Annotation of imach/src/imach.c, revision 1.322
1.322 ! brouard 1: /* $Id: imach.c,v 1.321 2022/07/22 12:04:24 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.322 ! brouard 4: Revision 1.321 2022/07/22 12:04:24 brouard
! 5: Summary: r28
! 6:
! 7: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
! 8:
1.321 brouard 9: Revision 1.320 2022/06/02 05:10:11 brouard
10: *** empty log message ***
11:
1.320 brouard 12: Revision 1.319 2022/06/02 04:45:11 brouard
13: * imach.c (Module): Adding the Wald tests from the log to the main
14: htm for better display of the maximum likelihood estimators.
15:
1.319 brouard 16: Revision 1.318 2022/05/24 08:10:59 brouard
17: * imach.c (Module): Some attempts to find a bug of wrong estimates
18: of confidencce intervals with product in the equation modelC
19:
1.318 brouard 20: Revision 1.317 2022/05/15 15:06:23 brouard
21: * imach.c (Module): Some minor improvements
22:
1.317 brouard 23: Revision 1.316 2022/05/11 15:11:31 brouard
24: Summary: r27
25:
1.316 brouard 26: Revision 1.315 2022/05/11 15:06:32 brouard
27: *** empty log message ***
28:
1.315 brouard 29: Revision 1.314 2022/04/13 17:43:09 brouard
30: * imach.c (Module): Adding link to text data files
31:
1.314 brouard 32: Revision 1.313 2022/04/11 15:57:42 brouard
33: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
34:
1.313 brouard 35: Revision 1.312 2022/04/05 21:24:39 brouard
36: *** empty log message ***
37:
1.312 brouard 38: Revision 1.311 2022/04/05 21:03:51 brouard
39: Summary: Fixed quantitative covariates
40:
41: Fixed covariates (dummy or quantitative)
42: with missing values have never been allowed but are ERRORS and
43: program quits. Standard deviations of fixed covariates were
44: wrongly computed. Mean and standard deviations of time varying
45: covariates are still not computed.
46:
1.311 brouard 47: Revision 1.310 2022/03/17 08:45:53 brouard
48: Summary: 99r25
49:
50: Improving detection of errors: result lines should be compatible with
51: the model.
52:
1.310 brouard 53: Revision 1.309 2021/05/20 12:39:14 brouard
54: Summary: Version 0.99r24
55:
1.309 brouard 56: Revision 1.308 2021/03/31 13:11:57 brouard
57: Summary: Version 0.99r23
58:
59:
60: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
61:
1.308 brouard 62: Revision 1.307 2021/03/08 18:11:32 brouard
63: Summary: 0.99r22 fixed bug on result:
64:
1.307 brouard 65: Revision 1.306 2021/02/20 15:44:02 brouard
66: Summary: Version 0.99r21
67:
68: * imach.c (Module): Fix bug on quitting after result lines!
69: (Module): Version 0.99r21
70:
1.306 brouard 71: Revision 1.305 2021/02/20 15:28:30 brouard
72: * imach.c (Module): Fix bug on quitting after result lines!
73:
1.305 brouard 74: Revision 1.304 2021/02/12 11:34:20 brouard
75: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
76:
1.304 brouard 77: Revision 1.303 2021/02/11 19:50:15 brouard
78: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
79:
1.303 brouard 80: Revision 1.302 2020/02/22 21:00:05 brouard
81: * (Module): imach.c Update mle=-3 (for computing Life expectancy
82: and life table from the data without any state)
83:
1.302 brouard 84: Revision 1.301 2019/06/04 13:51:20 brouard
85: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
86:
1.301 brouard 87: Revision 1.300 2019/05/22 19:09:45 brouard
88: Summary: version 0.99r19 of May 2019
89:
1.300 brouard 90: Revision 1.299 2019/05/22 18:37:08 brouard
91: Summary: Cleaned 0.99r19
92:
1.299 brouard 93: Revision 1.298 2019/05/22 18:19:56 brouard
94: *** empty log message ***
95:
1.298 brouard 96: Revision 1.297 2019/05/22 17:56:10 brouard
97: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
98:
1.297 brouard 99: Revision 1.296 2019/05/20 13:03:18 brouard
100: Summary: Projection syntax simplified
101:
102:
103: We can now start projections, forward or backward, from the mean date
104: of inteviews up to or down to a number of years of projection:
105: prevforecast=1 yearsfproj=15.3 mobil_average=0
106: or
107: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
108: or
109: prevbackcast=1 yearsbproj=12.3 mobil_average=1
110: or
111: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
112:
1.296 brouard 113: Revision 1.295 2019/05/18 09:52:50 brouard
114: Summary: doxygen tex bug
115:
1.295 brouard 116: Revision 1.294 2019/05/16 14:54:33 brouard
117: Summary: There was some wrong lines added
118:
1.294 brouard 119: Revision 1.293 2019/05/09 15:17:34 brouard
120: *** empty log message ***
121:
1.293 brouard 122: Revision 1.292 2019/05/09 14:17:20 brouard
123: Summary: Some updates
124:
1.292 brouard 125: Revision 1.291 2019/05/09 13:44:18 brouard
126: Summary: Before ncovmax
127:
1.291 brouard 128: Revision 1.290 2019/05/09 13:39:37 brouard
129: Summary: 0.99r18 unlimited number of individuals
130:
131: 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.
