1]); */
1.332 brouard 10737: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10738: break;
10739: }
10740: }
10741: if(match == 0){
1.338 brouard 10742: 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]);
10743: 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 10744: return 1;
10745: }
10746: }else if(Typevar[k1]==2){ /* Product No age We want to get the position in the resultline of the product in the model line*/
10747: /* resultmodel[nres][of such a Vn * Vm product k1] is not unique, so can't exist, we feed Tvard[k1][1] and [2] */
10748: match=0;
1.342 ! brouard 10749: /* 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 10750: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10751: if(Tvardk[k1][1]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10752: /* modelresult[k2]=k1; */
1.342 ! brouard 10753: /* printf("Decoderesult first Product modelresult[k2=%d]=%d (k1) V%d * \n",k2,k1,Tvarsel[k2]); */
1.332 brouard 10754: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10755: }
10756: }
10757: if(match == 0){
1.338 brouard 10758: 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);
10759: 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 10760: return 1;
10761: }
10762: match=0;
10763: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10764: if(Tvardk[k1][2]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10765: /* modelresult[k2]=k1;*/
1.342 ! brouard 10766: /* printf("Decoderesult second Product modelresult[k2=%d]=%d (k1) * V%d \n ",k2,k1,Tvarsel[k2]); */
1.332 brouard 10767: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10768: break;
10769: }
10770: }
10771: if(match == 0){
1.338 brouard 10772: 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);
10773: 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 10774: return 1;
10775: }
10776: }/* End of testing */
1.333 brouard 10777: }/* End loop cptcovt */
1.235 brouard 10778: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10779: /* Feeds resultmodel[nres][k1]=k2 for single covariate (k1) in the model equation */
1.334 brouard 10780: 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)
10781: * Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 10782: match=0;
1.318 brouard 10783: 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 10784: if(Typevar[k1]==0){ /* Single only */
1.237 brouard 10785: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.330 brouard 10786: 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 10787: 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 10788: ++match;
10789: }
10790: }
10791: }
10792: if(match == 0){
1.338 brouard 10793: printf("Error in result line: variable V%d is missing in model; result: %s, model=1+age+%s\n",Tvarsel[k2], resultline, model);
10794: 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 10795: return 1;
1.234 brouard 10796: }else if(match > 1){
1.338 brouard 10797: printf("Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
10798: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
1.310 brouard 10799: return 1;
1.234 brouard 10800: }
10801: }
1.334 brouard 10802: /* cptcovres=j /\* Number of variables in the resultline is equal to cptcovs and thus useless *\/ */
1.234 brouard 10803: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10804: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 brouard 10805: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
10806: /* 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*/
10807: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 10808: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
10809: /* 1 0 0 0 */
10810: /* 2 1 0 0 */
10811: /* 3 0 1 0 */
1.330 brouard 10812: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 10813: /* 5 0 0 1 */
1.330 brouard 10814: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 10815: /* 7 0 1 1 */
10816: /* 8 1 1 1 */
1.237 brouard 10817: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
10818: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
10819: /* V5*age V5 known which value for nres? */
10820: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.334 brouard 10821: 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.
10822: * loop on position k1 in the MODEL LINE */
1.331 brouard 10823: /* k counting number of combination of single dummies in the equation model */
10824: /* k4 counting single dummies in the equation model */
10825: /* k4q counting single quantitatives in the equation model */
1.334 brouard 10826: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Dummy and Single, k1 is sorting according to MODEL, but k3 to resultline */
10827: /* 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 10828: /* 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 10829: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
1.332 brouard 10830: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
10831: /* k3 is the position in the nres result line of the k1th variable of the model equation */
10832: /* Tvarsel[k3]: Name of the variable at the k3th position in the result line. */
10833: /* Tvalsel[k3]: Value of the variable at the k3th position in the result line. */
10834: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.334 brouard 10835: /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 10836: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1.330 brouard 10837: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.332 brouard 10838: k3= resultmodel[nres][k1]; /* From position k1 in model get position k3 in result line */
10839: /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
10840: 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 10841: 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 10842: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][Name]=Value; stores the value into the name of the variable. */
1.332 brouard 10843: /* Tinvresult[nres][4]=1 */
1.334 brouard 10844: /* Tresult[nres][k4+1]=Tvalsel[k3];/\* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) *\/ */
10845: Tresult[nres][k3]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
10846: /* Tvresult[nres][k4+1]=(int)Tvarsel[k3];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
10847: Tvresult[nres][k3]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
1.237 brouard 10848: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.334 brouard 10849: precov[nres][k1]=Tvalsel[k3]; /* Value from resultline of the variable at the k1 position in the model */
1.342 ! brouard 10850: /* 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 10851: k4++;;
1.331 brouard 10852: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Quantitative and single */
1.330 brouard 10853: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.332 brouard 10854: /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.330 brouard 10855: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
1.332 brouard 10856: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 5 =k3q */
10857: k2q=(int)Tvarsel[k3q]; /* Name of variable at k3q th position in the resultline */
10858: /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 10859: /* Tqresult[nres][k4q+1]=Tvalsel[k3q]; /\* Tqresult[nres][1]=25.1 *\/ */
10860: /* Tvresult[nres][k4q+1]=(int)Tvarsel[k3q];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
10861: /* Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /\* Tvqresult[nres][1]=5 *\/ */
10862: Tqresult[nres][k3q]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
10863: Tvresult[nres][k3q]=(int)Tvarsel[k3q];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
10864: Tvqresult[nres][k3q]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
1.237 brouard 10865: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 brouard 10866: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 10867: precov[nres][k1]=Tvalsel[k3q];
1.342 ! brouard 10868: /* 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 10869: k4q++;;
1.331 brouard 10870: }else if( Dummy[k1]==2 ){ /* For dummy with age product */
10871: /* Tvar[k1]; */ /* Age variable */
1.332 brouard 10872: /* Wrong we want the value of variable name Tvar[k1] */
10873:
10874: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
1.331 brouard 10875: 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 10876: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][4]=1 */
1.332 brouard 10877: precov[nres][k1]=Tvalsel[k3];
1.342 ! brouard 10878: /* 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 10879: }else if( Dummy[k1]==3 ){ /* For quant with age product */
1.332 brouard 10880: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
1.331 brouard 10881: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 10882: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* TinvDoQresult[nres][5]=25.1 */
1.332 brouard 10883: precov[nres][k1]=Tvalsel[k3q];
1.342 ! brouard 10884: /* 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 10885: }else if(Typevar[k1]==2 ){ /* For product quant or dummy (not with age) */
1.332 brouard 10886: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
1.342 ! brouard 10887: /* 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 10888: }else{
1.332 brouard 10889: printf("Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
10890: fprintf(ficlog,"Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 10891: }
10892: }
1.234 brouard 10893:
1.334 brouard 10894: TKresult[nres]=++k; /* Number of combinations of dummies for the nresult and the model =Tvalsel[k3]*pow(2,k4) + 1*/
1.230 brouard 10895: return (0);
10896: }
1.235 brouard 10897:
1.230 brouard 10898: int decodemodel( char model[], int lastobs)
10899: /**< This routine decodes the model and returns:
1.224 brouard 10900: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
10901: * - nagesqr = 1 if age*age in the model, otherwise 0.
10902: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
10903: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
10904: * - cptcovage number of covariates with age*products =2
10905: * - cptcovs number of simple covariates
1.339 brouard 10906: * ncovcolt=ncovcol+nqv+ntv+nqtv total of covariates in the data, not in the model equation
1.224 brouard 10907: * - 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 10908: * which is a new column after the 9 (ncovcol+nqv+ntv+nqtv) variables.
1.319 brouard 10909: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 10910: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
10911: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
10912: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
10913: */
1.319 brouard 10914: /* 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 10915: {
1.238 brouard 10916: int i, j, k, ks, v;
1.227 brouard 10917: int j1, k1, k2, k3, k4;
1.136 brouard 10918: char modelsav[80];
1.145 brouard 10919: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 10920: char *strpt;
1.136 brouard 10921:
1.145 brouard 10922: /*removespace(model);*/
1.136 brouard 10923: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 10924: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 10925: if (strstr(model,"AGE") !=0){
1.192 brouard 10926: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
10927: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 10928: return 1;
10929: }
1.141 brouard 10930: if (strstr(model,"v") !=0){
1.338 brouard 10931: printf("Error. 'v' must be in upper case 'V' model=1+age+%s ",model);
10932: fprintf(ficlog,"Error. 'v' must be in upper case model=1+age+%s ",model);fflush(ficlog);
1.141 brouard 10933: return 1;
10934: }
1.187 brouard 10935: strcpy(modelsav,model);
10936: if ((strpt=strstr(model,"age*age")) !=0){
1.338 brouard 10937: printf(" strpt=%s, model=1+age+%s\n",strpt, model);
1.187 brouard 10938: if(strpt != model){
1.338 brouard 10939: printf("Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10940: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10941: corresponding column of parameters.\n",model);
1.338 brouard 10942: fprintf(ficlog,"Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10943: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10944: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 10945: return 1;
1.225 brouard 10946: }
1.187 brouard 10947: nagesqr=1;
10948: if (strstr(model,"+age*age") !=0)
1.234 brouard 10949: substrchaine(modelsav, model, "+age*age");
1.187 brouard 10950: else if (strstr(model,"age*age+") !=0)
1.234 brouard 10951: substrchaine(modelsav, model, "age*age+");
1.187 brouard 10952: else
1.234 brouard 10953: substrchaine(modelsav, model, "age*age");
1.187 brouard 10954: }else
10955: nagesqr=0;
10956: if (strlen(modelsav) >1){
10957: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
10958: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 10959: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 10960: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 10961: * cst, age and age*age
10962: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
10963: /* including age products which are counted in cptcovage.
10964: * but the covariates which are products must be treated
10965: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 10966: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
10967: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 10968:
10969:
1.187 brouard 10970: /* Design
10971: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
10972: * < ncovcol=8 >
10973: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
10974: * k= 1 2 3 4 5 6 7 8
10975: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
10976: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 10977: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
10978: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 10979: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
10980: * Tage[++cptcovage]=k
10981: * if products, new covar are created after ncovcol with k1
10982: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
10983: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
10984: * 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
10985: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
10986: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
10987: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
10988: * < ncovcol=8 >
10989: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
10990: * k= 1 2 3 4 5 6 7 8 9 10 11 12
10991: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 10992: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 10993: * p Tprod[1]@2={ 6, 5}
10994: *p Tvard[1][1]@4= {7, 8, 5, 6}
10995: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
10996: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 10997: *How to reorganize? Tvars(orted)
1.187 brouard 10998: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
10999: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
11000: * {2, 1, 4, 8, 5, 6, 3, 7}
11001: * Struct []
11002: */
1.225 brouard 11003:
1.187 brouard 11004: /* This loop fills the array Tvar from the string 'model'.*/
11005: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
11006: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
11007: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
11008: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
11009: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
11010: /* k=1 Tvar[1]=2 (from V2) */
11011: /* k=5 Tvar[5] */
11012: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 11013: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 11014: /* } */
1.198 brouard 11015: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 11016: /*
11017: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 11018: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
11019: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
11020: }
1.187 brouard 11021: cptcovage=0;
1.319 brouard 11022: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
11023: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
11024: 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" */
11025: if (nbocc(modelsav,'+')==0)
11026: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 11027: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
11028: /*scanf("%d",i);*/
1.319 brouard 11029: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
11030: 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 11031: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
11032: /* covar is not filled and then is empty */
11033: cptcovprod--;
11034: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 11035: 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 11036: Typevar[k]=1; /* 1 for age product */
1.319 brouard 11037: cptcovage++; /* Counts the number of covariates which include age as a product */
11038: 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 11039: /*printf("stre=%s ", stre);*/
11040: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
11041: cptcovprod--;
11042: cutl(stre,strb,strc,'V');
11043: Tvar[k]=atoi(stre);
11044: Typevar[k]=1; /* 1 for age product */
11045: cptcovage++;
11046: Tage[cptcovage]=k;
11047: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
11048: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
11049: cptcovn++;
11050: cptcovprodnoage++;k1++;
11051: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
1.339 brouard 11052: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* ncovcolt+k1; For model-covariate k tells which data-covariate to use but
1.234 brouard 11053: because this model-covariate is a construction we invent a new column
11054: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.335 brouard 11055: If already ncovcol=4 and model= V2 + V1 + V1*V4 + age*V3 + V3*V2
1.319 brouard 11056: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
1.339 brouard 11057: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=3 etc */
1.335 brouard 11058: /* Please remark that the new variables are model dependent */
11059: /* If we have 4 variable but the model uses only 3, like in
11060: * model= V1 + age*V1 + V2 + V3 + age*V2 + age*V3 + V1*V2 + V1*V3
11061: * k= 1 2 3 4 5 6 7 8
11062: * Tvar[k]=1 1 2 3 2 3 (5 6) (and not 4 5 because of V4 missing)
11063: * Tage[kk] [1]= 2 [2]=5 [3]=6 kk=1 to cptcovage=3
11064: * Tvar[Tage[kk]][1]=2 [2]=2 [3]=3
11065: */
1.339 brouard 11066: Typevar[k]=2; /* 2 for product */
1.234 brouard 11067: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
11068: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 11069: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 11070: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
1.330 brouard 11071: Tvardk[k][1] =atoi(strc); /* m 1 for V1*/
1.234 brouard 11072: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
1.330 brouard 11073: Tvardk[k][2] =atoi(stre); /* n 4 for V4*/
1.234 brouard 11074: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
11075: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
11076: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 11077: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 11078: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
1.339 brouard 11079: 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 */
11080: for (i=1; i<=lastobs;i++){/* For fixed product */
1.234 brouard 11081: /* Computes the new covariate which is a product of
11082: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
1.339 brouard 11083: covar[ncovcolt+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
11084: }
11085: } /*End of FixedV */
1.234 brouard 11086: } /* End age is not in the model */
11087: } /* End if model includes a product */
1.319 brouard 11088: else { /* not a product */
1.234 brouard 11089: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
11090: /* scanf("%d",i);*/
11091: cutl(strd,strc,strb,'V');
11092: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
11093: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
11094: Tvar[k]=atoi(strd);
11095: Typevar[k]=0; /* 0 for simple covariates */
11096: }
11097: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 11098: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 11099: scanf("%d",i);*/
1.187 brouard 11100: } /* end of loop + on total covariates */