132:
1.290 brouard 133: Revision 1.289 2018/12/13 09:16:26 brouard
134: Summary: Bug for young ages (<-30) will be in r17
135:
1.289 brouard 136: Revision 1.288 2018/05/02 20:58:27 brouard
137: Summary: Some bugs fixed
138:
1.288 brouard 139: Revision 1.287 2018/05/01 17:57:25 brouard
140: Summary: Bug fixed by providing frequencies only for non missing covariates
141:
1.287 brouard 142: Revision 1.286 2018/04/27 14:27:04 brouard
143: Summary: some minor bugs
144:
1.286 brouard 145: Revision 1.285 2018/04/21 21:02:16 brouard
146: Summary: Some bugs fixed, valgrind tested
147:
1.285 brouard 148: Revision 1.284 2018/04/20 05:22:13 brouard
149: Summary: Computing mean and stdeviation of fixed quantitative variables
150:
1.284 brouard 151: Revision 1.283 2018/04/19 14:49:16 brouard
152: Summary: Some minor bugs fixed
153:
1.283 brouard 154: Revision 1.282 2018/02/27 22:50:02 brouard
155: *** empty log message ***
156:
1.282 brouard 157: Revision 1.281 2018/02/27 19:25:23 brouard
158: Summary: Adding second argument for quitting
159:
1.281 brouard 160: Revision 1.280 2018/02/21 07:58:13 brouard
161: Summary: 0.99r15
162:
163: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
164:
1.280 brouard 165: Revision 1.279 2017/07/20 13:35:01 brouard
166: Summary: temporary working
167:
1.279 brouard 168: Revision 1.278 2017/07/19 14:09:02 brouard
169: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
170:
1.278 brouard 171: Revision 1.277 2017/07/17 08:53:49 brouard
172: Summary: BOM files can be read now
173:
1.277 brouard 174: Revision 1.276 2017/06/30 15:48:31 brouard
175: Summary: Graphs improvements
176:
1.276 brouard 177: Revision 1.275 2017/06/30 13:39:33 brouard
178: Summary: Saito's color
179:
1.275 brouard 180: Revision 1.274 2017/06/29 09:47:08 brouard
181: Summary: Version 0.99r14
182:
1.274 brouard 183: Revision 1.273 2017/06/27 11:06:02 brouard
184: Summary: More documentation on projections
185:
1.273 brouard 186: Revision 1.272 2017/06/27 10:22:40 brouard
187: Summary: Color of backprojection changed from 6 to 5(yellow)
188:
1.272 brouard 189: Revision 1.271 2017/06/27 10:17:50 brouard
190: Summary: Some bug with rint
191:
1.271 brouard 192: Revision 1.270 2017/05/24 05:45:29 brouard
193: *** empty log message ***
194:
1.270 brouard 195: Revision 1.269 2017/05/23 08:39:25 brouard
196: Summary: Code into subroutine, cleanings
197:
1.269 brouard 198: Revision 1.268 2017/05/18 20:09:32 brouard
199: Summary: backprojection and confidence intervals of backprevalence
200:
1.268 brouard 201: Revision 1.267 2017/05/13 10:25:05 brouard
202: Summary: temporary save for backprojection
203:
1.267 brouard 204: Revision 1.266 2017/05/13 07:26:12 brouard
205: Summary: Version 0.99r13 (improvements and bugs fixed)
206:
1.266 brouard 207: Revision 1.265 2017/04/26 16:22:11 brouard
208: Summary: imach 0.99r13 Some bugs fixed
209:
1.265 brouard 210: Revision 1.264 2017/04/26 06:01:29 brouard
211: Summary: Labels in graphs
212:
1.264 brouard 213: Revision 1.263 2017/04/24 15:23:15 brouard
214: Summary: to save
215:
1.263 brouard 216: Revision 1.262 2017/04/18 16:48:12 brouard
217: *** empty log message ***
218:
1.262 brouard 219: Revision 1.261 2017/04/05 10:14:09 brouard
220: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
221:
1.261 brouard 222: Revision 1.260 2017/04/04 17:46:59 brouard
223: Summary: Gnuplot indexations fixed (humm)
224:
1.260 brouard 225: Revision 1.259 2017/04/04 13:01:16 brouard
226: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
227:
1.259 brouard 228: Revision 1.258 2017/04/03 10:17:47 brouard
229: Summary: Version 0.99r12
230:
231: Some cleanings, conformed with updated documentation.
232:
1.258 brouard 233: Revision 1.257 2017/03/29 16:53:30 brouard
234: Summary: Temp
235:
1.257 brouard 236: Revision 1.256 2017/03/27 05:50:23 brouard
237: Summary: Temporary
238:
1.256 brouard 239: Revision 1.255 2017/03/08 16:02:28 brouard
240: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
241:
1.255 brouard 242: Revision 1.254 2017/03/08 07:13:00 brouard
243: Summary: Fixing data parameter line
244:
1.254 brouard 245: Revision 1.253 2016/12/15 11:59:41 brouard
246: Summary: 0.99 in progress
247:
1.253 brouard 248: Revision 1.252 2016/09/15 21:15:37 brouard
249: *** empty log message ***
250:
1.252 brouard 251: Revision 1.251 2016/09/15 15:01:13 brouard
252: Summary: not working
253:
1.251 brouard 254: Revision 1.250 2016/09/08 16:07:27 brouard
255: Summary: continue
256:
1.250 brouard 257: Revision 1.249 2016/09/07 17:14:18 brouard
258: Summary: Starting values from frequencies
259:
1.249 brouard 260: Revision 1.248 2016/09/07 14:10:18 brouard
261: *** empty log message ***
262:
1.248 brouard 263: Revision 1.247 2016/09/02 11:11:21 brouard
264: *** empty log message ***
265:
1.247 brouard 266: Revision 1.246 2016/09/02 08:49:22 brouard
267: *** empty log message ***
268:
1.246 brouard 269: Revision 1.245 2016/09/02 07:25:01 brouard
270: *** empty log message ***
271:
1.245 brouard 272: Revision 1.244 2016/09/02 07:17:34 brouard
273: *** empty log message ***
274:
1.244 brouard 275: Revision 1.243 2016/09/02 06:45:35 brouard
276: *** empty log message ***
277:
1.243 brouard 278: Revision 1.242 2016/08/30 15:01:20 brouard
279: Summary: Fixing a lots
280:
1.242 brouard 281: Revision 1.241 2016/08/29 17:17:25 brouard
282: Summary: gnuplot problem in Back projection to fix
283:
1.241 brouard 284: Revision 1.240 2016/08/29 07:53:18 brouard
285: Summary: Better
286:
1.240 brouard 287: Revision 1.239 2016/08/26 15:51:03 brouard
288: Summary: Improvement in Powell output in order to copy and paste
289:
290: Author:
291:
1.239 brouard 292: Revision 1.238 2016/08/26 14:23:35 brouard
293: Summary: Starting tests of 0.99
294:
1.238 brouard 295: Revision 1.237 2016/08/26 09:20:19 brouard
296: Summary: to valgrind
297:
1.237 brouard 298: Revision 1.236 2016/08/25 10:50:18 brouard
299: *** empty log message ***
300:
1.236 brouard 301: Revision 1.235 2016/08/25 06:59:23 brouard
302: *** empty log message ***
303:
1.235 brouard 304: Revision 1.234 2016/08/23 16:51:20 brouard
305: *** empty log message ***
306:
1.234 brouard 307: Revision 1.233 2016/08/23 07:40:50 brouard
308: Summary: not working
309:
1.233 brouard 310: Revision 1.232 2016/08/22 14:20:21 brouard
311: Summary: not working
312:
1.232 brouard 313: Revision 1.231 2016/08/22 07:17:15 brouard
314: Summary: not working
315:
1.231 brouard 316: Revision 1.230 2016/08/22 06:55:53 brouard
317: Summary: Not working
318:
1.230 brouard 319: Revision 1.229 2016/07/23 09:45:53 brouard
320: Summary: Completing for func too
321:
1.229 brouard 322: Revision 1.228 2016/07/22 17:45:30 brouard
323: Summary: Fixing some arrays, still debugging
324:
1.227 brouard 325: Revision 1.226 2016/07/12 18:42:34 brouard
326: Summary: temp
327:
1.226 brouard 328: Revision 1.225 2016/07/12 08:40:03 brouard
329: Summary: saving but not running
330:
1.225 brouard 331: Revision 1.224 2016/07/01 13:16:01 brouard
332: Summary: Fixes
333:
1.224 brouard 334: Revision 1.223 2016/02/19 09:23:35 brouard
335: Summary: temporary
336:
1.223 brouard 337: Revision 1.222 2016/02/17 08:14:50 brouard
338: Summary: Probably last 0.98 stable version 0.98r6
339:
1.222 brouard 340: Revision 1.221 2016/02/15 23:35:36 brouard
341: Summary: minor bug
342:
1.220 brouard 343: Revision 1.219 2016/02/15 00:48:12 brouard
344: *** empty log message ***
345:
1.219 brouard 346: Revision 1.218 2016/02/12 11:29:23 brouard
347: Summary: 0.99 Back projections
348:
1.218 brouard 349: Revision 1.217 2015/12/23 17:18:31 brouard
350: Summary: Experimental backcast
351:
1.217 brouard 352: Revision 1.216 2015/12/18 17:32:11 brouard
353: Summary: 0.98r4 Warning and status=-2
354:
355: Version 0.98r4 is now:
356: - displaying an error when status is -1, date of interview unknown and date of death known;
357: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
358: Older changes concerning s=-2, dating from 2005 have been supersed.