11101: } /* end if strlen(modelsave == 0) age*age might exist */
11102: } /* end if strlen(model == 0) */
1.136 brouard 11103:
11104: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
11105: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 11106:
1.136 brouard 11107: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 11108: printf("cptcovprod=%d ", cptcovprod);
11109: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
11110: scanf("%d ",i);*/
11111:
11112:
1.230 brouard 11113: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
11114: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 11115: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
11116: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
11117: k = 1 2 3 4 5 6 7 8 9
11118: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 11119: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 11120: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
11121: Dummy[k] 1 0 0 0 3 1 1 2 3
11122: Tmodelind[combination of covar]=k;
1.225 brouard 11123: */
11124: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 11125: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 11126: /* 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 11127: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 11128: printf("Model=1+age+%s\n\
1.227 brouard 11129: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
11130: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
11131: 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 11132: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 11133: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
11134: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
11135: 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 11136: for(k=-1;k<=NCOVMAX; k++){ Fixed[k]=0; Dummy[k]=0;}
! 11137: for(k=1;k<=NCOVMAX; k++){TvarFind[k]=0; TvarVind[k]=0;}
1.339 brouard 11138: 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 */
1.234 brouard 11139: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 11140: Fixed[k]= 0;
11141: Dummy[k]= 0;
1.225 brouard 11142: ncoveff++;
1.232 brouard 11143: ncovf++;
1.234 brouard 11144: nsd++;
11145: modell[k].maintype= FTYPE;
11146: TvarsD[nsd]=Tvar[k];
11147: TvarsDind[nsd]=k;
1.330 brouard 11148: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 11149: TvarF[ncovf]=Tvar[k];
11150: TvarFind[ncovf]=k;
11151: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
11152: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.339 brouard 11153: /* }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /\* Product of fixed dummy (<=ncovcol) covariates For a fixed product k is higher than ncovcol *\/ */
11154: }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 11155: Fixed[k]= 0;
11156: Dummy[k]= 0;
11157: ncoveff++;
11158: ncovf++;
11159: modell[k].maintype= FTYPE;
11160: TvarF[ncovf]=Tvar[k];
1.330 brouard 11161: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
1.234 brouard 11162: TvarFind[ncovf]=k;
1.230 brouard 11163: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 11164: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 11165: }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 11166: Fixed[k]= 0;
11167: Dummy[k]= 1;
1.230 brouard 11168: nqfveff++;
1.234 brouard 11169: modell[k].maintype= FTYPE;
11170: modell[k].subtype= FQ;
11171: nsq++;
1.334 brouard 11172: TvarsQ[nsq]=Tvar[k]; /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary see below */
11173: TvarsQind[nsq]=k; /* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.232 brouard 11174: ncovf++;
1.234 brouard 11175: TvarF[ncovf]=Tvar[k];
11176: TvarFind[ncovf]=k;
1.231 brouard 11177: 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 11178: 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 11179: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.339 brouard 11180: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11181: /* model V1+V3+age*V1+age*V3+V1*V3 */
11182: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11183: ncovvt++;
11184: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
11185: TvarVVind[ncovvt]=k; /* TvarVVind[1]=2 (second position in the model equation */
11186:
1.227 brouard 11187: Fixed[k]= 1;
11188: Dummy[k]= 0;
1.225 brouard 11189: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 11190: modell[k].maintype= VTYPE;
11191: modell[k].subtype= VD;
11192: nsd++;
11193: TvarsD[nsd]=Tvar[k];
11194: TvarsDind[nsd]=k;
1.330 brouard 11195: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 11196: ncovv++; /* Only simple time varying variables */
11197: TvarV[ncovv]=Tvar[k];
1.242 brouard 11198: 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 11199: 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 */
11200: 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 11201: 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);
11202: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 11203: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.339 brouard 11204: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11205: /* model V1+V3+age*V1+age*V3+V1*V3 */
11206: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11207: ncovvt++;
11208: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
11209: TvarVVind[ncovvt]=k; /* TvarVV[1]=V3 (first time varying in the model equation */
11210:
1.234 brouard 11211: Fixed[k]= 1;
11212: Dummy[k]= 1;
11213: nqtveff++;
11214: modell[k].maintype= VTYPE;
11215: modell[k].subtype= VQ;
11216: ncovv++; /* Only simple time varying variables */
11217: nsq++;
1.334 brouard 11218: 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) */
11219: 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 11220: TvarV[ncovv]=Tvar[k];
1.242 brouard 11221: 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 11222: 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 */
11223: 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 11224: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
11225: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
1.342 ! brouard 11226: /* 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); */
! 11227: /* printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv); */
1.227 brouard 11228: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 11229: ncova++;
11230: TvarA[ncova]=Tvar[k];
11231: TvarAind[ncova]=k;
1.231 brouard 11232: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 11233: Fixed[k]= 2;
11234: Dummy[k]= 2;
11235: modell[k].maintype= ATYPE;
11236: modell[k].subtype= APFD;
11237: /* ncoveff++; */
1.227 brouard 11238: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 11239: Fixed[k]= 2;
11240: Dummy[k]= 3;
11241: modell[k].maintype= ATYPE;
11242: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
11243: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 11244: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 11245: Fixed[k]= 3;
11246: Dummy[k]= 2;
11247: modell[k].maintype= ATYPE;
11248: modell[k].subtype= APVD; /* Product age * varying dummy */
11249: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 11250: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 11251: Fixed[k]= 3;
11252: Dummy[k]= 3;
11253: modell[k].maintype= ATYPE;
11254: modell[k].subtype= APVQ; /* Product age * varying quantitative */
11255: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 11256: }
1.339 brouard 11257: }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 */
11258: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11259: /* model V1+V3+age*V1+age*V3+V1*V3 */
11260: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11261: 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 */
11262: ncovvt++;
11263: TvarVV[ncovvt]=Tvard[k1][1]; /* TvarVV[2]=V1 (because TvarVV[1] was V3, first time varying covariates */
11264: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
11265: ncovvt++;
11266: TvarVV[ncovvt]=Tvard[k1][2]; /* TvarVV[3]=V3 */
11267: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
11268:
11269:
11270: if(Tvard[k1][1] <=ncovcol){ /* Vn is dummy fixed, (Tvard[1][1]=V1), (Tvard[1][1]=V3 time varying) */
11271: if(Tvard[k1][2] <=ncovcol){ /* Vm is dummy fixed */
1.240 brouard 11272: Fixed[k]= 1;
11273: Dummy[k]= 0;
11274: modell[k].maintype= FTYPE;
11275: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
11276: ncovf++; /* Fixed variables without age */
11277: TvarF[ncovf]=Tvar[k];
11278: TvarFind[ncovf]=k;
1.339 brouard 11279: }else if(Tvard[k1][2] <=ncovcol+nqv){ /* Vm is quanti fixed */
11280: Fixed[k]= 0; /* Fixed product */
1.240 brouard 11281: Dummy[k]= 1;
11282: modell[k].maintype= FTYPE;
11283: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
11284: ncovf++; /* Varying variables without age */
11285: TvarF[ncovf]=Tvar[k];
11286: TvarFind[ncovf]=k;
1.339 brouard 11287: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is a time varying dummy covariate */
1.240 brouard 11288: Fixed[k]= 1;
11289: Dummy[k]= 0;
11290: modell[k].maintype= VTYPE;
11291: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
11292: ncovv++; /* Varying variables without age */
1.339 brouard 11293: TvarV[ncovv]=Tvar[k]; /* TvarV[1]=Tvar[5]=5 because there is a V4 */
11294: TvarVind[ncovv]=k;/* TvarVind[1]=5 */
11295: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is a time varying quantitative covariate */
1.240 brouard 11296: Fixed[k]= 1;
11297: Dummy[k]= 1;
11298: modell[k].maintype= VTYPE;
11299: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
11300: ncovv++; /* Varying variables without age */
11301: TvarV[ncovv]=Tvar[k];
11302: TvarVind[ncovv]=k;
11303: }
1.339 brouard 11304: }else if(Tvard[k1][1] <=ncovcol+nqv){ /* Vn is fixed quanti */
11305: if(Tvard[k1][2] <=ncovcol){ /* Vm is fixed dummy */
11306: Fixed[k]= 0; /* Fixed product */
1.240 brouard 11307: Dummy[k]= 1;
11308: modell[k].maintype= FTYPE;
11309: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
11310: ncovf++; /* Fixed variables without age */
11311: TvarF[ncovf]=Tvar[k];
11312: TvarFind[ncovf]=k;
1.339 brouard 11313: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is time varying */
1.240 brouard 11314: Fixed[k]= 1;
11315: Dummy[k]= 1;
11316: modell[k].maintype= VTYPE;
11317: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
11318: ncovv++; /* Varying variables without age */
11319: TvarV[ncovv]=Tvar[k];
11320: TvarVind[ncovv]=k;
1.339 brouard 11321: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is time varying quanti */
1.240 brouard 11322: Fixed[k]= 1;
11323: Dummy[k]= 1;
11324: modell[k].maintype= VTYPE;
11325: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
11326: ncovv++; /* Varying variables without age */
11327: TvarV[ncovv]=Tvar[k];
11328: TvarVind[ncovv]=k;
11329: ncovv++; /* Varying variables without age */
11330: TvarV[ncovv]=Tvar[k];
11331: TvarVind[ncovv]=k;
11332: }
1.339 brouard 11333: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){ /* Vn is time varying dummy */
1.240 brouard 11334: if(Tvard[k1][2] <=ncovcol){
11335: Fixed[k]= 1;
11336: Dummy[k]= 1;
11337: modell[k].maintype= VTYPE;
11338: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
11339: ncovv++; /* Varying variables without age */
11340: TvarV[ncovv]=Tvar[k];
11341: TvarVind[ncovv]=k;
11342: }else if(Tvard[k1][2] <=ncovcol+nqv){
11343: Fixed[k]= 1;
11344: Dummy[k]= 1;
11345: modell[k].maintype= VTYPE;
11346: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
11347: ncovv++; /* Varying variables without age */
11348: TvarV[ncovv]=Tvar[k];
11349: TvarVind[ncovv]=k;
11350: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11351: Fixed[k]= 1;
11352: Dummy[k]= 0;
11353: modell[k].maintype= VTYPE;
11354: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
11355: ncovv++; /* Varying variables without age */
11356: TvarV[ncovv]=Tvar[k];
11357: TvarVind[ncovv]=k;
11358: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11359: Fixed[k]= 1;
11360: Dummy[k]= 1;
11361: modell[k].maintype= VTYPE;
11362: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
11363: ncovv++; /* Varying variables without age */
11364: TvarV[ncovv]=Tvar[k];
11365: TvarVind[ncovv]=k;
11366: }
1.339 brouard 11367: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){ /* Vn is time varying quanti */
1.240 brouard 11368: if(Tvard[k1][2] <=ncovcol){
11369: Fixed[k]= 1;
11370: Dummy[k]= 1;
11371: modell[k].maintype= VTYPE;
11372: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
11373: ncovv++; /* Varying variables without age */
11374: TvarV[ncovv]=Tvar[k];
11375: TvarVind[ncovv]=k;
11376: }else if(Tvard[k1][2] <=ncovcol+nqv){
11377: Fixed[k]= 1;
11378: Dummy[k]= 1;
11379: modell[k].maintype= VTYPE;
11380: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
11381: ncovv++; /* Varying variables without age */
11382: TvarV[ncovv]=Tvar[k];
11383: TvarVind[ncovv]=k;
11384: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11385: Fixed[k]= 1;
11386: Dummy[k]= 1;
11387: modell[k].maintype= VTYPE;
11388: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
11389: ncovv++; /* Varying variables without age */
11390: TvarV[ncovv]=Tvar[k];
11391: TvarVind[ncovv]=k;
11392: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11393: Fixed[k]= 1;
11394: Dummy[k]= 1;
11395: modell[k].maintype= VTYPE;
11396: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
11397: ncovv++; /* Varying variables without age */
11398: TvarV[ncovv]=Tvar[k];
11399: TvarVind[ncovv]=k;
11400: }
1.227 brouard 11401: }else{
1.240 brouard 11402: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11403: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11404: } /*end k1*/
1.225 brouard 11405: }else{
1.226 brouard 11406: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
11407: 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 11408: }
1.342 ! brouard 11409: /* 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]); */
! 11410: /* printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype); */
1.227 brouard 11411: 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]);
11412: }
11413: /* Searching for doublons in the model */
11414: for(k1=1; k1<= cptcovt;k1++){
11415: for(k2=1; k2 <k1;k2++){
1.285 brouard 11416: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
11417: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 11418: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
11419: if(Tvar[k1]==Tvar[k2]){
1.338 brouard 11420: 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]);
11421: 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 11422: return(1);
11423: }
11424: }else if (Typevar[k1] ==2){
11425: k3=Tposprod[k1];
11426: k4=Tposprod[k2];
11427: 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 11428: 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]]);
11429: 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 11430: return(1);
11431: }
11432: }
1.227 brouard 11433: }
11434: }
1.225 brouard 11435: }
11436: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
11437: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 11438: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
11439: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 11440: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 11441: /*endread:*/
1.225 brouard 11442: printf("Exiting decodemodel: ");
11443: return (1);
1.136 brouard 11444: }
11445:
1.169 brouard 11446: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 11447: {/* Check ages at death */
1.136 brouard 11448: int i, m;
1.218 brouard 11449: int firstone=0;
11450:
1.136 brouard 11451: for (i=1; i<=imx; i++) {
11452: for(m=2; (m<= maxwav); m++) {
11453: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
11454: anint[m][i]=9999;
1.216 brouard 11455: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
11456: s[m][i]=-1;
1.136 brouard 11457: }
11458: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 11459: *nberr = *nberr + 1;
1.218 brouard 11460: if(firstone == 0){
11461: firstone=1;
1.260 brouard 11462: 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 11463: }
1.262 brouard 11464: 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 11465: s[m][i]=-1; /* Droping the death status */
1.136 brouard 11466: }
11467: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 11468: (*nberr)++;
1.259 brouard 11469: 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 11470: 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 11471: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 11472: }
11473: }
11474: }
11475:
11476: for (i=1; i<=imx; i++) {
11477: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
11478: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 11479: 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 11480: if (s[m][i] >= nlstate+1) {
1.169 brouard 11481: if(agedc[i]>0){
11482: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 11483: agev[m][i]=agedc[i];
1.214 brouard 11484: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 11485: }else {
1.136 brouard 11486: if ((int)andc[i]!=9999){
11487: nbwarn++;
11488: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
11489: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
11490: agev[m][i]=-1;
11491: }
11492: }
1.169 brouard 11493: } /* agedc > 0 */
1.214 brouard 11494: } /* end if */
1.136 brouard 11495: else if(s[m][i] !=9){ /* Standard case, age in fractional
11496: years but with the precision of a month */
11497: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
11498: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
11499: agev[m][i]=1;
11500: else if(agev[m][i] < *agemin){
11501: *agemin=agev[m][i];
11502: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
11503: }
11504: else if(agev[m][i] >*agemax){
11505: *agemax=agev[m][i];
1.156 brouard 11506: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 11507: }
11508: /*agev[m][i]=anint[m][i]-annais[i];*/
11509: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 11510: } /* en if 9*/
1.136 brouard 11511: else { /* =9 */
1.214 brouard 11512: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 11513: agev[m][i]=1;
11514: s[m][i]=-1;
11515: }
11516: }
1.214 brouard 11517: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 11518: agev[m][i]=1;
1.214 brouard 11519: else{
11520: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11521: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11522: agev[m][i]=0;
11523: }
11524: } /* End for lastpass */
11525: }
1.136 brouard 11526:
11527: for (i=1; i<=imx; i++) {
11528: for(m=firstpass; (m<=lastpass); m++){
11529: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 11530: (*nberr)++;
1.136 brouard 11531: 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);
11532: 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);
11533: return 1;
11534: }
11535: }
11536: }
11537:
11538: /*for (i=1; i<=imx; i++){
11539: for (m=firstpass; (m<lastpass); m++){
11540: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
11541: }
11542:
11543: }*/
11544:
11545:
1.139 brouard 11546: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
11547: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 11548:
11549: return (0);
1.164 brouard 11550: /* endread:*/
1.136 brouard 11551: printf("Exiting calandcheckages: ");
11552: return (1);
11553: }
11554:
1.172 brouard 11555: #if defined(_MSC_VER)
11556: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11557: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11558: //#include "stdafx.h"
11559: //#include <stdio.h>
11560: //#include <tchar.h>
11561: //#include <windows.h>
11562: //#include <iostream>
11563: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
11564:
11565: LPFN_ISWOW64PROCESS fnIsWow64Process;
11566:
11567: BOOL IsWow64()
11568: {
11569: BOOL bIsWow64 = FALSE;
11570:
11571: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
11572: // (HANDLE, PBOOL);
11573:
11574: //LPFN_ISWOW64PROCESS fnIsWow64Process;
11575:
11576: HMODULE module = GetModuleHandle(_T("kernel32"));
11577: const char funcName[] = "IsWow64Process";
11578: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
11579: GetProcAddress(module, funcName);
11580:
11581: if (NULL != fnIsWow64Process)
11582: {
11583: if (!fnIsWow64Process(GetCurrentProcess(),