359:
1.216 brouard 360: Revision 1.215 2015/12/16 08:52:24 brouard
361: Summary: 0.98r4 working
362:
1.215 brouard 363: Revision 1.214 2015/12/16 06:57:54 brouard
364: Summary: temporary not working
365:
1.214 brouard 366: Revision 1.213 2015/12/11 18:22:17 brouard
367: Summary: 0.98r4
368:
1.213 brouard 369: Revision 1.212 2015/11/21 12:47:24 brouard
370: Summary: minor typo
371:
1.212 brouard 372: Revision 1.211 2015/11/21 12:41:11 brouard
373: Summary: 0.98r3 with some graph of projected cross-sectional
374:
375: Author: Nicolas Brouard
376:
1.211 brouard 377: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 378: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 379: Summary: Adding ftolpl parameter
380: Author: N Brouard
381:
382: We had difficulties to get smoothed confidence intervals. It was due
383: to the period prevalence which wasn't computed accurately. The inner
384: parameter ftolpl is now an outer parameter of the .imach parameter
385: file after estepm. If ftolpl is small 1.e-4 and estepm too,
386: computation are long.
387:
1.209 brouard 388: Revision 1.208 2015/11/17 14:31:57 brouard
389: Summary: temporary
390:
1.208 brouard 391: Revision 1.207 2015/10/27 17:36:57 brouard
392: *** empty log message ***
393:
1.207 brouard 394: Revision 1.206 2015/10/24 07:14:11 brouard
395: *** empty log message ***
396:
1.206 brouard 397: Revision 1.205 2015/10/23 15:50:53 brouard
398: Summary: 0.98r3 some clarification for graphs on likelihood contributions
399:
1.205 brouard 400: Revision 1.204 2015/10/01 16:20:26 brouard
401: Summary: Some new graphs of contribution to likelihood
402:
1.204 brouard 403: Revision 1.203 2015/09/30 17:45:14 brouard
404: Summary: looking at better estimation of the hessian
405:
406: Also a better criteria for convergence to the period prevalence And
407: therefore adding the number of years needed to converge. (The
408: prevalence in any alive state shold sum to one
409:
1.203 brouard 410: Revision 1.202 2015/09/22 19:45:16 brouard
411: Summary: Adding some overall graph on contribution to likelihood. Might change
412:
1.202 brouard 413: Revision 1.201 2015/09/15 17:34:58 brouard
414: Summary: 0.98r0
415:
416: - Some new graphs like suvival functions
417: - Some bugs fixed like model=1+age+V2.
418:
1.201 brouard 419: Revision 1.200 2015/09/09 16:53:55 brouard
420: Summary: Big bug thanks to Flavia
421:
422: Even model=1+age+V2. did not work anymore
423:
1.200 brouard 424: Revision 1.199 2015/09/07 14:09:23 brouard
425: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
426:
1.199 brouard 427: Revision 1.198 2015/09/03 07:14:39 brouard
428: Summary: 0.98q5 Flavia
429:
1.198 brouard 430: Revision 1.197 2015/09/01 18:24:39 brouard
431: *** empty log message ***
432:
1.197 brouard 433: Revision 1.196 2015/08/18 23:17:52 brouard
434: Summary: 0.98q5
435:
1.196 brouard 436: Revision 1.195 2015/08/18 16:28:39 brouard
437: Summary: Adding a hack for testing purpose
438:
439: After reading the title, ftol and model lines, if the comment line has
440: a q, starting with #q, the answer at the end of the run is quit. It
441: permits to run test files in batch with ctest. The former workaround was
442: $ echo q | imach foo.imach
443:
1.195 brouard 444: Revision 1.194 2015/08/18 13:32:00 brouard
445: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
446:
1.194 brouard 447: Revision 1.193 2015/08/04 07:17:42 brouard
448: Summary: 0.98q4
449:
1.193 brouard 450: Revision 1.192 2015/07/16 16:49:02 brouard
451: Summary: Fixing some outputs
452:
1.192 brouard 453: Revision 1.191 2015/07/14 10:00:33 brouard
454: Summary: Some fixes
455:
1.191 brouard 456: Revision 1.190 2015/05/05 08:51:13 brouard
457: Summary: Adding digits in output parameters (7 digits instead of 6)
458:
459: Fix 1+age+.
460:
1.190 brouard 461: Revision 1.189 2015/04/30 14:45:16 brouard
462: Summary: 0.98q2
463:
1.189 brouard 464: Revision 1.188 2015/04/30 08:27:53 brouard
465: *** empty log message ***
466:
1.188 brouard 467: Revision 1.187 2015/04/29 09:11:15 brouard
468: *** empty log message ***
469:
1.187 brouard 470: Revision 1.186 2015/04/23 12:01:52 brouard
471: Summary: V1*age is working now, version 0.98q1
472:
473: Some codes had been disabled in order to simplify and Vn*age was
474: working in the optimization phase, ie, giving correct MLE parameters,
475: but, as usual, outputs were not correct and program core dumped.
476:
1.186 brouard 477: Revision 1.185 2015/03/11 13:26:42 brouard
478: Summary: Inclusion of compile and links command line for Intel Compiler
479:
1.185 brouard 480: Revision 1.184 2015/03/11 11:52:39 brouard
481: Summary: Back from Windows 8. Intel Compiler
482:
1.184 brouard 483: Revision 1.183 2015/03/10 20:34:32 brouard
484: Summary: 0.98q0, trying with directest, mnbrak fixed
485:
486: We use directest instead of original Powell test; probably no
487: incidence on the results, but better justifications;
488: We fixed Numerical Recipes mnbrak routine which was wrong and gave
489: wrong results.
490:
1.183 brouard 491: Revision 1.182 2015/02/12 08:19:57 brouard
492: Summary: Trying to keep directest which seems simpler and more general
493: Author: Nicolas Brouard
494:
1.182 brouard 495: Revision 1.181 2015/02/11 23:22:24 brouard
496: Summary: Comments on Powell added
497:
498: Author:
499:
1.181 brouard 500: Revision 1.180 2015/02/11 17:33:45 brouard
501: Summary: Finishing move from main to function (hpijx and prevalence_limit)
502:
1.180 brouard 503: Revision 1.179 2015/01/04 09:57:06 brouard
504: Summary: back to OS/X
505:
1.179 brouard 506: Revision 1.178 2015/01/04 09:35:48 brouard
507: *** empty log message ***
508:
1.178 brouard 509: Revision 1.177 2015/01/03 18:40:56 brouard
510: Summary: Still testing ilc32 on OSX
511:
1.177 brouard 512: Revision 1.176 2015/01/03 16:45:04 brouard
513: *** empty log message ***
514:
1.176 brouard 515: Revision 1.175 2015/01/03 16:33:42 brouard
516: *** empty log message ***
517:
1.175 brouard 518: Revision 1.174 2015/01/03 16:15:49 brouard
519: Summary: Still in cross-compilation
520:
1.174 brouard 521: Revision 1.173 2015/01/03 12:06:26 brouard
522: Summary: trying to detect cross-compilation
523:
1.173 brouard 524: Revision 1.172 2014/12/27 12:07:47 brouard
525: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
526:
1.172 brouard 527: Revision 1.171 2014/12/23 13:26:59 brouard
528: Summary: Back from Visual C
529:
530: Still problem with utsname.h on Windows
531:
1.171 brouard 532: Revision 1.170 2014/12/23 11:17:12 brouard
533: Summary: Cleaning some \%% back to %%
534:
535: The escape was mandatory for a specific compiler (which one?), but too many warnings.
536:
1.170 brouard 537: Revision 1.169 2014/12/22 23:08:31 brouard
538: Summary: 0.98p
539:
540: Outputs some informations on compiler used, OS etc. Testing on different platforms.
541:
1.169 brouard 542: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 543: Summary: update
1.169 brouard 544:
1.168 brouard 545: Revision 1.167 2014/12/22 13:50:56 brouard
546: Summary: Testing uname and compiler version and if compiled 32 or 64
547:
548: Testing on Linux 64
549:
1.167 brouard 550: Revision 1.166 2014/12/22 11:40:47 brouard
551: *** empty log message ***
552:
1.166 brouard 553: Revision 1.165 2014/12/16 11:20:36 brouard
554: Summary: After compiling on Visual C
555:
556: * imach.c (Module): Merging 1.61 to 1.162
557:
1.165 brouard 558: Revision 1.164 2014/12/16 10:52:11 brouard
559: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
560:
561: * imach.c (Module): Merging 1.61 to 1.162
562:
1.164 brouard 563: Revision 1.163 2014/12/16 10:30:11 brouard
564: * imach.c (Module): Merging 1.61 to 1.162
565:
1.163 brouard 566: Revision 1.162 2014/09/25 11:43:39 brouard
567: Summary: temporary backup 0.99!
568:
1.162 brouard 569: Revision 1.1 2014/09/16 11:06:58 brouard
570: Summary: With some code (wrong) for nlopt
571:
572: Author:
573:
574: Revision 1.161 2014/09/15 20:41:41 brouard
575: Summary: Problem with macro SQR on Intel compiler
576:
1.161 brouard 577: Revision 1.160 2014/09/02 09:24:05 brouard
578: *** empty log message ***
579:
1.160 brouard 580: Revision 1.159 2014/09/01 10:34:10 brouard
581: Summary: WIN32
582: Author: Brouard
583:
1.159 brouard 584: Revision 1.158 2014/08/27 17:11:51 brouard
585: *** empty log message ***
586:
1.158 brouard 587: Revision 1.157 2014/08/27 16:26:55 brouard
588: Summary: Preparing windows Visual studio version
589: Author: Brouard
590:
591: In order to compile on Visual studio, time.h is now correct and time_t
592: and tm struct should be used. difftime should be used but sometimes I
593: just make the differences in raw time format (time(&now).
594: Trying to suppress #ifdef LINUX
595: Add xdg-open for __linux in order to open default browser.
596:
1.157 brouard 597: Revision 1.156 2014/08/25 20:10:10 brouard
598: *** empty log message ***
599:
1.156 brouard 600: Revision 1.155 2014/08/25 18:32:34 brouard
601: Summary: New compile, minor changes
602: Author: Brouard
603:
1.155 brouard 604: Revision 1.154 2014/06/20 17:32:08 brouard
605: Summary: Outputs now all graphs of convergence to period prevalence
606:
1.154 brouard 607: Revision 1.153 2014/06/20 16:45:46 brouard
608: Summary: If 3 live state, convergence to period prevalence on same graph
609: Author: Brouard
610:
1.153 brouard 611: Revision 1.152 2014/06/18 17:54:09 brouard
612: Summary: open browser, use gnuplot on same dir than imach if not found in the path
613:
1.152 brouard 614: Revision 1.151 2014/06/18 16:43:30 brouard
615: *** empty log message ***
616:
1.151 brouard 617: Revision 1.150 2014/06/18 16:42:35 brouard
618: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
619: Author: brouard
620:
1.150 brouard 621: Revision 1.149 2014/06/18 15:51:14 brouard
622: Summary: Some fixes in parameter files errors
623: Author: Nicolas Brouard
624:
1.149 brouard 625: Revision 1.148 2014/06/17 17:38:48 brouard
626: Summary: Nothing new
627: Author: Brouard
628:
629: Just a new packaging for OS/X version 0.98nS
630:
1.148 brouard 631: Revision 1.147 2014/06/16 10:33:11 brouard
632: *** empty log message ***
633:
1.147 brouard 634: Revision 1.146 2014/06/16 10:20:28 brouard
635: Summary: Merge
636: Author: Brouard
637:
638: Merge, before building revised version.
639:
1.146 brouard 640: Revision 1.145 2014/06/10 21:23:15 brouard
641: Summary: Debugging with valgrind
642: Author: Nicolas Brouard
643:
644: Lot of changes in order to output the results with some covariates
645: After the Edimburgh REVES conference 2014, it seems mandatory to
646: improve the code.
647: No more memory valgrind error but a lot has to be done in order to
648: continue the work of splitting the code into subroutines.
649: Also, decodemodel has been improved. Tricode is still not
650: optimal. nbcode should be improved. Documentation has been added in
651: the source code.
652:
1.144 brouard 653: Revision 1.143 2014/01/26 09:45:38 brouard
654: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
655:
656: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
657: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
658:
1.143 brouard 659: Revision 1.142 2014/01/26 03:57:36 brouard
660: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
661:
662: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
663:
1.142 brouard 664: Revision 1.141 2014/01/26 02:42:01 brouard
665: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
666:
1.141 brouard 667: Revision 1.140 2011/09/02 10:37:54 brouard
668: Summary: times.h is ok with mingw32 now.
669:
1.140 brouard 670: Revision 1.139 2010/06/14 07:50:17 brouard
671: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
672: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
673:
1.139 brouard 674: Revision 1.138 2010/04/30 18:19:40 brouard
675: *** empty log message ***
676:
1.138 brouard 677: Revision 1.137 2010/04/29 18:11:38 brouard
678: (Module): Checking covariates for more complex models
679: than V1+V2. A lot of change to be done. Unstable.
680:
1.137 brouard 681: Revision 1.136 2010/04/26 20:30:53 brouard
682: (Module): merging some libgsl code. Fixing computation
683: of likelione (using inter/intrapolation if mle = 0) in order to
684: get same likelihood as if mle=1.
685: Some cleaning of code and comments added.