11584: &bIsWow64))
11585: //throw std::exception("Unknown error");
11586: printf("Unknown error\n");
11587: }
11588: return bIsWow64 != FALSE;
11589: }
11590: #endif
1.177 brouard 11591:
1.191 brouard 11592: void syscompilerinfo(int logged)
1.292 brouard 11593: {
11594: #include <stdint.h>
11595:
11596: /* #include "syscompilerinfo.h"*/
1.185 brouard 11597: /* command line Intel compiler 32bit windows, XP compatible:*/
11598: /* /GS /W3 /Gy
11599: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
11600: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
11601: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 11602: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
11603: */
11604: /* 64 bits */
1.185 brouard 11605: /*
11606: /GS /W3 /Gy
11607: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
11608: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
11609: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
11610: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
11611: /* Optimization are useless and O3 is slower than O2 */
11612: /*
11613: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
11614: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
11615: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
11616: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
11617: */
1.186 brouard 11618: /* Link is */ /* /OUT:"visual studio
1.185 brouard 11619: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
11620: /PDB:"visual studio
11621: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
11622: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
11623: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
11624: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
11625: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
11626: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
11627: uiAccess='false'"
11628: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
11629: /NOLOGO /TLBID:1
11630: */
1.292 brouard 11631:
11632:
1.177 brouard 11633: #if defined __INTEL_COMPILER
1.178 brouard 11634: #if defined(__GNUC__)
11635: struct utsname sysInfo; /* For Intel on Linux and OS/X */
11636: #endif
1.177 brouard 11637: #elif defined(__GNUC__)
1.179 brouard 11638: #ifndef __APPLE__
1.174 brouard 11639: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 11640: #endif
1.177 brouard 11641: struct utsname sysInfo;
1.178 brouard 11642: int cross = CROSS;
11643: if (cross){
11644: printf("Cross-");
1.191 brouard 11645: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 11646: }
1.174 brouard 11647: #endif
11648:
1.191 brouard 11649: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 11650: #if defined(__clang__)
1.191 brouard 11651: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 11652: #endif
11653: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 11654: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 11655: #endif
11656: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 11657: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 11658: #endif
11659: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 11660: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 11661: #endif
11662: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 11663: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 11664: #endif
11665: #if defined(_MSC_VER)
1.191 brouard 11666: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 11667: #endif
11668: #if defined(__PGI)
1.191 brouard 11669: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 11670: #endif
11671: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 11672: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 11673: #endif
1.191 brouard 11674: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 11675:
1.167 brouard 11676: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
11677: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
11678: // Windows (x64 and x86)
1.191 brouard 11679: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 11680: #elif __unix__ // all unices, not all compilers
11681: // Unix
1.191 brouard 11682: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 11683: #elif __linux__
11684: // linux
1.191 brouard 11685: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 11686: #elif __APPLE__
1.174 brouard 11687: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 11688: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 11689: #endif
11690:
11691: /* __MINGW32__ */
11692: /* __CYGWIN__ */
11693: /* __MINGW64__ */
11694: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
11695: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
11696: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
11697: /* _WIN64 // Defined for applications for Win64. */
11698: /* _M_X64 // Defined for compilations that target x64 processors. */
11699: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 11700:
1.167 brouard 11701: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 11702: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 11703: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 11704: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 11705: #else
1.191 brouard 11706: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 11707: #endif
11708:
1.169 brouard 11709: #if defined(__GNUC__)
11710: # if defined(__GNUC_PATCHLEVEL__)
11711: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11712: + __GNUC_MINOR__ * 100 \
11713: + __GNUC_PATCHLEVEL__)
11714: # else
11715: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11716: + __GNUC_MINOR__ * 100)
11717: # endif
1.174 brouard 11718: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 11719: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 11720:
11721: if (uname(&sysInfo) != -1) {
11722: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 11723: 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 11724: }
11725: else
11726: perror("uname() error");
1.179 brouard 11727: //#ifndef __INTEL_COMPILER
11728: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 11729: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 11730: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 11731: #endif
1.169 brouard 11732: #endif
1.172 brouard 11733:
1.286 brouard 11734: // void main ()
1.172 brouard 11735: // {
1.169 brouard 11736: #if defined(_MSC_VER)
1.174 brouard 11737: if (IsWow64()){
1.191 brouard 11738: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
11739: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 11740: }
11741: else{
1.191 brouard 11742: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
11743: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 11744: }
1.172 brouard 11745: // printf("\nPress Enter to continue...");
11746: // getchar();
11747: // }
11748:
1.169 brouard 11749: #endif
11750:
1.167 brouard 11751:
1.219 brouard 11752: }
1.136 brouard 11753:
1.219 brouard 11754: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 11755: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.332 brouard 11756: /* Computes the prevalence limit for each combination of the dummy covariates */
1.235 brouard 11757: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 11758: /* double ftolpl = 1.e-10; */
1.180 brouard 11759: double age, agebase, agelim;
1.203 brouard 11760: double tot;
1.180 brouard 11761:
1.202 brouard 11762: strcpy(filerespl,"PL_");
11763: strcat(filerespl,fileresu);
11764: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 11765: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
11766: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 11767: }
1.288 brouard 11768: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
11769: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 11770: pstamp(ficrespl);
1.288 brouard 11771: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 11772: fprintf(ficrespl,"#Age ");
11773: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
11774: fprintf(ficrespl,"\n");
1.180 brouard 11775:
1.219 brouard 11776: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 11777:
1.219 brouard 11778: agebase=ageminpar;
11779: agelim=agemaxpar;
1.180 brouard 11780:
1.227 brouard 11781: /* i1=pow(2,ncoveff); */
1.234 brouard 11782: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 11783: if (cptcovn < 1){i1=1;}
1.180 brouard 11784:
1.337 brouard 11785: /* for(k=1; k<=i1;k++){ /\* For each combination k of dummy covariates in the model *\/ */
1.238 brouard 11786: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 11787: k=TKresult[nres];
1.338 brouard 11788: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 11789: /* if(i1 != 1 && TKresult[nres]!= k) /\* We found the combination k corresponding to the resultline value of dummies *\/ */
11790: /* continue; */
1.235 brouard 11791:
1.238 brouard 11792: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11793: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
11794: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
11795: /* k=k+1; */
11796: /* to clean */
1.332 brouard 11797: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11798: fprintf(ficrespl,"#******");
11799: printf("#******");
11800: fprintf(ficlog,"#******");
1.337 brouard 11801: 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 11802: /* fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Here problem for varying dummy*\/ */
1.337 brouard 11803: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11804: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11805: fprintf(ficrespl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11806: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11807: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11808: }
11809: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
11810: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11811: /* fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11812: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11813: /* } */
1.238 brouard 11814: fprintf(ficrespl,"******\n");
11815: printf("******\n");
11816: fprintf(ficlog,"******\n");
11817: if(invalidvarcomb[k]){
11818: printf("\nCombination (%d) ignored because no case \n",k);
11819: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
11820: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
11821: continue;
11822: }
1.219 brouard 11823:
1.238 brouard 11824: fprintf(ficrespl,"#Age ");
1.337 brouard 11825: /* for(j=1;j<=cptcoveff;j++) { */
11826: /* fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11827: /* } */
11828: for(j=1;j<=cptcovs;j++) { /* New the quanti variable is added */
11829: fprintf(ficrespl,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 11830: }
11831: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
11832: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 11833:
1.238 brouard 11834: for (age=agebase; age<=agelim; age++){
11835: /* for (age=agebase; age<=agebase; age++){ */
1.337 brouard 11836: /**< Computes the prevalence limit in each live state at age x and for covariate combination (k and) nres */
11837: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres); /* Nicely done */
1.238 brouard 11838: fprintf(ficrespl,"%.0f ",age );
1.337 brouard 11839: /* for(j=1;j<=cptcoveff;j++) */
11840: /* fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11841: for(j=1;j<=cptcovs;j++)
11842: fprintf(ficrespl,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 11843: tot=0.;
11844: for(i=1; i<=nlstate;i++){
11845: tot += prlim[i][i];
11846: fprintf(ficrespl," %.5f", prlim[i][i]);
11847: }
11848: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
11849: } /* Age */
11850: /* was end of cptcod */
1.337 brouard 11851: } /* nres */
11852: /* } /\* for each combination *\/ */
1.219 brouard 11853: return 0;
1.180 brouard 11854: }
11855:
1.218 brouard 11856: 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 11857: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 11858:
11859: /* Computes the back prevalence limit for any combination of covariate values
11860: * at any age between ageminpar and agemaxpar
11861: */
1.235 brouard 11862: int i, j, k, i1, nres=0 ;
1.217 brouard 11863: /* double ftolpl = 1.e-10; */
11864: double age, agebase, agelim;
11865: double tot;
1.218 brouard 11866: /* double ***mobaverage; */
11867: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 11868:
11869: strcpy(fileresplb,"PLB_");
11870: strcat(fileresplb,fileresu);
11871: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 11872: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
11873: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 11874: }
1.288 brouard 11875: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
11876: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 11877: pstamp(ficresplb);
1.288 brouard 11878: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 11879: fprintf(ficresplb,"#Age ");
11880: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
11881: fprintf(ficresplb,"\n");
11882:
1.218 brouard 11883:
11884: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
11885:
11886: agebase=ageminpar;
11887: agelim=agemaxpar;
11888:
11889:
1.227 brouard 11890: i1=pow(2,cptcoveff);
1.218 brouard 11891: if (cptcovn < 1){i1=1;}
1.227 brouard 11892:
1.238 brouard 11893: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.338 brouard 11894: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
11895: k=TKresult[nres];
11896: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
11897: /* if(i1 != 1 && TKresult[nres]!= k) */
11898: /* continue; */
11899: /* /\*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*\/ */
1.238 brouard 11900: fprintf(ficresplb,"#******");
11901: printf("#******");
11902: fprintf(ficlog,"#******");
1.338 brouard 11903: 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) */
11904: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11905: fprintf(ficresplb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11906: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 11907: }
1.338 brouard 11908: /* for(j=1;j<=cptcoveff ;j++) {/\* all covariates *\/ */
11909: /* fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11910: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11911: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11912: /* } */
11913: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
11914: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
11915: /* fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
11916: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
11917: /* } */
1.238 brouard 11918: fprintf(ficresplb,"******\n");
11919: printf("******\n");
11920: fprintf(ficlog,"******\n");
11921: if(invalidvarcomb[k]){
11922: printf("\nCombination (%d) ignored because no cases \n",k);
11923: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
11924: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
11925: continue;
11926: }
1.218 brouard 11927:
1.238 brouard 11928: fprintf(ficresplb,"#Age ");
1.338 brouard 11929: for(j=1;j<=cptcovs;j++) {
11930: fprintf(ficresplb,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 11931: }
11932: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
11933: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 11934:
11935:
1.238 brouard 11936: for (age=agebase; age<=agelim; age++){
11937: /* for (age=agebase; age<=agebase; age++){ */
11938: if(mobilavproj > 0){
11939: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
11940: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11941: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 11942: }else if (mobilavproj == 0){
11943: 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);
11944: 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);
11945: exit(1);
11946: }else{
11947: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11948: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 11949: /* printf("TOTOT\n"); */
11950: /* exit(1); */
1.238 brouard 11951: }
11952: fprintf(ficresplb,"%.0f ",age );
1.338 brouard 11953: for(j=1;j<=cptcovs;j++)
11954: fprintf(ficresplb,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 11955: tot=0.;
11956: for(i=1; i<=nlstate;i++){
11957: tot += bprlim[i][i];
11958: fprintf(ficresplb," %.5f", bprlim[i][i]);
11959: }
11960: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
11961: } /* Age */
11962: /* was end of cptcod */
1.255 brouard 11963: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.338 brouard 11964: /* } /\* end of any combination *\/ */
1.238 brouard 11965: } /* end of nres */
1.218 brouard 11966: /* hBijx(p, bage, fage); */
11967: /* fclose(ficrespijb); */
11968:
11969: return 0;
1.217 brouard 11970: }
1.218 brouard 11971:
1.180 brouard 11972: int hPijx(double *p, int bage, int fage){
11973: /*------------- h Pij x at various ages ------------*/
1.336 brouard 11974: /* to be optimized with precov */
1.180 brouard 11975: int stepsize;
11976: int agelim;
11977: int hstepm;
11978: int nhstepm;
1.235 brouard 11979: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 11980:
11981: double agedeb;
11982: double ***p3mat;
11983:
1.337 brouard 11984: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
11985: if((ficrespij=fopen(filerespij,"w"))==NULL) {
11986: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
11987: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
11988: }
11989: printf("Computing pij: result on file '%s' \n", filerespij);
11990: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
11991:
11992: stepsize=(int) (stepm+YEARM-1)/YEARM;
11993: /*if (stepm<=24) stepsize=2;*/
11994:
11995: agelim=AGESUP;
11996: hstepm=stepsize*YEARM; /* Every year of age */
11997: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
11998:
11999: /* hstepm=1; aff par mois*/
12000: pstamp(ficrespij);
12001: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
12002: i1= pow(2,cptcoveff);
12003: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
12004: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
12005: /* k=k+1; */
12006: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
12007: k=TKresult[nres];
1.338 brouard 12008: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 12009: /* for(k=1; k<=i1;k++){ */
12010: /* if(i1 != 1 && TKresult[nres]!= k) */
12011: /* continue; */
12012: fprintf(ficrespij,"\n#****** ");
12013: for(j=1;j<=cptcovs;j++){
12014: fprintf(ficrespij," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12015: /* fprintf(ficrespij,"@wV%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12016: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
12017: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12018: /* fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12019: }
12020: fprintf(ficrespij,"******\n");
12021:
12022: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
12023: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
12024: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
12025:
12026: /* nhstepm=nhstepm*YEARM; aff par mois*/
12027:
12028: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12029: oldm=oldms;savm=savms;
12030: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
12031: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
12032: for(i=1; i<=nlstate;i++)
12033: for(j=1; j<=nlstate+ndeath;j++)
12034: fprintf(ficrespij," %1d-%1d",i,j);
12035: fprintf(ficrespij,"\n");
12036: for (h=0; h<=nhstepm; h++){
12037: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
12038: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.183 brouard 12039: for(i=1; i<=nlstate;i++)
12040: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 12041: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.183 brouard 12042: fprintf(ficrespij,"\n");
12043: }
1.337 brouard 12044: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12045: fprintf(ficrespij,"\n");
1.180 brouard 12046: }
1.337 brouard 12047: }
12048: /*}*/
12049: return 0;
1.180 brouard 12050: }
1.218 brouard 12051:
12052: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 12053: /*------------- h Bij x at various ages ------------*/
1.336 brouard 12054: /* To be optimized with precov */
1.217 brouard 12055: int stepsize;
1.218 brouard 12056: /* int agelim; */
12057: int ageminl;