686:
1.136 brouard 687: Revision 1.135 2009/10/29 15:33:14 brouard
688: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
689:
1.135 brouard 690: Revision 1.134 2009/10/29 13:18:53 brouard
691: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
692:
1.134 brouard 693: Revision 1.133 2009/07/06 10:21:25 brouard
694: just nforces
695:
1.133 brouard 696: Revision 1.132 2009/07/06 08:22:05 brouard
697: Many tings
698:
1.132 brouard 699: Revision 1.131 2009/06/20 16:22:47 brouard
700: Some dimensions resccaled
701:
1.131 brouard 702: Revision 1.130 2009/05/26 06:44:34 brouard
703: (Module): Max Covariate is now set to 20 instead of 8. A
704: lot of cleaning with variables initialized to 0. Trying to make
705: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
706:
1.130 brouard 707: Revision 1.129 2007/08/31 13:49:27 lievre
708: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
709:
1.129 lievre 710: Revision 1.128 2006/06/30 13:02:05 brouard
711: (Module): Clarifications on computing e.j
712:
1.128 brouard 713: Revision 1.127 2006/04/28 18:11:50 brouard
714: (Module): Yes the sum of survivors was wrong since
715: imach-114 because nhstepm was no more computed in the age
716: loop. Now we define nhstepma in the age loop.
717: (Module): In order to speed up (in case of numerous covariates) we
718: compute health expectancies (without variances) in a first step
719: and then all the health expectancies with variances or standard
720: deviation (needs data from the Hessian matrices) which slows the
721: computation.
722: In the future we should be able to stop the program is only health
723: expectancies and graph are needed without standard deviations.
724:
1.127 brouard 725: Revision 1.126 2006/04/28 17:23:28 brouard
726: (Module): Yes the sum of survivors was wrong since
727: imach-114 because nhstepm was no more computed in the age
728: loop. Now we define nhstepma in the age loop.
729: Version 0.98h
730:
1.126 brouard 731: Revision 1.125 2006/04/04 15:20:31 lievre
732: Errors in calculation of health expectancies. Age was not initialized.
733: Forecasting file added.
734:
735: Revision 1.124 2006/03/22 17:13:53 lievre
736: Parameters are printed with %lf instead of %f (more numbers after the comma).
737: The log-likelihood is printed in the log file
738:
739: Revision 1.123 2006/03/20 10:52:43 brouard
740: * imach.c (Module): <title> changed, corresponds to .htm file
741: name. <head> headers where missing.
742:
743: * imach.c (Module): Weights can have a decimal point as for
744: English (a comma might work with a correct LC_NUMERIC environment,
745: otherwise the weight is truncated).
746: Modification of warning when the covariates values are not 0 or
747: 1.
748: Version 0.98g
749:
750: Revision 1.122 2006/03/20 09:45:41 brouard
751: (Module): Weights can have a decimal point as for
752: English (a comma might work with a correct LC_NUMERIC environment,
753: otherwise the weight is truncated).
754: Modification of warning when the covariates values are not 0 or
755: 1.
756: Version 0.98g
757:
758: Revision 1.121 2006/03/16 17:45:01 lievre
759: * imach.c (Module): Comments concerning covariates added
760:
761: * imach.c (Module): refinements in the computation of lli if
762: status=-2 in order to have more reliable computation if stepm is
763: not 1 month. Version 0.98f
764:
765: Revision 1.120 2006/03/16 15:10:38 lievre
766: (Module): refinements in the computation of lli if
767: status=-2 in order to have more reliable computation if stepm is
768: not 1 month. Version 0.98f
769:
770: Revision 1.119 2006/03/15 17:42:26 brouard
771: (Module): Bug if status = -2, the loglikelihood was
772: computed as likelihood omitting the logarithm. Version O.98e
773:
774: Revision 1.118 2006/03/14 18:20:07 brouard
775: (Module): varevsij Comments added explaining the second
776: table of variances if popbased=1 .
777: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
778: (Module): Function pstamp added
779: (Module): Version 0.98d
780:
781: Revision 1.117 2006/03/14 17:16:22 brouard
782: (Module): varevsij Comments added explaining the second
783: table of variances if popbased=1 .
784: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
785: (Module): Function pstamp added
786: (Module): Version 0.98d
787:
788: Revision 1.116 2006/03/06 10:29:27 brouard
789: (Module): Variance-covariance wrong links and
790: varian-covariance of ej. is needed (Saito).
791:
792: Revision 1.115 2006/02/27 12:17:45 brouard
793: (Module): One freematrix added in mlikeli! 0.98c
794:
795: Revision 1.114 2006/02/26 12:57:58 brouard
796: (Module): Some improvements in processing parameter
797: filename with strsep.
798:
799: Revision 1.113 2006/02/24 14:20:24 brouard
800: (Module): Memory leaks checks with valgrind and:
801: datafile was not closed, some imatrix were not freed and on matrix
802: allocation too.
803:
804: Revision 1.112 2006/01/30 09:55:26 brouard
805: (Module): Back to gnuplot.exe instead of wgnuplot.exe
806:
807: Revision 1.111 2006/01/25 20:38:18 brouard
808: (Module): Lots of cleaning and bugs added (Gompertz)
809: (Module): Comments can be added in data file. Missing date values
810: can be a simple dot '.'.
811:
812: Revision 1.110 2006/01/25 00:51:50 brouard
813: (Module): Lots of cleaning and bugs added (Gompertz)
814:
815: Revision 1.109 2006/01/24 19:37:15 brouard
816: (Module): Comments (lines starting with a #) are allowed in data.
817:
818: Revision 1.108 2006/01/19 18:05:42 lievre
819: Gnuplot problem appeared...
820: To be fixed
821:
822: Revision 1.107 2006/01/19 16:20:37 brouard
823: Test existence of gnuplot in imach path
824:
825: Revision 1.106 2006/01/19 13:24:36 brouard
826: Some cleaning and links added in html output
827:
828: Revision 1.105 2006/01/05 20:23:19 lievre
829: *** empty log message ***
830:
831: Revision 1.104 2005/09/30 16:11:43 lievre
832: (Module): sump fixed, loop imx fixed, and simplifications.
833: (Module): If the status is missing at the last wave but we know
834: that the person is alive, then we can code his/her status as -2
835: (instead of missing=-1 in earlier versions) and his/her
836: contributions to the likelihood is 1 - Prob of dying from last
837: health status (= 1-p13= p11+p12 in the easiest case of somebody in
838: the healthy state at last known wave). Version is 0.98
839:
840: Revision 1.103 2005/09/30 15:54:49 lievre
841: (Module): sump fixed, loop imx fixed, and simplifications.
842:
843: Revision 1.102 2004/09/15 17:31:30 brouard
844: Add the possibility to read data file including tab characters.
845:
846: Revision 1.101 2004/09/15 10:38:38 brouard
847: Fix on curr_time
848:
849: Revision 1.100 2004/07/12 18:29:06 brouard
850: Add version for Mac OS X. Just define UNIX in Makefile
851:
852: Revision 1.99 2004/06/05 08:57:40 brouard
853: *** empty log message ***
854:
855: Revision 1.98 2004/05/16 15:05:56 brouard
856: New version 0.97 . First attempt to estimate force of mortality
857: directly from the data i.e. without the need of knowing the health
858: state at each age, but using a Gompertz model: log u =a + b*age .