1.217 brouard 12058: int hstepm;
12059: int nhstepm;
1.238 brouard 12060: int h, i, i1, j, k, nres;
1.218 brouard 12061:
1.217 brouard 12062: double agedeb;
12063: double ***p3mat;
1.218 brouard 12064:
12065: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
12066: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
12067: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
12068: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
12069: }
12070: printf("Computing pij back: result on file '%s' \n", filerespijb);
12071: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
12072:
12073: stepsize=(int) (stepm+YEARM-1)/YEARM;
12074: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 12075:
1.218 brouard 12076: /* agelim=AGESUP; */
1.289 brouard 12077: ageminl=AGEINF; /* was 30 */
1.218 brouard 12078: hstepm=stepsize*YEARM; /* Every year of age */
12079: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
12080:
12081: /* hstepm=1; aff par mois*/
12082: pstamp(ficrespijb);
1.255 brouard 12083: 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 12084: i1= pow(2,cptcoveff);
1.218 brouard 12085: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
12086: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
12087: /* k=k+1; */
1.238 brouard 12088: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 12089: k=TKresult[nres];
1.338 brouard 12090: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 12091: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
12092: /* if(i1 != 1 && TKresult[nres]!= k) */
12093: /* continue; */
12094: fprintf(ficrespijb,"\n#****** ");
12095: for(j=1;j<=cptcovs;j++){
1.338 brouard 12096: fprintf(ficrespijb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.337 brouard 12097: /* for(j=1;j<=cptcoveff;j++) */
12098: /* fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12099: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
12100: /* fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12101: }
12102: fprintf(ficrespijb,"******\n");
12103: if(invalidvarcomb[k]){ /* Is it necessary here? */
12104: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
12105: continue;
12106: }
12107:
12108: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
12109: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
12110: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
12111: 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 */
12112: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
12113:
12114: /* nhstepm=nhstepm*YEARM; aff par mois*/
12115:
12116: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
12117: /* and memory limitations if stepm is small */
12118:
12119: /* oldm=oldms;savm=savms; */
12120: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
12121: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
12122: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
12123: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
12124: for(i=1; i<=nlstate;i++)
12125: for(j=1; j<=nlstate+ndeath;j++)
12126: fprintf(ficrespijb," %1d-%1d",i,j);
12127: fprintf(ficrespijb,"\n");
12128: for (h=0; h<=nhstepm; h++){
12129: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
12130: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
12131: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
1.217 brouard 12132: for(i=1; i<=nlstate;i++)
12133: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 12134: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.217 brouard 12135: fprintf(ficrespijb,"\n");
1.337 brouard 12136: }
12137: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12138: fprintf(ficrespijb,"\n");
12139: } /* end age deb */
12140: /* } /\* end combination *\/ */
1.238 brouard 12141: } /* end nres */
1.218 brouard 12142: return 0;
12143: } /* hBijx */
1.217 brouard 12144:
1.180 brouard 12145:
1.136 brouard 12146: /***********************************************/
12147: /**************** Main Program *****************/
12148: /***********************************************/
12149:
12150: int main(int argc, char *argv[])
12151: {
12152: #ifdef GSL
12153: const gsl_multimin_fminimizer_type *T;
12154: size_t iteri = 0, it;
12155: int rval = GSL_CONTINUE;
12156: int status = GSL_SUCCESS;
12157: double ssval;
12158: #endif
12159: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 12160: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
12161: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 12162: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 12163: int jj, ll, li, lj, lk;
1.136 brouard 12164: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 12165: int num_filled;
1.136 brouard 12166: int itimes;
12167: int NDIM=2;
12168: int vpopbased=0;
1.235 brouard 12169: int nres=0;
1.258 brouard 12170: int endishere=0;
1.277 brouard 12171: int noffset=0;
1.274 brouard 12172: int ncurrv=0; /* Temporary variable */
12173:
1.164 brouard 12174: char ca[32], cb[32];
1.136 brouard 12175: /* FILE *fichtm; *//* Html File */
12176: /* FILE *ficgp;*/ /*Gnuplot File */
12177: struct stat info;
1.191 brouard 12178: double agedeb=0.;
1.194 brouard 12179:
12180: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 12181: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 12182:
1.165 brouard 12183: double fret;
1.191 brouard 12184: double dum=0.; /* Dummy variable */
1.136 brouard 12185: double ***p3mat;
1.218 brouard 12186: /* double ***mobaverage; */
1.319 brouard 12187: double wald;
1.164 brouard 12188:
12189: char line[MAXLINE];
1.197 brouard 12190: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
12191:
1.234 brouard 12192: char modeltemp[MAXLINE];
1.332 brouard 12193: char resultline[MAXLINE], resultlineori[MAXLINE];
1.230 brouard 12194:
1.136 brouard 12195: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 12196: char *tok, *val; /* pathtot */
1.334 brouard 12197: /* int firstobs=1, lastobs=10; /\* nobs = lastobs-firstobs declared globally ;*\/ */
1.195 brouard 12198: int c, h , cpt, c2;
1.191 brouard 12199: int jl=0;
12200: int i1, j1, jk, stepsize=0;
1.194 brouard 12201: int count=0;
12202:
1.164 brouard 12203: int *tab;
1.136 brouard 12204: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 12205: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
12206: /* double anprojf, mprojf, jprojf; */
12207: /* double jintmean,mintmean,aintmean; */
12208: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
12209: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
12210: double yrfproj= 10.0; /* Number of years of forward projections */
12211: double yrbproj= 10.0; /* Number of years of backward projections */
12212: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 12213: int mobilav=0,popforecast=0;
1.191 brouard 12214: int hstepm=0, nhstepm=0;
1.136 brouard 12215: int agemortsup;
12216: float sumlpop=0.;
12217: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
12218: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
12219:
1.191 brouard 12220: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 12221: double ftolpl=FTOL;
12222: double **prlim;
1.217 brouard 12223: double **bprlim;
1.317 brouard 12224: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
12225: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 12226: double ***paramstart; /* Matrix of starting parameter values */
12227: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 12228: double **matcov; /* Matrix of covariance */
1.203 brouard 12229: double **hess; /* Hessian matrix */
1.136 brouard 12230: double ***delti3; /* Scale */
12231: double *delti; /* Scale */
12232: double ***eij, ***vareij;
12233: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 12234:
1.136 brouard 12235: double *epj, vepp;
1.164 brouard 12236:
1.273 brouard 12237: double dateprev1, dateprev2;
1.296 brouard 12238: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
12239: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
12240:
1.217 brouard 12241:
1.136 brouard 12242: double **ximort;
1.145 brouard 12243: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 12244: int *dcwave;
12245:
1.164 brouard 12246: char z[1]="c";
1.136 brouard 12247:
12248: /*char *strt;*/
12249: char strtend[80];
1.126 brouard 12250:
1.164 brouard 12251:
1.126 brouard 12252: /* setlocale (LC_ALL, ""); */
12253: /* bindtextdomain (PACKAGE, LOCALEDIR); */
12254: /* textdomain (PACKAGE); */
12255: /* setlocale (LC_CTYPE, ""); */
12256: /* setlocale (LC_MESSAGES, ""); */
12257:
12258: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 12259: rstart_time = time(NULL);
12260: /* (void) gettimeofday(&start_time,&tzp);*/
12261: start_time = *localtime(&rstart_time);
1.126 brouard 12262: curr_time=start_time;
1.157 brouard 12263: /*tml = *localtime(&start_time.tm_sec);*/
12264: /* strcpy(strstart,asctime(&tml)); */
12265: strcpy(strstart,asctime(&start_time));
1.126 brouard 12266:
12267: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 12268: /* tp.tm_sec = tp.tm_sec +86400; */
12269: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 12270: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
12271: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
12272: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 12273: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 12274: /* strt=asctime(&tmg); */
12275: /* printf("Time(after) =%s",strstart); */
12276: /* (void) time (&time_value);
12277: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
12278: * tm = *localtime(&time_value);
12279: * strstart=asctime(&tm);
12280: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
12281: */
12282:
12283: nberr=0; /* Number of errors and warnings */
12284: nbwarn=0;
1.184 brouard 12285: #ifdef WIN32
12286: _getcwd(pathcd, size);
12287: #else
1.126 brouard 12288: getcwd(pathcd, size);
1.184 brouard 12289: #endif
1.191 brouard 12290: syscompilerinfo(0);
1.196 brouard 12291: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 12292: if(argc <=1){
12293: printf("\nEnter the parameter file name: ");
1.205 brouard 12294: if(!fgets(pathr,FILENAMELENGTH,stdin)){
12295: printf("ERROR Empty parameter file name\n");
12296: goto end;
12297: }
1.126 brouard 12298: i=strlen(pathr);
12299: if(pathr[i-1]=='\n')
12300: pathr[i-1]='\0';
1.156 brouard 12301: i=strlen(pathr);
1.205 brouard 12302: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 12303: pathr[i-1]='\0';
1.205 brouard 12304: }
12305: i=strlen(pathr);
12306: if( i==0 ){
12307: printf("ERROR Empty parameter file name\n");
12308: goto end;
12309: }
12310: for (tok = pathr; tok != NULL; ){
1.126 brouard 12311: printf("Pathr |%s|\n",pathr);
12312: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
12313: printf("val= |%s| pathr=%s\n",val,pathr);
12314: strcpy (pathtot, val);
12315: if(pathr[0] == '\0') break; /* Dirty */
12316: }
12317: }
1.281 brouard 12318: else if (argc<=2){
12319: strcpy(pathtot,argv[1]);
12320: }
1.126 brouard 12321: else{
12322: strcpy(pathtot,argv[1]);
1.281 brouard 12323: strcpy(z,argv[2]);
12324: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 12325: }
12326: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
12327: /*cygwin_split_path(pathtot,path,optionfile);
12328: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
12329: /* cutv(path,optionfile,pathtot,'\\');*/
12330:
12331: /* Split argv[0], imach program to get pathimach */
12332: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
12333: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12334: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12335: /* strcpy(pathimach,argv[0]); */
12336: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
12337: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
12338: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 12339: #ifdef WIN32
12340: _chdir(path); /* Can be a relative path */
12341: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
12342: #else
1.126 brouard 12343: chdir(path); /* Can be a relative path */
1.184 brouard 12344: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
12345: #endif
12346: printf("Current directory %s!\n",pathcd);
1.126 brouard 12347: strcpy(command,"mkdir ");
12348: strcat(command,optionfilefiname);
12349: if((outcmd=system(command)) != 0){
1.169 brouard 12350: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 12351: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
12352: /* fclose(ficlog); */
12353: /* exit(1); */
12354: }
12355: /* if((imk=mkdir(optionfilefiname))<0){ */
12356: /* perror("mkdir"); */
12357: /* } */
12358:
12359: /*-------- arguments in the command line --------*/
12360:
1.186 brouard 12361: /* Main Log file */
1.126 brouard 12362: strcat(filelog, optionfilefiname);
12363: strcat(filelog,".log"); /* */
12364: if((ficlog=fopen(filelog,"w"))==NULL) {
12365: printf("Problem with logfile %s\n",filelog);
12366: goto end;
12367: }
12368: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 12369: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 12370: fprintf(ficlog,"\nEnter the parameter file name: \n");
12371: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
12372: path=%s \n\
12373: optionfile=%s\n\
12374: optionfilext=%s\n\
1.156 brouard 12375: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 12376:
1.197 brouard 12377: syscompilerinfo(1);
1.167 brouard 12378:
1.126 brouard 12379: printf("Local time (at start):%s",strstart);
12380: fprintf(ficlog,"Local time (at start): %s",strstart);
12381: fflush(ficlog);
12382: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 12383: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 12384:
12385: /* */
12386: strcpy(fileres,"r");
12387: strcat(fileres, optionfilefiname);
1.201 brouard 12388: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 12389: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 12390: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 12391:
1.186 brouard 12392: /* Main ---------arguments file --------*/
1.126 brouard 12393:
12394: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 12395: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
12396: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 12397: fflush(ficlog);
1.149 brouard 12398: /* goto end; */
12399: exit(70);
1.126 brouard 12400: }
12401:
12402: strcpy(filereso,"o");
1.201 brouard 12403: strcat(filereso,fileresu);
1.126 brouard 12404: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
12405: printf("Problem with Output resultfile: %s\n", filereso);
12406: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
12407: fflush(ficlog);
12408: goto end;
12409: }
1.278 brouard 12410: /*-------- Rewriting parameter file ----------*/
12411: strcpy(rfileres,"r"); /* "Rparameterfile */
12412: strcat(rfileres,optionfilefiname); /* Parameter file first name */
12413: strcat(rfileres,"."); /* */
12414: strcat(rfileres,optionfilext); /* Other files have txt extension */
12415: if((ficres =fopen(rfileres,"w"))==NULL) {
12416: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
12417: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
12418: fflush(ficlog);
12419: goto end;
12420: }
12421: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 12422:
1.278 brouard 12423:
1.126 brouard 12424: /* Reads comments: lines beginning with '#' */
12425: numlinepar=0;
1.277 brouard 12426: /* Is it a BOM UTF-8 Windows file? */
12427: /* First parameter line */
1.197 brouard 12428: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 12429: noffset=0;
12430: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
12431: {
12432: noffset=noffset+3;
12433: printf("# File is an UTF8 Bom.\n"); // 0xBF
12434: }
1.302 brouard 12435: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
12436: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 12437: {
12438: noffset=noffset+2;
12439: printf("# File is an UTF16BE BOM file\n");
12440: }
12441: else if( line[0] == 0 && line[1] == 0)
12442: {
12443: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
12444: noffset=noffset+4;
12445: printf("# File is an UTF16BE BOM file\n");
12446: }
12447: } else{
12448: ;/*printf(" Not a BOM file\n");*/
12449: }
12450:
1.197 brouard 12451: /* If line starts with a # it is a comment */
1.277 brouard 12452: if (line[noffset] == '#') {
1.197 brouard 12453: numlinepar++;
12454: fputs(line,stdout);
12455: fputs(line,ficparo);
1.278 brouard 12456: fputs(line,ficres);
1.197 brouard 12457: fputs(line,ficlog);
12458: continue;
12459: }else
12460: break;
12461: }
12462: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
12463: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
12464: if (num_filled != 5) {
12465: printf("Should be 5 parameters\n");
1.283 brouard 12466: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 12467: }
1.126 brouard 12468: numlinepar++;
1.197 brouard 12469: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 12470: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12471: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12472: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 12473: }
12474: /* Second parameter line */
12475: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 12476: /* while(fscanf(ficpar,"%[^\n]", line)) { */
12477: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 12478: if (line[0] == '#') {
12479: numlinepar++;
1.283 brouard 12480: printf("%s",line);
12481: fprintf(ficres,"%s",line);
12482: fprintf(ficparo,"%s",line);
12483: fprintf(ficlog,"%s",line);
1.197 brouard 12484: continue;
12485: }else
12486: break;
12487: }
1.223 brouard 12488: 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", \
12489: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
12490: if (num_filled != 11) {
12491: 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 12492: printf("but line=%s\n",line);
1.283 brouard 12493: 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");
12494: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 12495: }
1.286 brouard 12496: if( lastpass > maxwav){
12497: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12498: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12499: fflush(ficlog);
12500: goto end;
12501: }
12502: 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 12503: 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 12504: 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 12505: 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 12506: }
1.203 brouard 12507: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 12508: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 12509: /* Third parameter line */
12510: while(fgets(line, MAXLINE, ficpar)) {
12511: /* If line starts with a # it is a comment */
12512: if (line[0] == '#') {
12513: numlinepar++;
1.283 brouard 12514: printf("%s",line);
12515: fprintf(ficres,"%s",line);
12516: fprintf(ficparo,"%s",line);
12517: fprintf(ficlog,"%s",line);
1.197 brouard 12518: continue;
12519: }else
12520: break;
12521: }
1.201 brouard 12522: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 12523: if (num_filled != 1){
1.302 brouard 12524: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
12525: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 12526: model[0]='\0';
12527: goto end;
12528: }
12529: else{
12530: if (model[0]=='+'){
12531: for(i=1; i<=strlen(model);i++)
12532: modeltemp[i-1]=model[i];
1.201 brouard 12533: strcpy(model,modeltemp);
1.197 brouard 12534: }
12535: }
1.338 brouard 12536: /* printf(" model=1+age%s modeltemp= %s, model=1+age+%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 12537: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 12538: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
12539: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
12540: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 12541: }
12542: /* 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); */
12543: /* numlinepar=numlinepar+3; /\* In general *\/ */
12544: /* 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 12545: /* 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); */
12546: /* 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 12547: fflush(ficlog);
1.190 brouard 12548: /* if(model[0]=='#'|| model[0]== '\0'){ */
12549: if(model[0]=='#'){
1.279 brouard 12550: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