859: This is the basic analysis of mortality and should be done before any
860: other analysis, in order to test if the mortality estimated from the
861: cross-longitudinal survey is different from the mortality estimated
862: from other sources like vital statistic data.
863:
864: The same imach parameter file can be used but the option for mle should be -3.
865:
1.133 brouard 866: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 867: former routines in order to include the new code within the former code.
868:
869: The output is very simple: only an estimate of the intercept and of
870: the slope with 95% confident intervals.
871:
872: Current limitations:
873: A) Even if you enter covariates, i.e. with the
874: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
875: B) There is no computation of Life Expectancy nor Life Table.
876:
877: Revision 1.97 2004/02/20 13:25:42 lievre
878: Version 0.96d. Population forecasting command line is (temporarily)
879: suppressed.
880:
881: Revision 1.96 2003/07/15 15:38:55 brouard
882: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
883: rewritten within the same printf. Workaround: many printfs.
884:
885: Revision 1.95 2003/07/08 07:54:34 brouard
886: * imach.c (Repository):
887: (Repository): Using imachwizard code to output a more meaningful covariance
888: matrix (cov(a12,c31) instead of numbers.
889:
890: Revision 1.94 2003/06/27 13:00:02 brouard
891: Just cleaning
892:
893: Revision 1.93 2003/06/25 16:33:55 brouard
894: (Module): On windows (cygwin) function asctime_r doesn't
895: exist so I changed back to asctime which exists.
896: (Module): Version 0.96b
897:
898: Revision 1.92 2003/06/25 16:30:45 brouard
899: (Module): On windows (cygwin) function asctime_r doesn't
900: exist so I changed back to asctime which exists.
901:
902: Revision 1.91 2003/06/25 15:30:29 brouard
903: * imach.c (Repository): Duplicated warning errors corrected.
904: (Repository): Elapsed time after each iteration is now output. It
905: helps to forecast when convergence will be reached. Elapsed time
906: is stamped in powell. We created a new html file for the graphs
907: concerning matrix of covariance. It has extension -cov.htm.
908:
909: Revision 1.90 2003/06/24 12:34:15 brouard
910: (Module): Some bugs corrected for windows. Also, when
911: mle=-1 a template is output in file "or"mypar.txt with the design
912: of the covariance matrix to be input.
913:
914: Revision 1.89 2003/06/24 12:30:52 brouard
915: (Module): Some bugs corrected for windows. Also, when
916: mle=-1 a template is output in file "or"mypar.txt with the design
917: of the covariance matrix to be input.
918:
919: Revision 1.88 2003/06/23 17:54:56 brouard
920: * 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.
921:
922: Revision 1.87 2003/06/18 12:26:01 brouard
923: Version 0.96
924:
925: Revision 1.86 2003/06/17 20:04:08 brouard
926: (Module): Change position of html and gnuplot routines and added
927: routine fileappend.
928:
929: Revision 1.85 2003/06/17 13:12:43 brouard
930: * imach.c (Repository): Check when date of death was earlier that
931: current date of interview. It may happen when the death was just
932: prior to the death. In this case, dh was negative and likelihood
933: was wrong (infinity). We still send an "Error" but patch by
934: assuming that the date of death was just one stepm after the
935: interview.
936: (Repository): Because some people have very long ID (first column)
937: we changed int to long in num[] and we added a new lvector for
938: memory allocation. But we also truncated to 8 characters (left
939: truncation)
940: (Repository): No more line truncation errors.
941:
942: Revision 1.84 2003/06/13 21:44:43 brouard
943: * imach.c (Repository): Replace "freqsummary" at a correct
944: place. It differs from routine "prevalence" which may be called
945: many times. Probs is memory consuming and must be used with
946: parcimony.
947: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
948:
949: Revision 1.83 2003/06/10 13:39:11 lievre
950: *** empty log message ***
951:
952: Revision 1.82 2003/06/05 15:57:20 brouard
953: Add log in imach.c and fullversion number is now printed.
954:
955: */
956: /*
957: Interpolated Markov Chain
958:
959: Short summary of the programme:
960:
1.227 brouard 961: This program computes Healthy Life Expectancies or State-specific
962: (if states aren't health statuses) Expectancies from
963: cross-longitudinal data. Cross-longitudinal data consist in:
964:
965: -1- a first survey ("cross") where individuals from different ages
966: are interviewed on their health status or degree of disability (in
967: the case of a health survey which is our main interest)
968:
969: -2- at least a second wave of interviews ("longitudinal") which
970: measure each change (if any) in individual health status. Health
971: expectancies are computed from the time spent in each health state
972: according to a model. More health states you consider, more time is
973: necessary to reach the Maximum Likelihood of the parameters involved
974: in the model. The simplest model is the multinomial logistic model
975: where pij is the probability to be observed in state j at the second
976: wave conditional to be observed in state i at the first
977: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
978: etc , where 'age' is age and 'sex' is a covariate. If you want to
979: have a more complex model than "constant and age", you should modify
980: the program where the markup *Covariates have to be included here
981: again* invites you to do it. More covariates you add, slower the
1.126 brouard 982: convergence.
983:
984: The advantage of this computer programme, compared to a simple
985: multinomial logistic model, is clear when the delay between waves is not
986: identical for each individual. Also, if a individual missed an
987: intermediate interview, the information is lost, but taken into
988: account using an interpolation or extrapolation.
989:
990: hPijx is the probability to be observed in state i at age x+h
991: conditional to the observed state i at age x. The delay 'h' can be
992: split into an exact number (nh*stepm) of unobserved intermediate
993: states. This elementary transition (by month, quarter,
994: semester or year) is modelled as a multinomial logistic. The hPx
995: matrix is simply the matrix product of nh*stepm elementary matrices
996: and the contribution of each individual to the likelihood is simply
997: hPijx.
998:
999: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1000: of the life expectancies. It also computes the period (stable) prevalence.
1001:
1002: Back prevalence and projections:
1.227 brouard 1003:
1004: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1005: double agemaxpar, double ftolpl, int *ncvyearp, double
1006: dateprev1,double dateprev2, int firstpass, int lastpass, int
1007: mobilavproj)
1008:
1009: Computes the back prevalence limit for any combination of
1010: covariate values k at any age between ageminpar and agemaxpar and
1011: returns it in **bprlim. In the loops,
1012:
1013: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1014: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1015:
1016: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1017: Computes for any combination of covariates k and any age between bage and fage
1018: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1019: oldm=oldms;savm=savms;
1.227 brouard 1020:
1.267 brouard 1021: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1022: Computes the transition matrix starting at age 'age' over
1023: 'nhstepm*hstepm*stepm' months (i.e. until
1024: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1025: nhstepm*hstepm matrices.