12551: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
12552: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 12553: if(mle != -1){
1.279 brouard 12554: 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 12555: exit(1);
12556: }
12557: }
1.126 brouard 12558: while((c=getc(ficpar))=='#' && c!= EOF){
12559: ungetc(c,ficpar);
12560: fgets(line, MAXLINE, ficpar);
12561: numlinepar++;
1.195 brouard 12562: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
12563: z[0]=line[1];
1.342 ! brouard 12564: }else if(line[1]=='d'){ /* For debugging individual values of covariates in ficresilk */
! 12565: debugILK=1;
1.195 brouard 12566: }
12567: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 12568: fputs(line, stdout);
12569: //puts(line);
1.126 brouard 12570: fputs(line,ficparo);
12571: fputs(line,ficlog);
12572: }
12573: ungetc(c,ficpar);
12574:
12575:
1.290 brouard 12576: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
12577: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
12578: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
1.341 brouard 12579: /* if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /\**< Time varying covariate (dummy and quantitative)*\/ */
12580: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs); /**< Might be better */
1.136 brouard 12581: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
12582: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
12583: v1+v2*age+v2*v3 makes cptcovn = 3
12584: */
12585: if (strlen(model)>1)
1.187 brouard 12586: 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 12587: else
1.187 brouard 12588: ncovmodel=2; /* Constant and age */
1.133 brouard 12589: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
12590: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 12591: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
12592: 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);
12593: 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);
12594: fflush(stdout);
12595: fclose (ficlog);
12596: goto end;
12597: }
1.126 brouard 12598: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12599: delti=delti3[1][1];
12600: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
12601: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 12602: /* We could also provide initial parameters values giving by simple logistic regression
12603: * only one way, that is without matrix product. We will have nlstate maximizations */
12604: /* for(i=1;i<nlstate;i++){ */
12605: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12606: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12607: /* } */
1.126 brouard 12608: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 12609: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
12610: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12611: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12612: fclose (ficparo);
12613: fclose (ficlog);
12614: goto end;
12615: exit(0);
1.220 brouard 12616: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 12617: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 12618: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
12619: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12620: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12621: matcov=matrix(1,npar,1,npar);
1.203 brouard 12622: hess=matrix(1,npar,1,npar);
1.220 brouard 12623: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 12624: /* Read guessed parameters */
1.126 brouard 12625: /* Reads comments: lines beginning with '#' */
12626: while((c=getc(ficpar))=='#' && c!= EOF){
12627: ungetc(c,ficpar);
12628: fgets(line, MAXLINE, ficpar);
12629: numlinepar++;
1.141 brouard 12630: fputs(line,stdout);
1.126 brouard 12631: fputs(line,ficparo);
12632: fputs(line,ficlog);
12633: }
12634: ungetc(c,ficpar);
12635:
12636: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 12637: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 12638: for(i=1; i <=nlstate; i++){
1.234 brouard 12639: j=0;
1.126 brouard 12640: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 12641: if(jj==i) continue;
12642: j++;
1.292 brouard 12643: while((c=getc(ficpar))=='#' && c!= EOF){
12644: ungetc(c,ficpar);
12645: fgets(line, MAXLINE, ficpar);
12646: numlinepar++;
12647: fputs(line,stdout);
12648: fputs(line,ficparo);
12649: fputs(line,ficlog);
12650: }
12651: ungetc(c,ficpar);
1.234 brouard 12652: fscanf(ficpar,"%1d%1d",&i1,&j1);
12653: if ((i1 != i) || (j1 != jj)){
12654: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 12655: It might be a problem of design; if ncovcol and the model are correct\n \
12656: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 12657: exit(1);
12658: }
12659: fprintf(ficparo,"%1d%1d",i1,j1);
12660: if(mle==1)
12661: printf("%1d%1d",i,jj);
12662: fprintf(ficlog,"%1d%1d",i,jj);
12663: for(k=1; k<=ncovmodel;k++){
12664: fscanf(ficpar," %lf",¶m[i][j][k]);
12665: if(mle==1){
12666: printf(" %lf",param[i][j][k]);
12667: fprintf(ficlog," %lf",param[i][j][k]);
12668: }
12669: else
12670: fprintf(ficlog," %lf",param[i][j][k]);
12671: fprintf(ficparo," %lf",param[i][j][k]);
12672: }
12673: fscanf(ficpar,"\n");
12674: numlinepar++;
12675: if(mle==1)
12676: printf("\n");
12677: fprintf(ficlog,"\n");
12678: fprintf(ficparo,"\n");
1.126 brouard 12679: }
12680: }
12681: fflush(ficlog);
1.234 brouard 12682:
1.251 brouard 12683: /* Reads parameters values */
1.126 brouard 12684: p=param[1][1];
1.251 brouard 12685: pstart=paramstart[1][1];
1.126 brouard 12686:
12687: /* Reads comments: lines beginning with '#' */
12688: while((c=getc(ficpar))=='#' && c!= EOF){
12689: ungetc(c,ficpar);
12690: fgets(line, MAXLINE, ficpar);
12691: numlinepar++;
1.141 brouard 12692: fputs(line,stdout);
1.126 brouard 12693: fputs(line,ficparo);
12694: fputs(line,ficlog);
12695: }
12696: ungetc(c,ficpar);
12697:
12698: for(i=1; i <=nlstate; i++){
12699: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 12700: fscanf(ficpar,"%1d%1d",&i1,&j1);
12701: if ( (i1-i) * (j1-j) != 0){
12702: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
12703: exit(1);
12704: }
12705: printf("%1d%1d",i,j);
12706: fprintf(ficparo,"%1d%1d",i1,j1);
12707: fprintf(ficlog,"%1d%1d",i1,j1);
12708: for(k=1; k<=ncovmodel;k++){
12709: fscanf(ficpar,"%le",&delti3[i][j][k]);
12710: printf(" %le",delti3[i][j][k]);
12711: fprintf(ficparo," %le",delti3[i][j][k]);
12712: fprintf(ficlog," %le",delti3[i][j][k]);
12713: }
12714: fscanf(ficpar,"\n");
12715: numlinepar++;
12716: printf("\n");
12717: fprintf(ficparo,"\n");
12718: fprintf(ficlog,"\n");
1.126 brouard 12719: }
12720: }
12721: fflush(ficlog);
1.234 brouard 12722:
1.145 brouard 12723: /* Reads covariance matrix */
1.126 brouard 12724: delti=delti3[1][1];
1.220 brouard 12725:
12726:
1.126 brouard 12727: /* 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 12728:
1.126 brouard 12729: /* Reads comments: lines beginning with '#' */
12730: while((c=getc(ficpar))=='#' && c!= EOF){
12731: ungetc(c,ficpar);
12732: fgets(line, MAXLINE, ficpar);
12733: numlinepar++;
1.141 brouard 12734: fputs(line,stdout);
1.126 brouard 12735: fputs(line,ficparo);
12736: fputs(line,ficlog);
12737: }
12738: ungetc(c,ficpar);
1.220 brouard 12739:
1.126 brouard 12740: matcov=matrix(1,npar,1,npar);
1.203 brouard 12741: hess=matrix(1,npar,1,npar);
1.131 brouard 12742: for(i=1; i <=npar; i++)
12743: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 12744:
1.194 brouard 12745: /* Scans npar lines */
1.126 brouard 12746: for(i=1; i <=npar; i++){
1.226 brouard 12747: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 12748: if(count != 3){
1.226 brouard 12749: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12750: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12751: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12752: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12753: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12754: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12755: exit(1);
1.220 brouard 12756: }else{
1.226 brouard 12757: if(mle==1)
12758: printf("%1d%1d%d",i1,j1,jk);
12759: }
12760: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
12761: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 12762: for(j=1; j <=i; j++){
1.226 brouard 12763: fscanf(ficpar," %le",&matcov[i][j]);
12764: if(mle==1){
12765: printf(" %.5le",matcov[i][j]);
12766: }
12767: fprintf(ficlog," %.5le",matcov[i][j]);
12768: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 12769: }
12770: fscanf(ficpar,"\n");
12771: numlinepar++;
12772: if(mle==1)
1.220 brouard 12773: printf("\n");
1.126 brouard 12774: fprintf(ficlog,"\n");
12775: fprintf(ficparo,"\n");
12776: }
1.194 brouard 12777: /* End of read covariance matrix npar lines */
1.126 brouard 12778: for(i=1; i <=npar; i++)
12779: for(j=i+1;j<=npar;j++)
1.226 brouard 12780: matcov[i][j]=matcov[j][i];
1.126 brouard 12781:
12782: if(mle==1)
12783: printf("\n");
12784: fprintf(ficlog,"\n");
12785:
12786: fflush(ficlog);
12787:
12788: } /* End of mle != -3 */
1.218 brouard 12789:
1.186 brouard 12790: /* Main data
12791: */
1.290 brouard 12792: nobs=lastobs-firstobs+1; /* was = lastobs;*/
12793: /* num=lvector(1,n); */
12794: /* moisnais=vector(1,n); */
12795: /* annais=vector(1,n); */
12796: /* moisdc=vector(1,n); */
12797: /* andc=vector(1,n); */
12798: /* weight=vector(1,n); */
12799: /* agedc=vector(1,n); */
12800: /* cod=ivector(1,n); */
12801: /* for(i=1;i<=n;i++){ */
12802: num=lvector(firstobs,lastobs);
12803: moisnais=vector(firstobs,lastobs);
12804: annais=vector(firstobs,lastobs);
12805: moisdc=vector(firstobs,lastobs);
12806: andc=vector(firstobs,lastobs);
12807: weight=vector(firstobs,lastobs);
12808: agedc=vector(firstobs,lastobs);
12809: cod=ivector(firstobs,lastobs);
12810: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 12811: num[i]=0;
12812: moisnais[i]=0;
12813: annais[i]=0;
12814: moisdc[i]=0;
12815: andc[i]=0;
12816: agedc[i]=0;
12817: cod[i]=0;
12818: weight[i]=1.0; /* Equal weights, 1 by default */
12819: }
1.290 brouard 12820: mint=matrix(1,maxwav,firstobs,lastobs);
12821: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 12822: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.336 brouard 12823: /* printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel)); */
1.126 brouard 12824: tab=ivector(1,NCOVMAX);
1.144 brouard 12825: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 12826: 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 12827:
1.136 brouard 12828: /* Reads data from file datafile */
12829: if (readdata(datafile, firstobs, lastobs, &imx)==1)
12830: goto end;
12831:
12832: /* Calculation of the number of parameters from char model */
1.234 brouard 12833: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 12834: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
12835: k=3 V4 Tvar[k=3]= 4 (from V4)
12836: k=2 V1 Tvar[k=2]= 1 (from V1)
12837: k=1 Tvar[1]=2 (from V2)
1.234 brouard 12838: */
12839:
12840: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
12841: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 brouard 12842: TnsdVar=ivector(1,NCOVMAX); /* */
1.335 brouard 12843: /* for(i=1; i<=NCOVMAX;i++) TnsdVar[i]=3; */
1.234 brouard 12844: TvarsD=ivector(1,NCOVMAX); /* */
12845: TvarsQind=ivector(1,NCOVMAX); /* */
12846: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 12847: TvarF=ivector(1,NCOVMAX); /* */
12848: TvarFind=ivector(1,NCOVMAX); /* */
12849: TvarV=ivector(1,NCOVMAX); /* */
12850: TvarVind=ivector(1,NCOVMAX); /* */
12851: TvarA=ivector(1,NCOVMAX); /* */
12852: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 12853: TvarFD=ivector(1,NCOVMAX); /* */
12854: TvarFDind=ivector(1,NCOVMAX); /* */
12855: TvarFQ=ivector(1,NCOVMAX); /* */
12856: TvarFQind=ivector(1,NCOVMAX); /* */
12857: TvarVD=ivector(1,NCOVMAX); /* */
12858: TvarVDind=ivector(1,NCOVMAX); /* */
12859: TvarVQ=ivector(1,NCOVMAX); /* */
12860: TvarVQind=ivector(1,NCOVMAX); /* */
1.339 brouard 12861: TvarVV=ivector(1,NCOVMAX); /* */
12862: TvarVVind=ivector(1,NCOVMAX); /* */
1.231 brouard 12863:
1.230 brouard 12864: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 12865: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 12866: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
12867: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
12868: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 12869: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
12870: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
12871: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
12872: */
12873: /* For model-covariate k tells which data-covariate to use but
12874: because this model-covariate is a construction we invent a new column
12875: ncovcol + k1
12876: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
12877: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 12878: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
12879: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 12880: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
12881: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 12882: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 12883: */
1.145 brouard 12884: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
12885: 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 12886: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
12887: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.330 brouard 12888: Tvardk=imatrix(1,NCOVMAX,1,2);
1.145 brouard 12889: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 12890: 4 covariates (3 plus signs)
12891: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 12892: */
12893: for(i=1;i<NCOVMAX;i++)
12894: Tage[i]=0;
1.230 brouard 12895: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 12896: * individual dummy, fixed or varying:
12897: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
12898: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 12899: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
12900: * V1 df, V2 qf, V3 & V4 dv, V5 qv
12901: * Tmodelind[1]@9={9,0,3,2,}*/
12902: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
12903: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 12904: * individual quantitative, fixed or varying:
12905: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
12906: * 3, 1, 0, 0, 0, 0, 0, 0},
12907: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 12908: /* Main decodemodel */
12909:
1.187 brouard 12910:
1.223 brouard 12911: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 12912: goto end;
12913:
1.137 brouard 12914: if((double)(lastobs-imx)/(double)imx > 1.10){
12915: nbwarn++;
12916: 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);
12917: 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);
12918: }
1.136 brouard 12919: /* if(mle==1){*/
1.137 brouard 12920: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
12921: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 12922: }
12923:
12924: /*-calculation of age at interview from date of interview and age at death -*/
12925: agev=matrix(1,maxwav,1,imx);
12926:
12927: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
12928: goto end;
12929:
1.126 brouard 12930:
1.136 brouard 12931: agegomp=(int)agemin;
1.290 brouard 12932: free_vector(moisnais,firstobs,lastobs);
12933: free_vector(annais,firstobs,lastobs);
1.126 brouard 12934: /* free_matrix(mint,1,maxwav,1,n);
12935: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 12936: /* free_vector(moisdc,1,n); */
12937: /* free_vector(andc,1,n); */
1.145 brouard 12938: /* */
12939:
1.126 brouard 12940: wav=ivector(1,imx);
1.214 brouard 12941: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
12942: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
12943: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
12944: 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.*/
12945: bh=imatrix(1,lastpass-firstpass+2,1,imx);
12946: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 12947:
12948: /* Concatenates waves */
1.214 brouard 12949: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
12950: Death is a valid wave (if date is known).
12951: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
12952: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
12953: and mw[mi+1][i]. dh depends on stepm.
12954: */
12955:
1.126 brouard 12956: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 12957: /* Concatenates waves */
1.145 brouard 12958:
1.290 brouard 12959: free_vector(moisdc,firstobs,lastobs);
12960: free_vector(andc,firstobs,lastobs);
1.215 brouard 12961:
1.126 brouard 12962: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
12963: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
12964: ncodemax[1]=1;
1.145 brouard 12965: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 12966: cptcoveff=0;
1.220 brouard 12967: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
1.335 brouard 12968: 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 12969: }
12970:
12971: ncovcombmax=pow(2,cptcoveff);
1.338 brouard 12972: invalidvarcomb=ivector(0, ncovcombmax);
12973: for(i=0;i<ncovcombmax;i++)
1.227 brouard 12974: invalidvarcomb[i]=0;
12975:
1.211 brouard 12976: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 12977: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 12978: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 12979:
1.200 brouard 12980: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 12981: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 12982: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 12983: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
12984: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
12985: * (currently 0 or 1) in the data.
12986: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
12987: * corresponding modality (h,j).
12988: */
12989:
1.145 brouard 12990: h=0;
12991: /*if (cptcovn > 0) */
1.126 brouard 12992: m=pow(2,cptcoveff);
12993:
1.144 brouard 12994: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 12995: * For k=4 covariates, h goes from 1 to m=2**k
12996: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
12997: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 12998: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
12999: *______________________________ *______________________
13000: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
13001: * 2 2 1 1 1 * 1 0 0 0 1
13002: * 3 i=2 1 2 1 1 * 2 0 0 1 0
13003: * 4 2 2 1 1 * 3 0 0 1 1
13004: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
13005: * 6 2 1 2 1 * 5 0 1 0 1
13006: * 7 i=4 1 2 2 1 * 6 0 1 1 0
13007: * 8 2 2 2 1 * 7 0 1 1 1
13008: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
13009: * 10 2 1 1 2 * 9 1 0 0 1
13010: * 11 i=6 1 2 1 2 * 10 1 0 1 0
13011: * 12 2 2 1 2 * 11 1 0 1 1
13012: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
13013: * 14 2 1 2 2 * 13 1 1 0 1
13014: * 15 i=8 1 2 2 2 * 14 1 1 1 0
13015: * 16 2 2 2 2 * 15 1 1 1 1
13016: */
1.212 brouard 13017: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 13018: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
13019: * and the value of each covariate?
13020: * V1=1, V2=1, V3=2, V4=1 ?
13021: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
13022: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
13023: * In order to get the real value in the data, we use nbcode
13024: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
13025: * We are keeping this crazy system in order to be able (in the future?)
13026: * to have more than 2 values (0 or 1) for a covariate.
13027: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
13028: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
13029: * bbbbbbbb
13030: * 76543210
13031: * h-1 00000101 (6-1=5)
1.219 brouard 13032: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 13033: * &
13034: * 1 00000001 (1)
1.219 brouard 13035: * 00000000 = 1 & ((h-1) >> (k-1))
13036: * +1= 00000001 =1
1.211 brouard 13037: *
13038: * h=14, k=3 => h'=h-1=13, k'=k-1=2
13039: * h' 1101 =2^3+2^2+0x2^1+2^0
13040: * >>k' 11
13041: * & 00000001
13042: * = 00000001
13043: * +1 = 00000010=2 = codtabm(14,3)
13044: * Reverse h=6 and m=16?
13045: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
13046: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
13047: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
13048: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
13049: * V3=decodtabm(14,3,2**4)=2
13050: * h'=13 1101 =2^3+2^2+0x2^1+2^0
13051: *(h-1) >> (j-1) 0011 =13 >> 2
13052: * &1 000000001
13053: * = 000000001
13054: * +1= 000000010 =2
13055: * 2211
13056: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
13057: * V3=2
1.220 brouard 13058: * codtabm and decodtabm are identical
1.211 brouard 13059: */
13060:
1.145 brouard 13061:
13062: free_ivector(Ndum,-1,NCOVMAX);
13063:
13064:
1.126 brouard 13065:
1.186 brouard 13066: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 13067: strcpy(optionfilegnuplot,optionfilefiname);
13068: if(mle==-3)
1.201 brouard 13069: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 13070: strcat(optionfilegnuplot,".gp");
13071:
13072: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
13073: printf("Problem with file %s",optionfilegnuplot);
13074: }
13075: else{
1.204 brouard 13076: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 13077: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 13078: //fprintf(ficgp,"set missing 'NaNq'\n");
13079: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 13080: }
13081: /* fclose(ficgp);*/
1.186 brouard 13082:
13083:
13084: /* Initialisation of --------- index.htm --------*/
1.126 brouard 13085:
13086: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
13087: if(mle==-3)
1.201 brouard 13088: strcat(optionfilehtm,"-MORT_");
1.126 brouard 13089: strcat(optionfilehtm,".htm");
13090: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 13091: printf("Problem with %s \n",optionfilehtm);
13092: exit(0);
1.126 brouard 13093: }
13094:
13095: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
13096: strcat(optionfilehtmcov,"-cov.htm");
13097: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
13098: printf("Problem with %s \n",optionfilehtmcov), exit(0);
13099: }
13100: else{
13101: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
13102: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 13103: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 13104: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
13105: }
13106:
1.335 brouard 13107: fprintf(fichtm,"<html><head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n\
13108: <title>IMaCh %s</title></head>\n\
13109: <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n\
13110: <font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>\
13111: -EUROREVES-Institut de longévité-2013-2022-Japan Society for the Promotion of Sciences 日本学術振興会 \
13112: (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - \
13113: <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \n", optionfilehtm);
13114:
13115: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 13116: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 13117: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.337 brouard 13118: 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 13119: \n\
13120: <hr size=\"2\" color=\"#EC5E5E\">\
13121: <ul><li><h4>Parameter files</h4>\n\
13122: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
13123: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
13124: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
13125: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
13126: - Date and time at start: %s</ul>\n",\
1.335 brouard 13127: version,fullversion,optionfilehtm,optionfilehtm,title,datafile,datafile,firstpass,lastpass,stepm, weightopt, model, \
1.126 brouard 13128: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
13129: fileres,fileres,\
13130: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
13131: fflush(fichtm);
13132:
13133: strcpy(pathr,path);
13134: strcat(pathr,optionfilefiname);
1.184 brouard 13135: #ifdef WIN32
13136: _chdir(optionfilefiname); /* Move to directory named optionfile */
13137: #else
1.126 brouard 13138: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 13139: #endif
13140:
1.126 brouard 13141:
1.220 brouard 13142: /* Calculates basic frequencies. Computes observed prevalence at single age
13143: and for any valid combination of covariates
1.126 brouard 13144: and prints on file fileres'p'. */
1.251 brouard 13145: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 13146: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 13147:
13148: fprintf(fichtm,"\n");
1.286 brouard 13149: 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 13150: ftol, stepm);
13151: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
13152: ncurrv=1;
13153: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
13154: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
13155: ncurrv=i;
13156: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 13157: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 13158: ncurrv=i;
13159: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 13160: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 13161: ncurrv=i;
13162: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
13163: 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", \
13164: nlstate, ndeath, maxwav, mle, weightopt);
13165:
13166: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
13167: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
13168:
13169:
1.317 brouard 13170: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 13171: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
13172: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 13173: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 13174: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 13175: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13176: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13177: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13178: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 13179:
1.126 brouard 13180: /* For Powell, parameters are in a vector p[] starting at p[1]
13181: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
13182: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
13183:
13184: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 13185: /* For mortality only */
1.126 brouard 13186: if (mle==-3){
1.136 brouard 13187: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 13188: for(i=1;i<=NDIM;i++)
13189: for(j=1;j<=NDIM;j++)
13190: ximort[i][j]=0.;
1.186 brouard 13191: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 13192: cens=ivector(firstobs,lastobs);
13193: ageexmed=vector(firstobs,lastobs);
13194: agecens=vector(firstobs,lastobs);
13195: dcwave=ivector(firstobs,lastobs);
1.223 brouard 13196:
1.126 brouard 13197: for (i=1; i<=imx; i++){
13198: dcwave[i]=-1;
13199: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 13200: if (s[m][i]>nlstate) {
13201: dcwave[i]=m;
13202: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
13203: break;
13204: }
1.126 brouard 13205: }
1.226 brouard 13206:
1.126 brouard 13207: for (i=1; i<=imx; i++) {
13208: if (wav[i]>0){
1.226 brouard 13209: ageexmed[i]=agev[mw[1][i]][i];
13210: j=wav[i];
13211: agecens[i]=1.;
13212:
13213: if (ageexmed[i]> 1 && wav[i] > 0){
13214: agecens[i]=agev[mw[j][i]][i];
13215: cens[i]= 1;
13216: }else if (ageexmed[i]< 1)
13217: cens[i]= -1;
13218: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
13219: cens[i]=0 ;
1.126 brouard 13220: }
13221: else cens[i]=-1;
13222: }
13223:
13224: for (i=1;i<=NDIM;i++) {
13225: for (j=1;j<=NDIM;j++)
1.226 brouard 13226: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 13227: }
13228:
1.302 brouard 13229: p[1]=0.0268; p[NDIM]=0.083;
13230: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 13231:
13232:
1.136 brouard 13233: #ifdef GSL
13234: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 13235: #else
1.126 brouard 13236: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 13237: #endif
1.201 brouard 13238: strcpy(filerespow,"POW-MORT_");
13239: strcat(filerespow,fileresu);
1.126 brouard 13240: if((ficrespow=fopen(filerespow,"w"))==NULL) {
13241: printf("Problem with resultfile: %s\n", filerespow);
13242: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
13243: }
1.136 brouard 13244: #ifdef GSL
13245: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 13246: #else
1.126 brouard 13247: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 13248: #endif
1.126 brouard 13249: /* for (i=1;i<=nlstate;i++)
13250: for(j=1;j<=nlstate+ndeath;j++)
13251: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
13252: */
13253: fprintf(ficrespow,"\n");
1.136 brouard 13254: #ifdef GSL
13255: /* gsl starts here */
13256: T = gsl_multimin_fminimizer_nmsimplex;
13257: gsl_multimin_fminimizer *sfm = NULL;
13258: gsl_vector *ss, *x;
13259: gsl_multimin_function minex_func;
13260:
13261: /* Initial vertex size vector */
13262: ss = gsl_vector_alloc (NDIM);
13263:
13264: if (ss == NULL){
13265: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
13266: }
13267: /* Set all step sizes to 1 */
13268: gsl_vector_set_all (ss, 0.001);
13269:
13270: /* Starting point */
1.126 brouard 13271:
1.136 brouard 13272: x = gsl_vector_alloc (NDIM);
13273:
13274: if (x == NULL){
13275: gsl_vector_free(ss);
13276: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
13277: }
13278:
13279: /* Initialize method and iterate */
13280: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 13281: /* gsl_vector_set(x, 0, 0.0268); */
13282: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 13283: gsl_vector_set(x, 0, p[1]);
13284: gsl_vector_set(x, 1, p[2]);
13285:
13286: minex_func.f = &gompertz_f;
13287: minex_func.n = NDIM;
13288: minex_func.params = (void *)&p; /* ??? */
13289:
13290: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
13291: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
13292:
13293: printf("Iterations beginning .....\n\n");
13294: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
13295:
13296: iteri=0;
13297: while (rval == GSL_CONTINUE){
13298: iteri++;
13299: status = gsl_multimin_fminimizer_iterate(sfm);
13300:
13301: if (status) printf("error: %s\n", gsl_strerror (status));
13302: fflush(0);
13303:
13304: if (status)
13305: break;
13306:
13307: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
13308: ssval = gsl_multimin_fminimizer_size (sfm);
13309:
13310: if (rval == GSL_SUCCESS)
13311: printf ("converged to a local maximum at\n");
13312:
13313: printf("%5d ", iteri);
13314: for (it = 0; it < NDIM; it++){
13315: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
13316: }
13317: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
13318: }
13319:
13320: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
13321:
13322: gsl_vector_free(x); /* initial values */
13323: gsl_vector_free(ss); /* inital step size */
13324: for (it=0; it<NDIM; it++){
13325: p[it+1]=gsl_vector_get(sfm->x,it);
13326: fprintf(ficrespow," %.12lf", p[it]);
13327: }
13328: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
13329: #endif
13330: #ifdef POWELL
13331: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
13332: #endif
1.126 brouard 13333: fclose(ficrespow);
13334:
1.203 brouard 13335: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 13336:
13337: for(i=1; i <=NDIM; i++)
13338: for(j=i+1;j<=NDIM;j++)
1.220 brouard 13339: matcov[i][j]=matcov[j][i];
1.126 brouard 13340:
13341: printf("\nCovariance matrix\n ");
1.203 brouard 13342: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 13343: for(i=1; i <=NDIM; i++) {
13344: for(j=1;j<=NDIM;j++){
1.220 brouard 13345: printf("%f ",matcov[i][j]);
13346: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 13347: }
1.203 brouard 13348: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 13349: }
13350:
13351: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 13352: for (i=1;i<=NDIM;i++) {
1.126 brouard 13353: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 13354: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
13355: }
1.302 brouard 13356: lsurv=vector(agegomp,AGESUP);
13357: lpop=vector(agegomp,AGESUP);
13358: tpop=vector(agegomp,AGESUP);
1.126 brouard 13359: lsurv[agegomp]=100000;
13360:
13361: for (k=agegomp;k<=AGESUP;k++) {
13362: agemortsup=k;
13363: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
13364: }
13365:
13366: for (k=agegomp;k<agemortsup;k++)
13367: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
13368:
13369: for (k=agegomp;k<agemortsup;k++){
13370: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
13371: sumlpop=sumlpop+lpop[k];
13372: }
13373:
13374: tpop[agegomp]=sumlpop;
13375: for (k=agegomp;k<(agemortsup-3);k++){
13376: /* tpop[k+1]=2;*/
13377: tpop[k+1]=tpop[k]-lpop[k];
13378: }
13379:
13380:
13381: printf("\nAge lx qx dx Lx Tx e(x)\n");
13382: for (k=agegomp;k<(agemortsup-2);k++)
13383: 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]);
13384:
13385:
13386: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 13387: ageminpar=50;
13388: agemaxpar=100;
1.194 brouard 13389: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
13390: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13391: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13392: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
13393: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13394: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13395: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13396: }else{
13397: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
13398: 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 13399: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 13400: }
1.201 brouard 13401: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 13402: stepm, weightopt,\
13403: model,imx,p,matcov,agemortsup);
13404:
1.302 brouard 13405: free_vector(lsurv,agegomp,AGESUP);
13406: free_vector(lpop,agegomp,AGESUP);
13407: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 13408: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 13409: free_ivector(dcwave,firstobs,lastobs);
13410: free_vector(agecens,firstobs,lastobs);
13411: free_vector(ageexmed,firstobs,lastobs);
13412: free_ivector(cens,firstobs,lastobs);
1.220 brouard 13413: #ifdef GSL
1.136 brouard 13414: #endif
1.186 brouard 13415: } /* Endof if mle==-3 mortality only */
1.205 brouard 13416: /* Standard */
13417: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
13418: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13419: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 13420: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 13421: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
13422: for (k=1; k<=npar;k++)
13423: printf(" %d %8.5f",k,p[k]);
13424: printf("\n");
1.205 brouard 13425: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
13426: /* mlikeli uses func not funcone */
1.247 brouard 13427: /* for(i=1;i<nlstate;i++){ */
13428: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
13429: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
13430: /* } */
1.205 brouard 13431: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
13432: }
13433: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
13434: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13435: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
13436: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13437: }
13438: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 13439: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13440: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
1.335 brouard 13441: /* exit(0); */
1.126 brouard 13442: for (k=1; k<=npar;k++)
13443: printf(" %d %8.5f",k,p[k]);
13444: printf("\n");
13445:
13446: /*--------- results files --------------*/
1.283 brouard 13447: /* 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 13448:
13449:
13450: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13451: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 13452: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13453:
13454: printf("#model= 1 + age ");
13455: fprintf(ficres,"#model= 1 + age ");
13456: fprintf(ficlog,"#model= 1 + age ");
13457: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
13458: </ul>", model);
13459:
13460: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
13461: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
13462: if(nagesqr==1){
13463: printf(" + age*age ");
13464: fprintf(ficres," + age*age ");
13465: fprintf(ficlog," + age*age ");
13466: fprintf(fichtm, "<th>+ age*age</th>");
13467: }
13468: for(j=1;j <=ncovmodel-2;j++){
13469: if(Typevar[j]==0) {
13470: printf(" + V%d ",Tvar[j]);
13471: fprintf(ficres," + V%d ",Tvar[j]);
13472: fprintf(ficlog," + V%d ",Tvar[j]);
13473: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13474: }else if(Typevar[j]==1) {
13475: printf(" + V%d*age ",Tvar[j]);
13476: fprintf(ficres," + V%d*age ",Tvar[j]);
13477: fprintf(ficlog," + V%d*age ",Tvar[j]);
13478: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13479: }else if(Typevar[j]==2) {
13480: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13481: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13482: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13483: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13484: }
13485: }
13486: printf("\n");
13487: fprintf(ficres,"\n");
13488: fprintf(ficlog,"\n");
13489: fprintf(fichtm, "</tr>");
13490: fprintf(fichtm, "\n");
13491:
13492:
1.126 brouard 13493: for(i=1,jk=1; i <=nlstate; i++){
13494: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 13495: if (k != i) {
1.319 brouard 13496: fprintf(fichtm, "<tr>");
1.225 brouard 13497: printf("%d%d ",i,k);
13498: fprintf(ficlog,"%d%d ",i,k);
13499: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 13500: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13501: for(j=1; j <=ncovmodel; j++){
13502: printf("%12.7f ",p[jk]);
13503: fprintf(ficlog,"%12.7f ",p[jk]);
13504: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 13505: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 13506: jk++;
13507: }
13508: printf("\n");
13509: fprintf(ficlog,"\n");
13510: fprintf(ficres,"\n");
1.319 brouard 13511: fprintf(fichtm, "</tr>\n");
1.225 brouard 13512: }
1.126 brouard 13513: }
13514: }
1.319 brouard 13515: /* fprintf(fichtm,"</tr>\n"); */
13516: fprintf(fichtm,"</table>\n");
13517: fprintf(fichtm, "\n");
13518:
1.203 brouard 13519: if(mle != 0){
13520: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 13521: ftolhess=ftol; /* Usually correct */
1.203 brouard 13522: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
13523: 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");
13524: 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 13525: 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 13526: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
13527: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
13528: if(nagesqr==1){
13529: printf(" + age*age ");
13530: fprintf(ficres," + age*age ");
13531: fprintf(ficlog," + age*age ");
13532: fprintf(fichtm, "<th>+ age*age</th>");
13533: }
13534: for(j=1;j <=ncovmodel-2;j++){
13535: if(Typevar[j]==0) {
13536: printf(" + V%d ",Tvar[j]);
13537: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13538: }else if(Typevar[j]==1) {
13539: printf(" + V%d*age ",Tvar[j]);
13540: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13541: }else if(Typevar[j]==2) {
13542: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13543: }
13544: }
13545: fprintf(fichtm, "</tr>\n");
13546:
1.203 brouard 13547: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 13548: for(k=1; k <=(nlstate+ndeath); k++){
13549: if (k != i) {
1.319 brouard 13550: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 13551: printf("%d%d ",i,k);
13552: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 13553: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13554: for(j=1; j <=ncovmodel; j++){
1.319 brouard 13555: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 13556: 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]));
13557: 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 13558: if(fabs(wald) > 1.96){
1.321 brouard 13559: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 13560: }else{
13561: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
13562: }
1.324 brouard 13563: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 13564: 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 13565: jk++;
13566: }
13567: printf("\n");
13568: fprintf(ficlog,"\n");
1.319 brouard 13569: fprintf(fichtm, "</tr>\n");
1.225 brouard 13570: }
13571: }
1.193 brouard 13572: }
1.203 brouard 13573: } /* end of hesscov and Wald tests */
1.319 brouard 13574: fprintf(fichtm,"</table>\n");
1.225 brouard 13575:
1.203 brouard 13576: /* */
1.126 brouard 13577: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
13578: printf("# Scales (for hessian or gradient estimation)\n");
13579: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
13580: for(i=1,jk=1; i <=nlstate; i++){