1026:
1027: Returns p3mat[i][j][h] after calling
1028: p3mat[i][j][h]=matprod2(newm,
1029: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1030: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1031: oldm);
1.226 brouard 1032:
1033: Important routines
1034:
1035: - func (or funcone), computes logit (pij) distinguishing
1036: o fixed variables (single or product dummies or quantitative);
1037: o varying variables by:
1038: (1) wave (single, product dummies, quantitative),
1039: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1040: % fixed dummy (treated) or quantitative (not done because time-consuming);
1041: % varying dummy (not done) or quantitative (not done);
1042: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1043: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1044: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1045: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1046: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1047:
1.226 brouard 1048:
1049:
1.133 brouard 1050: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1051: Institut national d'études démographiques, Paris.
1.126 brouard 1052: This software have been partly granted by Euro-REVES, a concerted action
1053: from the European Union.
1054: It is copyrighted identically to a GNU software product, ie programme and
1055: software can be distributed freely for non commercial use. Latest version
1056: can be accessed at http://euroreves.ined.fr/imach .
1057:
1058: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1059: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1060:
1061: **********************************************************************/
1062: /*
1063: main
1064: read parameterfile
1065: read datafile
1066: concatwav
1067: freqsummary
1068: if (mle >= 1)
1069: mlikeli
1070: print results files
1071: if mle==1
1072: computes hessian
1073: read end of parameter file: agemin, agemax, bage, fage, estepm
1074: begin-prev-date,...
1075: open gnuplot file
1076: open html file
1.145 brouard 1077: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1078: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1079: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1080: freexexit2 possible for memory heap.
1081:
1082: h Pij x | pij_nom ficrestpij
1083: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1084: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1085: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1086:
1087: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1088: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1089: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1090: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1091: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1092:
1.126 brouard 1093: forecasting if prevfcast==1 prevforecast call prevalence()
1094: health expectancies
1095: Variance-covariance of DFLE
1096: prevalence()
1097: movingaverage()
1098: varevsij()
1099: if popbased==1 varevsij(,popbased)
1100: total life expectancies
1101: Variance of period (stable) prevalence
1102: end
1103: */
1104:
1.187 brouard 1105: /* #define DEBUG */
1106: /* #define DEBUGBRENT */
1.203 brouard 1107: /* #define DEBUGLINMIN */
1108: /* #define DEBUGHESS */
1109: #define DEBUGHESSIJ
1.224 brouard 1110: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1111: #define POWELL /* Instead of NLOPT */
1.224 brouard 1112: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1113: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1114: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1115: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1116:
1117: #include <math.h>
1118: #include <stdio.h>
1119: #include <stdlib.h>
1120: #include <string.h>
1.226 brouard 1121: #include <ctype.h>
1.159 brouard 1122:
1123: #ifdef _WIN32
1124: #include <io.h>
1.172 brouard 1125: #include <windows.h>
1126: #include <tchar.h>
1.159 brouard 1127: #else
1.126 brouard 1128: #include <unistd.h>
1.159 brouard 1129: #endif
1.126 brouard 1130:
1131: #include <limits.h>
1132: #include <sys/types.h>
1.171 brouard 1133:
1134: #if defined(__GNUC__)
1135: #include <sys/utsname.h> /* Doesn't work on Windows */
1136: #endif
1137:
1.126 brouard 1138: #include <sys/stat.h>
1139: #include <errno.h>
1.159 brouard 1140: /* extern int errno; */
1.126 brouard 1141:
1.157 brouard 1142: /* #ifdef LINUX */
1143: /* #include <time.h> */
1144: /* #include "timeval.h" */
1145: /* #else */
1146: /* #include <sys/time.h> */
1147: /* #endif */
1148:
1.126 brouard 1149: #include <time.h>
1150:
1.136 brouard 1151: #ifdef GSL
1152: #include <gsl/gsl_errno.h>
1153: #include <gsl/gsl_multimin.h>
1154: #endif
1155:
1.167 brouard 1156:
1.162 brouard 1157: #ifdef NLOPT
1158: #include <nlopt.h>
1159: typedef struct {
1160: double (* function)(double [] );
1161: } myfunc_data ;
1162: #endif
1163:
1.126 brouard 1164: /* #include <libintl.h> */
1165: /* #define _(String) gettext (String) */
1166:
1.251 brouard 1167: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1168:
1169: #define GNUPLOTPROGRAM "gnuplot"
1170: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1171: #define FILENAMELENGTH 132
1172:
1173: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1174: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1175:
1.144 brouard 1176: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1177: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1178:
1179: #define NINTERVMAX 8
1.144 brouard 1180: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1181: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.318 brouard 1182: #define NCOVMAX 30 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1183: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1184: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1185: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1186: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1187: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1188: /* #define AGESUP 130 */
1.288 brouard 1189: /* #define AGESUP 150 */
1190: #define AGESUP 200
1.268 brouard 1191: #define AGEINF 0
1.218 brouard 1192: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1193: #define AGEBASE 40
1.194 brouard 1194: #define AGEOVERFLOW 1.e20
1.164 brouard 1195: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1196: #ifdef _WIN32
1197: #define DIRSEPARATOR '\\'
1198: #define CHARSEPARATOR "\\"
1199: #define ODIRSEPARATOR '/'
1200: #else
1.126 brouard 1201: #define DIRSEPARATOR '/'
1202: #define CHARSEPARATOR "/"
1203: #define ODIRSEPARATOR '\\'
1204: #endif
1205:
1.322 ! brouard 1206: /* $Id: imach.c,v 1.321 2022/07/22 12:04:24 brouard Exp $ */
1.126 brouard 1207: /* $State: Exp $ */
1.196 brouard 1208: #include "version.h"
1209: char version[]=__IMACH_VERSION__;
1.316 brouard 1210: char copyright[]="May 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";
1.322 ! brouard 1211: char fullversion[]="$Revision: 1.321 $ $Date: 2022/07/22 12:04:24 $";
1.126 brouard 1212: char strstart[80];
1213: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1214: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1215: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1216: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1217: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1218: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1219: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1220: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1221: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1222: int cptcovprodnoage=0; /**< Number of covariate products without age */
1223: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1224: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1225: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1226: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1227: int nsd=0; /**< Total number of single dummy variables (output) */
1228: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1229: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1230: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1231: int ntveff=0; /**< ntveff number of effective time varying variables */
1232: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1233: int cptcov=0; /* Working variable */
1.290 brouard 1234: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1235: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1236: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1237: int nlstate=2; /* Number of live states */
1238: int ndeath=1; /* Number of dead states */
1.130 brouard 1239: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1240: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1241: int popbased=0;
1242:
1243: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1244: int maxwav=0; /* Maxim number of waves */
1245: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1246: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1247: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1248: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1249: int mle=1, weightopt=0;
1.126 brouard 1250: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1251: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1252: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1253: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1254: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1255: int selected(int kvar); /* Is covariate kvar selected for printing results */
1256:
1.130 brouard 1257: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1258: double **matprod2(); /* test */
1.126 brouard 1259: double **oldm, **newm, **savm; /* Working pointers to matrices */
1260: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1261: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1262:
1.136 brouard 1263: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1264: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1265: FILE *ficlog, *ficrespow;
1.130 brouard 1266: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1267: double fretone; /* Only one call to likelihood */
1.130 brouard 1268: long ipmx=0; /* Number of contributions */
1.126 brouard 1269: double sw; /* Sum of weights */
1270: char filerespow[FILENAMELENGTH];
1271: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1272: FILE *ficresilk;
1273: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1274: FILE *ficresprobmorprev;
1275: FILE *fichtm, *fichtmcov; /* Html File */
1276: FILE *ficreseij;
1277: char filerese[FILENAMELENGTH];
1278: FILE *ficresstdeij;
1279: char fileresstde[FILENAMELENGTH];
1280: FILE *ficrescveij;
1281: char filerescve[FILENAMELENGTH];
1282: FILE *ficresvij;
1283: char fileresv[FILENAMELENGTH];