13581: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 13582: if (j!=i) {
13583: fprintf(ficres,"%1d%1d",i,j);
13584: printf("%1d%1d",i,j);
13585: fprintf(ficlog,"%1d%1d",i,j);
13586: for(k=1; k<=ncovmodel;k++){
13587: printf(" %.5e",delti[jk]);
13588: fprintf(ficlog," %.5e",delti[jk]);
13589: fprintf(ficres," %.5e",delti[jk]);
13590: jk++;
13591: }
13592: printf("\n");
13593: fprintf(ficlog,"\n");
13594: fprintf(ficres,"\n");
13595: }
1.126 brouard 13596: }
13597: }
13598:
13599: 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 13600: if(mle >= 1) /* To big for the screen */
1.126 brouard 13601: 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");
13602: 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");
13603: /* # 121 Var(a12)\n\ */
13604: /* # 122 Cov(b12,a12) Var(b12)\n\ */
13605: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
13606: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
13607: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
13608: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
13609: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
13610: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
13611:
13612:
13613: /* Just to have a covariance matrix which will be more understandable
13614: even is we still don't want to manage dictionary of variables
13615: */
13616: for(itimes=1;itimes<=2;itimes++){
13617: jj=0;
13618: for(i=1; i <=nlstate; i++){
1.225 brouard 13619: for(j=1; j <=nlstate+ndeath; j++){
13620: if(j==i) continue;
13621: for(k=1; k<=ncovmodel;k++){
13622: jj++;
13623: ca[0]= k+'a'-1;ca[1]='\0';
13624: if(itimes==1){
13625: if(mle>=1)
13626: printf("#%1d%1d%d",i,j,k);
13627: fprintf(ficlog,"#%1d%1d%d",i,j,k);
13628: fprintf(ficres,"#%1d%1d%d",i,j,k);
13629: }else{
13630: if(mle>=1)
13631: printf("%1d%1d%d",i,j,k);
13632: fprintf(ficlog,"%1d%1d%d",i,j,k);
13633: fprintf(ficres,"%1d%1d%d",i,j,k);
13634: }
13635: ll=0;
13636: for(li=1;li <=nlstate; li++){
13637: for(lj=1;lj <=nlstate+ndeath; lj++){
13638: if(lj==li) continue;
13639: for(lk=1;lk<=ncovmodel;lk++){
13640: ll++;
13641: if(ll<=jj){
13642: cb[0]= lk +'a'-1;cb[1]='\0';
13643: if(ll<jj){
13644: if(itimes==1){
13645: if(mle>=1)
13646: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13647: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13648: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13649: }else{
13650: if(mle>=1)
13651: printf(" %.5e",matcov[jj][ll]);
13652: fprintf(ficlog," %.5e",matcov[jj][ll]);
13653: fprintf(ficres," %.5e",matcov[jj][ll]);
13654: }
13655: }else{
13656: if(itimes==1){
13657: if(mle>=1)
13658: printf(" Var(%s%1d%1d)",ca,i,j);
13659: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
13660: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
13661: }else{
13662: if(mle>=1)
13663: printf(" %.7e",matcov[jj][ll]);
13664: fprintf(ficlog," %.7e",matcov[jj][ll]);
13665: fprintf(ficres," %.7e",matcov[jj][ll]);
13666: }
13667: }
13668: }
13669: } /* end lk */
13670: } /* end lj */
13671: } /* end li */
13672: if(mle>=1)
13673: printf("\n");
13674: fprintf(ficlog,"\n");
13675: fprintf(ficres,"\n");
13676: numlinepar++;
13677: } /* end k*/
13678: } /*end j */
1.126 brouard 13679: } /* end i */
13680: } /* end itimes */
13681:
13682: fflush(ficlog);
13683: fflush(ficres);
1.225 brouard 13684: while(fgets(line, MAXLINE, ficpar)) {
13685: /* If line starts with a # it is a comment */
13686: if (line[0] == '#') {
13687: numlinepar++;
13688: fputs(line,stdout);
13689: fputs(line,ficparo);
13690: fputs(line,ficlog);
1.299 brouard 13691: fputs(line,ficres);
1.225 brouard 13692: continue;
13693: }else
13694: break;
13695: }
13696:
1.209 brouard 13697: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
13698: /* ungetc(c,ficpar); */
13699: /* fgets(line, MAXLINE, ficpar); */
13700: /* fputs(line,stdout); */
13701: /* fputs(line,ficparo); */
13702: /* } */
13703: /* ungetc(c,ficpar); */
1.126 brouard 13704:
13705: estepm=0;
1.209 brouard 13706: 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 13707:
13708: if (num_filled != 6) {
13709: 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);
13710: 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);
13711: goto end;
13712: }
13713: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
13714: }
13715: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
13716: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
13717:
1.209 brouard 13718: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 13719: if (estepm==0 || estepm < stepm) estepm=stepm;
13720: if (fage <= 2) {
13721: bage = ageminpar;
13722: fage = agemaxpar;
13723: }
13724:
13725: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 13726: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
13727: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 13728:
1.186 brouard 13729: /* Other stuffs, more or less useful */
1.254 brouard 13730: while(fgets(line, MAXLINE, ficpar)) {
13731: /* If line starts with a # it is a comment */
13732: if (line[0] == '#') {
13733: numlinepar++;
13734: fputs(line,stdout);
13735: fputs(line,ficparo);
13736: fputs(line,ficlog);
1.299 brouard 13737: fputs(line,ficres);
1.254 brouard 13738: continue;
13739: }else
13740: break;
13741: }
13742:
13743: 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){
13744:
13745: if (num_filled != 7) {
13746: 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);
13747: 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);
13748: goto end;
13749: }
13750: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13751: 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);
13752: 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);
13753: 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 13754: }
1.254 brouard 13755:
13756: while(fgets(line, MAXLINE, ficpar)) {
13757: /* If line starts with a # it is a comment */
13758: if (line[0] == '#') {
13759: numlinepar++;
13760: fputs(line,stdout);
13761: fputs(line,ficparo);
13762: fputs(line,ficlog);
1.299 brouard 13763: fputs(line,ficres);
1.254 brouard 13764: continue;
13765: }else
13766: break;
1.126 brouard 13767: }
13768:
13769:
13770: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
13771: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
13772:
1.254 brouard 13773: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
13774: if (num_filled != 1) {
13775: 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);
13776: 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);
13777: goto end;
13778: }
13779: printf("pop_based=%d\n",popbased);
13780: fprintf(ficlog,"pop_based=%d\n",popbased);
13781: fprintf(ficparo,"pop_based=%d\n",popbased);
13782: fprintf(ficres,"pop_based=%d\n",popbased);
13783: }
13784:
1.258 brouard 13785: /* Results */
1.332 brouard 13786: /* Value of covariate in each resultine will be compututed (if product) and sorted according to model rank */
13787: /* It is precov[] because we need the varying age in order to compute the real cov[] of the model equation */
13788: precov=matrix(1,MAXRESULTLINESPONE,1,NCOVMAX+1);
1.307 brouard 13789: endishere=0;
1.258 brouard 13790: nresult=0;
1.308 brouard 13791: parameterline=0;
1.258 brouard 13792: do{
13793: if(!fgets(line, MAXLINE, ficpar)){
13794: endishere=1;
1.308 brouard 13795: parameterline=15;
1.258 brouard 13796: }else if (line[0] == '#') {
13797: /* If line starts with a # it is a comment */
1.254 brouard 13798: numlinepar++;
13799: fputs(line,stdout);
13800: fputs(line,ficparo);
13801: fputs(line,ficlog);
1.299 brouard 13802: fputs(line,ficres);
1.254 brouard 13803: continue;
1.258 brouard 13804: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
13805: parameterline=11;
1.296 brouard 13806: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 13807: parameterline=12;
1.307 brouard 13808: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 13809: parameterline=13;
1.307 brouard 13810: }
1.258 brouard 13811: else{
13812: parameterline=14;
1.254 brouard 13813: }
1.308 brouard 13814: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 13815: case 11:
1.296 brouard 13816: 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)){
13817: 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 13818: 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);
13819: 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);
13820: 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);
13821: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 13822: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
13823: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 13824: prvforecast = 1;
13825: }
13826: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 13827: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13828: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13829: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 13830: prvforecast = 2;
13831: }
13832: else {
13833: 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);
13834: 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);
13835: goto end;
1.258 brouard 13836: }
1.254 brouard 13837: break;
1.258 brouard 13838: case 12:
1.296 brouard 13839: 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)){
13840: 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);
13841: 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);
13842: 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);
13843: 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);
13844: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 13845: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
13846: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 13847: prvbackcast = 1;
13848: }
13849: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 13850: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13851: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13852: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 13853: prvbackcast = 2;
13854: }
13855: else {
13856: 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);
13857: 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);
13858: goto end;
1.258 brouard 13859: }
1.230 brouard 13860: break;
1.258 brouard 13861: case 13:
1.332 brouard 13862: num_filled=sscanf(line,"result:%[^\n]\n",resultlineori);
1.307 brouard 13863: nresult++; /* Sum of resultlines */
1.342 ! brouard 13864: /* printf("Result %d: result:%s\n",nresult, resultlineori); */
1.332 brouard 13865: /* removefirstspace(&resultlineori); */
13866:
13867: if(strstr(resultlineori,"v") !=0){
13868: printf("Error. 'v' must be in upper case 'V' result: %s ",resultlineori);
13869: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultlineori);fflush(ficlog);
13870: return 1;
13871: }
13872: trimbb(resultline, resultlineori); /* Suppressing double blank in the resultline */
1.342 ! brouard 13873: /* printf("Decoderesult resultline=\"%s\" resultlineori=\"%s\"\n", resultline, resultlineori); */
1.318 brouard 13874: if(nresult > MAXRESULTLINESPONE-1){
13875: 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);
13876: 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 13877: goto end;
13878: }
1.332 brouard 13879:
1.310 brouard 13880: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 13881: fprintf(ficparo,"result: %s\n",resultline);
13882: fprintf(ficres,"result: %s\n",resultline);
13883: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 13884: } else
13885: goto end;
1.307 brouard 13886: break;
13887: case 14:
13888: printf("Error: Unknown command '%s'\n",line);
13889: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 13890: if(line[0] == ' ' || line[0] == '\n'){
13891: printf("It should not be an empty line '%s'\n",line);
13892: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
13893: }
1.307 brouard 13894: if(ncovmodel >=2 && nresult==0 ){
13895: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
13896: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 13897: }
1.307 brouard 13898: /* goto end; */
13899: break;
1.308 brouard 13900: case 15:
13901: printf("End of resultlines.\n");
13902: fprintf(ficlog,"End of resultlines.\n");
13903: break;
13904: default: /* parameterline =0 */
1.307 brouard 13905: nresult=1;
13906: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 13907: } /* End switch parameterline */
13908: }while(endishere==0); /* End do */
1.126 brouard 13909:
1.230 brouard 13910: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 13911: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 13912:
13913: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 13914: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 13915: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13916: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13917: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 13918: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13919: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13920: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13921: }else{
1.270 brouard 13922: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 13923: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
13924: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
13925: if(prvforecast==1){
13926: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
13927: jprojd=jproj1;
13928: mprojd=mproj1;
13929: anprojd=anproj1;
13930: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
13931: jprojf=jproj2;
13932: mprojf=mproj2;
13933: anprojf=anproj2;
13934: } else if(prvforecast == 2){
13935: dateprojd=dateintmean;
13936: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
13937: dateprojf=dateintmean+yrfproj;
13938: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
13939: }
13940: if(prvbackcast==1){
13941: datebackd=(jback1+12*mback1+365*anback1)/365;
13942: jbackd=jback1;
13943: mbackd=mback1;
13944: anbackd=anback1;
13945: datebackf=(jback2+12*mback2+365*anback2)/365;
13946: jbackf=jback2;
13947: mbackf=mback2;
13948: anbackf=anback2;
13949: } else if(prvbackcast == 2){
13950: datebackd=dateintmean;
13951: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
13952: datebackf=dateintmean-yrbproj;
13953: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
13954: }
13955:
13956: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 13957: }
13958: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 13959: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
13960: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 13961:
1.225 brouard 13962: /*------------ free_vector -------------*/
13963: /* chdir(path); */
1.220 brouard 13964:
1.215 brouard 13965: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
13966: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
13967: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
13968: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 13969: free_lvector(num,firstobs,lastobs);
13970: free_vector(agedc,firstobs,lastobs);
1.126 brouard 13971: /*free_matrix(covar,0,NCOVMAX,1,n);*/
13972: /*free_matrix(covar,1,NCOVMAX,1,n);*/
13973: fclose(ficparo);
13974: fclose(ficres);
1.220 brouard 13975:
13976:
1.186 brouard 13977: /* Other results (useful)*/
1.220 brouard 13978:
13979:
1.126 brouard 13980: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 13981: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
13982: prlim=matrix(1,nlstate,1,nlstate);
1.332 brouard 13983: /* Computes the prevalence limit for each combination k of the dummy covariates by calling prevalim(k) */
1.209 brouard 13984: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 13985: fclose(ficrespl);
13986:
13987: /*------------- h Pij x at various ages ------------*/
1.180 brouard 13988: /*#include "hpijx.h"*/
1.332 brouard 13989: /** 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?*/
13990: /* calls hpxij with combination k */
1.180 brouard 13991: hPijx(p, bage, fage);
1.145 brouard 13992: fclose(ficrespij);
1.227 brouard 13993:
1.220 brouard 13994: /* ncovcombmax= pow(2,cptcoveff); */
1.332 brouard 13995: /*-------------- Variance of one-step probabilities for a combination ij or for nres ?---*/
1.145 brouard 13996: k=1;
1.126 brouard 13997: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 13998:
1.269 brouard 13999: /* Prevalence for each covariate combination in probs[age][status][cov] */
14000: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
14001: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 14002: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 14003: for(k=1;k<=ncovcombmax;k++)
14004: probs[i][j][k]=0.;
1.269 brouard 14005: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
14006: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 14007: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 14008: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
14009: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 14010: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 14011: for(k=1;k<=ncovcombmax;k++)
14012: mobaverages[i][j][k]=0.;
1.219 brouard 14013: mobaverage=mobaverages;
14014: if (mobilav!=0) {
1.235 brouard 14015: printf("Movingaveraging observed prevalence\n");
1.258 brouard 14016: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 14017: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
14018: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
14019: printf(" Error in movingaverage mobilav=%d\n",mobilav);
14020: }
1.269 brouard 14021: } else if (mobilavproj !=0) {
1.235 brouard 14022: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 14023: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 14024: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
14025: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
14026: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
14027: }
1.269 brouard 14028: }else{
14029: printf("Internal error moving average\n");
14030: fflush(stdout);
14031: exit(1);
1.219 brouard 14032: }
14033: }/* end if moving average */
1.227 brouard 14034:
1.126 brouard 14035: /*---------- Forecasting ------------------*/
1.296 brouard 14036: if(prevfcast==1){
14037: /* /\* if(stepm ==1){*\/ */
14038: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
14039: /*This done previously after freqsummary.*/
14040: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
14041: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
14042:
14043: /* } else if (prvforecast==2){ */
14044: /* /\* if(stepm ==1){*\/ */
14045: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
14046: /* } */
14047: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
14048: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 14049: }
1.269 brouard 14050:
1.296 brouard 14051: /* Prevbcasting */
14052: if(prevbcast==1){
1.219 brouard 14053: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14054: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14055: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14056:
14057: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
14058:
14059: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 14060:
1.219 brouard 14061: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
14062: fclose(ficresplb);
14063:
1.222 brouard 14064: hBijx(p, bage, fage, mobaverage);
14065: fclose(ficrespijb);
1.219 brouard 14066:
1.296 brouard 14067: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
14068: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
14069: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
14070: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
14071: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
14072: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
14073:
14074:
1.269 brouard 14075: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14076:
14077:
1.269 brouard 14078: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 14079: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
14080: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
14081: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 14082: } /* end Prevbcasting */
1.268 brouard 14083:
1.186 brouard 14084:
14085: /* ------ Other prevalence ratios------------ */
1.126 brouard 14086:
1.215 brouard 14087: free_ivector(wav,1,imx);
14088: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
14089: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
14090: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 14091:
14092:
1.127 brouard 14093: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 14094:
1.201 brouard 14095: strcpy(filerese,"E_");
14096: strcat(filerese,fileresu);
1.126 brouard 14097: if((ficreseij=fopen(filerese,"w"))==NULL) {
14098: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
14099: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
14100: }
1.208 brouard 14101: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
14102: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 14103:
14104: pstamp(ficreseij);
1.219 brouard 14105:
1.235 brouard 14106: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
14107: if (cptcovn < 1){i1=1;}
14108:
14109: for(nres=1; nres <= nresult; nres++) /* For each resultline */
14110: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 14111: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 14112: continue;
1.219 brouard 14113: fprintf(ficreseij,"\n#****** ");
1.235 brouard 14114: printf("\n#****** ");
1.225 brouard 14115: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 14116: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
14117: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 14118: }
14119: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.337 brouard 14120: printf(" V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]); /* TvarsQ[j] gives the name of the jth quantitative (fixed or time v) */
14121: fprintf(ficreseij,"V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]);
1.219 brouard 14122: }
14123: fprintf(ficreseij,"******\n");
1.235 brouard 14124: printf("******\n");
1.219 brouard 14125:
14126: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14127: oldm=oldms;savm=savms;
1.330 brouard 14128: /* 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 14129: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 14130:
1.219 brouard 14131: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 14132: }
14133: fclose(ficreseij);
1.208 brouard 14134: printf("done evsij\n");fflush(stdout);
14135: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 14136:
1.218 brouard 14137:
1.227 brouard 14138: /*---------- State-specific expectancies and variances ------------*/
1.336 brouard 14139: /* Should be moved in a function */
1.201 brouard 14140: strcpy(filerest,"T_");
14141: strcat(filerest,fileresu);
1.127 brouard 14142: if((ficrest=fopen(filerest,"w"))==NULL) {
14143: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
14144: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
14145: }
1.208 brouard 14146: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
14147: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 14148: strcpy(fileresstde,"STDE_");
14149: strcat(fileresstde,fileresu);
1.126 brouard 14150: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 14151: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
14152: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 14153: }
1.227 brouard 14154: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
14155: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 14156:
1.201 brouard 14157: strcpy(filerescve,"CVE_");
14158: strcat(filerescve,fileresu);
1.126 brouard 14159: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 14160: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
14161: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 14162: }
1.227 brouard 14163: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
14164: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 14165:
1.201 brouard 14166: strcpy(fileresv,"V_");
14167: strcat(fileresv,fileresu);
1.126 brouard 14168: if((ficresvij=fopen(fileresv,"w"))==NULL) {
14169: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
14170: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
14171: }
1.227 brouard 14172: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
14173: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 14174:
1.235 brouard 14175: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
14176: if (cptcovn < 1){i1=1;}
14177:
1.334 brouard 14178: for(nres=1; nres <= nresult; nres++) /* For each resultline, find the combination and output results according to the values of dummies and then quanti. */
14179: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying. For each nres and each value at position k
14180: * we know Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline
14181: * Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline
14182: * and Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
14183: /* */
14184: 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 14185: continue;
1.321 brouard 14186: printf("\n# model %s \n#****** Result for:", model);
14187: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
14188: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.334 brouard 14189: /* It might not be a good idea to mix dummies and quantitative */
14190: /* 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 *\/ */
14191: 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 */
14192: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* Output by variables in the resultline *\/ */
14193: /* Tvaraff[j] is the name of the dummy variable in position j in the equation model:
14194: * Tvaraff[1]@9={4, 3, 0, 0, 0, 0, 0, 0, 0}, in model=V5+V4+V3+V4*V3+V5*age
14195: * (V5 is quanti) V4 and V3 are dummies
14196: * TnsdVar[4] is the position 1 and TnsdVar[3]=2 in codtabm(k,l)(V4 V3)=V4 V3
14197: * l=1 l=2
14198: * k=1 1 1 0 0
14199: * k=2 2 1 1 0
14200: * k=3 [1] [2] 0 1
14201: * k=4 2 2 1 1
14202: * If nres=1 result: V3=1 V4=0 then k=3 and outputs
14203: * If nres=2 result: V4=1 V3=0 then k=2 and outputs
14204: * nres=1 =>k=3 j=1 V4= nbcode[4][codtabm(3,1)=1)=0; j=2 V3= nbcode[3][codtabm(3,2)=2]=1
14205: * nres=2 =>k=2 j=1 V4= nbcode[4][codtabm(2,1)=2)=1; j=2 V3= nbcode[3][codtabm(2,2)=1]=0
14206: */
14207: /* Tvresult[nres][j] Name of the variable at position j in this resultline */
14208: /* Tresult[nres][j] Value of this variable at position j could be a float if quantitative */
14209: /* We give up with the combinations!! */
1.342 ! brouard 14210: /* if(debugILK) */
! 14211: /* 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 14212:
14213: 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 14214: 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 */
14215: 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 */
14216: 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 14217: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
14218: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
14219: }else{
14220: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
14221: }
14222: /* fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14223: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14224: }else if(Dummy[modelresult[nres][j]]==1){ /* Quanti variable */
14225: /* For each selected (single) quantitative value */
1.337 brouard 14226: printf(" V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
14227: fprintf(ficlog," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
14228: fprintf(ficrest," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
1.334 brouard 14229: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
14230: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
14231: }else{
14232: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
14233: }
14234: }else{
14235: 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 */
14236: 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 */
14237: exit(1);
14238: }
1.335 brouard 14239: } /* End loop for each variable in the resultline */
1.334 brouard 14240: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
14241: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /\* Wrong j is not in the equation model *\/ */
14242: /* fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14243: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14244: /* } */
1.208 brouard 14245: fprintf(ficrest,"******\n");
1.227 brouard 14246: fprintf(ficlog,"******\n");
14247: printf("******\n");
1.208 brouard 14248:
14249: fprintf(ficresstdeij,"\n#****** ");
14250: fprintf(ficrescveij,"\n#****** ");
1.337 brouard 14251: /* It could have been: for(j=1;j<=cptcoveff;j++) {printf("V=%d=%lg",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);} */
14252: /* But it won't be sorted and depends on how the resultline is ordered */
1.225 brouard 14253: for(j=1;j<=cptcoveff;j++) {
1.334 brouard 14254: fprintf(ficresstdeij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14255: /* fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14256: /* fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14257: }
14258: 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 14259: fprintf(ficresstdeij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
14260: fprintf(ficrescveij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
1.235 brouard 14261: }
1.208 brouard 14262: fprintf(ficresstdeij,"******\n");
14263: fprintf(ficrescveij,"******\n");
14264:
14265: fprintf(ficresvij,"\n#****** ");
1.238 brouard 14266: /* pstamp(ficresvij); */
1.225 brouard 14267: for(j=1;j<=cptcoveff;j++)
1.335 brouard 14268: fprintf(ficresvij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14269: /* fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[TnsdVar[Tvaraff[j]]])]); */
1.235 brouard 14270: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 14271: /* fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]); /\* To solve *\/ */
1.337 brouard 14272: fprintf(ficresvij," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /* Solved */
1.235 brouard 14273: }
1.208 brouard 14274: fprintf(ficresvij,"******\n");
14275:
14276: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14277: oldm=oldms;savm=savms;
1.235 brouard 14278: printf(" cvevsij ");
14279: fprintf(ficlog, " cvevsij ");
14280: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 14281: printf(" end cvevsij \n ");
14282: fprintf(ficlog, " end cvevsij \n ");
14283:
14284: /*
14285: */
14286: /* goto endfree; */
14287:
14288: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14289: pstamp(ficrest);
14290:
1.269 brouard 14291: epj=vector(1,nlstate+1);
1.208 brouard 14292: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 14293: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
14294: cptcod= 0; /* To be deleted */
14295: printf("varevsij vpopbased=%d \n",vpopbased);
14296: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 14297: 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 14298: 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 ");
14299: if(vpopbased==1)
14300: 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);
14301: else
1.288 brouard 14302: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.335 brouard 14303: fprintf(ficrest,"# Age popbased mobilav e.. (std) "); /* Adding covariate values? */
1.227 brouard 14304: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
14305: fprintf(ficrest,"\n");
14306: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 14307: printf("Computing age specific forward period (stable) prevalences in each health state \n");
14308: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 14309: for(age=bage; age <=fage ;age++){
1.235 brouard 14310: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 14311: if (vpopbased==1) {
14312: if(mobilav ==0){
14313: for(i=1; i<=nlstate;i++)
14314: prlim[i][i]=probs[(int)age][i][k];
14315: }else{ /* mobilav */
14316: for(i=1; i<=nlstate;i++)
14317: prlim[i][i]=mobaverage[(int)age][i][k];
14318: }
14319: }
1.219 brouard 14320:
1.227 brouard 14321: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
14322: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
14323: /* printf(" age %4.0f ",age); */
14324: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
14325: for(i=1, epj[j]=0.;i <=nlstate;i++) {
14326: epj[j] += prlim[i][i]*eij[i][j][(int)age];
14327: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
14328: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
14329: }
14330: epj[nlstate+1] +=epj[j];
14331: }
14332: /* printf(" age %4.0f \n",age); */
1.219 brouard 14333:
1.227 brouard 14334: for(i=1, vepp=0.;i <=nlstate;i++)
14335: for(j=1;j <=nlstate;j++)
14336: vepp += vareij[i][j][(int)age];
14337: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
14338: for(j=1;j <=nlstate;j++){
14339: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
14340: }
14341: fprintf(ficrest,"\n");
14342: }
1.208 brouard 14343: } /* End vpopbased */
1.269 brouard 14344: free_vector(epj,1,nlstate+1);
1.208 brouard 14345: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
14346: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 14347: printf("done selection\n");fflush(stdout);
14348: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 14349:
1.335 brouard 14350: } /* End k selection or end covariate selection for nres */
1.227 brouard 14351:
14352: printf("done State-specific expectancies\n");fflush(stdout);
14353: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
14354:
1.335 brouard 14355: /* variance-covariance of forward period prevalence */
1.269 brouard 14356: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14357:
1.227 brouard 14358:
1.290 brouard 14359: free_vector(weight,firstobs,lastobs);
1.330 brouard 14360: free_imatrix(Tvardk,1,NCOVMAX,1,2);
1.227 brouard 14361: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 14362: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
14363: free_matrix(anint,1,maxwav,firstobs,lastobs);
14364: free_matrix(mint,1,maxwav,firstobs,lastobs);
14365: free_ivector(cod,firstobs,lastobs);
1.227 brouard 14366: free_ivector(tab,1,NCOVMAX);
14367: fclose(ficresstdeij);
14368: fclose(ficrescveij);
14369: fclose(ficresvij);
14370: fclose(ficrest);
14371: fclose(ficpar);
14372:
14373:
1.126 brouard 14374: /*---------- End : free ----------------*/
1.219 brouard 14375: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 14376: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
14377: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 14378: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
14379: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 14380: } /* mle==-3 arrives here for freeing */
1.227 brouard 14381: /* endfree:*/
14382: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
14383: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
14384: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.341 brouard 14385: /* if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs); */
14386: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs);
1.290 brouard 14387: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
14388: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
14389: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 14390: free_matrix(matcov,1,npar,1,npar);
14391: free_matrix(hess,1,npar,1,npar);
14392: /*free_vector(delti,1,npar);*/
14393: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14394: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 14395: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 14396: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14397:
14398: free_ivector(ncodemax,1,NCOVMAX);
14399: free_ivector(ncodemaxwundef,1,NCOVMAX);
14400: free_ivector(Dummy,-1,NCOVMAX);
14401: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 14402: free_ivector(DummyV,1,NCOVMAX);
14403: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 14404: free_ivector(Typevar,-1,NCOVMAX);
14405: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 14406: free_ivector(TvarsQ,1,NCOVMAX);
14407: free_ivector(TvarsQind,1,NCOVMAX);
14408: free_ivector(TvarsD,1,NCOVMAX);
1.330 brouard 14409: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 14410: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 14411: free_ivector(TvarFD,1,NCOVMAX);
14412: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 14413: free_ivector(TvarF,1,NCOVMAX);
14414: free_ivector(TvarFind,1,NCOVMAX);
14415: free_ivector(TvarV,1,NCOVMAX);
14416: free_ivector(TvarVind,1,NCOVMAX);
14417: free_ivector(TvarA,1,NCOVMAX);
14418: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 14419: free_ivector(TvarFQ,1,NCOVMAX);
14420: free_ivector(TvarFQind,1,NCOVMAX);
14421: free_ivector(TvarVD,1,NCOVMAX);
14422: free_ivector(TvarVDind,1,NCOVMAX);
14423: free_ivector(TvarVQ,1,NCOVMAX);
14424: free_ivector(TvarVQind,1,NCOVMAX);
1.339 brouard 14425: free_ivector(TvarVV,1,NCOVMAX);
14426: free_ivector(TvarVVind,1,NCOVMAX);
14427:
1.230 brouard 14428: free_ivector(Tvarsel,1,NCOVMAX);
14429: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 14430: free_ivector(Tposprod,1,NCOVMAX);
14431: free_ivector(Tprod,1,NCOVMAX);
14432: free_ivector(Tvaraff,1,NCOVMAX);
1.338 brouard 14433: free_ivector(invalidvarcomb,0,ncovcombmax);
1.227 brouard 14434: free_ivector(Tage,1,NCOVMAX);
14435: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 14436: free_ivector(TmodelInvind,1,NCOVMAX);
14437: free_ivector(TmodelInvQind,1,NCOVMAX);
1.332 brouard 14438:
14439: free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /* Could be elsewhere ?*/
14440:
1.227 brouard 14441: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
14442: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 14443: fflush(fichtm);
14444: fflush(ficgp);
14445:
1.227 brouard 14446:
1.126 brouard 14447: if((nberr >0) || (nbwarn>0)){
1.216 brouard 14448: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
14449: 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 14450: }else{
14451: printf("End of Imach\n");
14452: fprintf(ficlog,"End of Imach\n");
14453: }
14454: printf("See log file on %s\n",filelog);
14455: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 14456: /*(void) gettimeofday(&end_time,&tzp);*/
14457: rend_time = time(NULL);
14458: end_time = *localtime(&rend_time);
14459: /* tml = *localtime(&end_time.tm_sec); */
14460: strcpy(strtend,asctime(&end_time));
1.126 brouard 14461: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
14462: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 14463: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 14464:
1.157 brouard 14465: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
14466: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
14467: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 14468: /* printf("Total time was %d uSec.\n", total_usecs);*/
14469: /* if(fileappend(fichtm,optionfilehtm)){ */
14470: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14471: fclose(fichtm);
14472: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14473: fclose(fichtmcov);
14474: fclose(ficgp);
14475: fclose(ficlog);
14476: /*------ End -----------*/
1.227 brouard 14477:
1.281 brouard 14478:
14479: /* Executes gnuplot */
1.227 brouard 14480:
14481: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 14482: #ifdef WIN32
1.227 brouard 14483: if (_chdir(pathcd) != 0)
14484: printf("Can't move to directory %s!\n",path);
14485: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 14486: #else
1.227 brouard 14487: if(chdir(pathcd) != 0)
14488: printf("Can't move to directory %s!\n", path);
14489: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 14490: #endif
1.126 brouard 14491: printf("Current directory %s!\n",pathcd);
14492: /*strcat(plotcmd,CHARSEPARATOR);*/
14493: sprintf(plotcmd,"gnuplot");
1.157 brouard 14494: #ifdef _WIN32
1.126 brouard 14495: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
14496: #endif
14497: if(!stat(plotcmd,&info)){
1.158 brouard 14498: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14499: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 14500: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 14501: }else
14502: strcpy(pplotcmd,plotcmd);
1.157 brouard 14503: #ifdef __unix
1.126 brouard 14504: strcpy(plotcmd,GNUPLOTPROGRAM);
14505: if(!stat(plotcmd,&info)){
1.158 brouard 14506: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14507: }else
14508: strcpy(pplotcmd,plotcmd);
14509: #endif
14510: }else
14511: strcpy(pplotcmd,plotcmd);
14512:
14513: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 14514: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 14515: strcpy(pplotcmd,plotcmd);
1.227 brouard 14516:
1.126 brouard 14517: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 14518: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 14519: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 14520: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 14521: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 14522: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 14523: strcpy(plotcmd,pplotcmd);
14524: }
1.126 brouard 14525: }
1.158 brouard 14526: printf(" Successful, please wait...");
1.126 brouard 14527: while (z[0] != 'q') {
14528: /* chdir(path); */
1.154 brouard 14529: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 14530: scanf("%s",z);
14531: /* if (z[0] == 'c') system("./imach"); */
14532: if (z[0] == 'e') {
1.158 brouard 14533: #ifdef __APPLE__
1.152 brouard 14534: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 14535: #elif __linux
14536: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 14537: #else
1.152 brouard 14538: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 14539: #endif
14540: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
14541: system(pplotcmd);
1.126 brouard 14542: }
14543: else if (z[0] == 'g') system(plotcmd);
14544: else if (z[0] == 'q') exit(0);
14545: }
1.227 brouard 14546: end:
1.126 brouard 14547: while (z[0] != 'q') {
1.195 brouard 14548: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 14549: scanf("%s",z);
14550: }
1.283 brouard 14551: printf("End\n");
1.282 brouard 14552: exit(0);
1.126 brouard 14553: }
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