1.269 brouard 1284:
1.126 brouard 1285: char title[MAXLINE];
1.234 brouard 1286: char model[MAXLINE]; /**< The model line */
1.217 brouard 1287: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1288: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1289: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1290: char command[FILENAMELENGTH];
1291: int outcmd=0;
1292:
1.217 brouard 1293: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1294: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1295: char filelog[FILENAMELENGTH]; /* Log file */
1296: char filerest[FILENAMELENGTH];
1297: char fileregp[FILENAMELENGTH];
1298: char popfile[FILENAMELENGTH];
1299:
1300: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1301:
1.157 brouard 1302: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1303: /* struct timezone tzp; */
1304: /* extern int gettimeofday(); */
1305: struct tm tml, *gmtime(), *localtime();
1306:
1307: extern time_t time();
1308:
1309: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1310: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1311: struct tm tm;
1312:
1.126 brouard 1313: char strcurr[80], strfor[80];
1314:
1315: char *endptr;
1316: long lval;
1317: double dval;
1318:
1319: #define NR_END 1
1320: #define FREE_ARG char*
1321: #define FTOL 1.0e-10
1322:
1323: #define NRANSI
1.240 brouard 1324: #define ITMAX 200
1325: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1326:
1327: #define TOL 2.0e-4
1328:
1329: #define CGOLD 0.3819660
1330: #define ZEPS 1.0e-10
1331: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1332:
1333: #define GOLD 1.618034
1334: #define GLIMIT 100.0
1335: #define TINY 1.0e-20
1336:
1337: static double maxarg1,maxarg2;
1338: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1339: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1340:
1341: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1342: #define rint(a) floor(a+0.5)
1.166 brouard 1343: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1344: #define mytinydouble 1.0e-16
1.166 brouard 1345: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1346: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1347: /* static double dsqrarg; */
1348: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1349: static double sqrarg;
1350: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1351: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1352: int agegomp= AGEGOMP;
1353:
1354: int imx;
1355: int stepm=1;
1356: /* Stepm, step in month: minimum step interpolation*/
1357:
1358: int estepm;
1359: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1360:
1361: int m,nb;
1362: long *num;
1.197 brouard 1363: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1364: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1365: covariate for which somebody answered excluding
1366: undefined. Usually 2: 0 and 1. */
1367: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1368: covariate for which somebody answered including
1369: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1370: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1371: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1372: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1373: double *ageexmed,*agecens;
1374: double dateintmean=0;
1.296 brouard 1375: double anprojd, mprojd, jprojd; /* For eventual projections */
1376: double anprojf, mprojf, jprojf;
1.126 brouard 1377:
1.296 brouard 1378: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1379: double anbackf, mbackf, jbackf;
1380: double jintmean,mintmean,aintmean;
1.126 brouard 1381: double *weight;
1382: int **s; /* Status */
1.141 brouard 1383: double *agedc;
1.145 brouard 1384: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1385: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1386: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1387: double **coqvar; /* Fixed quantitative covariate nqv */
1388: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1389: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1390: double idx;
1391: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1392: /* Some documentation */
1393: /* Design original data
1394: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1395: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1396: * ntv=3 nqtv=1
1397: * cptcovn number of covariates (not including constant and age) = # of + plus 1 = 10+1=11
1398: * For time varying covariate, quanti or dummies
1399: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1400: * cotvar[wav][ntv+iv][i]= [3+(1 to nqtv)][i]=(V12) quanti
1401: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1402: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1403: * covar[k,i], value of kth fixed covariate dummy or quanti :
1404: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1405: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1406: * k= 1 2 3 4 5 6 7 8 9 10 11
1407: */
1408: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1409: /* 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
1410: # States 1=Coresidence, 2 Living alone, 3 Institution
1411: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1412: */
1.319 brouard 1413: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1414: /* k 1 2 3 4 5 6 7 8 9 */
1415: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1416: /* fixed or varying), 1 for age product, 2 for*/
1417: /* product */
1418: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1419: /*(single or product without age), 2 dummy*/
1420: /* with age product, 3 quant with age product*/
1421: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1422: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1423: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1424: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1425: /* nsq 1 2 */ /* Counting single quantit tv */
1426: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1427: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1428: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1429: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1430: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1431: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1432: /* 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 1433: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1434: /* Type */
1435: /* V 1 2 3 4 5 */
1436: /* F F V V V */
1437: /* D Q D D Q */
1438: /* */
1439: int *TvarsD;
1440: int *TvarsDind;
1441: int *TvarsQ;
1442: int *TvarsQind;
1443:
1.318 brouard 1444: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1445: int nresult=0;
1.258 brouard 1446: int parameterline=0; /* # of the parameter (type) line */
1.318 brouard 1447: int TKresult[MAXRESULTLINESPONE];
1448: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable , value (output) */
1449: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable , value (output) */
1450: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For dummy variable , variable # (output) */
1451: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1452: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1453: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , variable # (output) */
1454:
1455: /* 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
1456: # States 1=Coresidence, 2 Living alone, 3 Institution
1457: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1458: */
1.234 brouard 1459: /* 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 1460: 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 */
1461: 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 */
1462: 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 */
1463: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1464: 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 */
1465: 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 1466: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1467: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1468: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1469: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1470: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1471: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1472: 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 */
1473: 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 */
1474:
1.230 brouard 1475: int *Tvarsel; /**< Selected covariates for output */
1476: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1477: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1478: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1479: 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 1480: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1481: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1482: int *Tage;
1.227 brouard 1483: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1484: 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 1485: 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*/
1486: 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 1487: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1488: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1489: int **Tvard;
1490: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1491: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1492: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1493: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1494: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1495: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1496: double *lsurv, *lpop, *tpop;
1497:
1.231 brouard 1498: #define FD 1; /* Fixed dummy covariate */
1499: #define FQ 2; /* Fixed quantitative covariate */
1500: #define FP 3; /* Fixed product covariate */
1501: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1502: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1503: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1504: #define VD 10; /* Varying dummy covariate */
1505: #define VQ 11; /* Varying quantitative covariate */
1506: #define VP 12; /* Varying product covariate */
1507: #define VPDD 13; /* Varying product dummy*dummy covariate */
1508: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1509: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1510: #define APFD 16; /* Age product * fixed dummy covariate */
1511: #define APFQ 17; /* Age product * fixed quantitative covariate */
1512: #define APVD 18; /* Age product * varying dummy covariate */
1513: #define APVQ 19; /* Age product * varying quantitative covariate */
1514:
1515: #define FTYPE 1; /* Fixed covariate */
1516: #define VTYPE 2; /* Varying covariate (loop in wave) */
1517: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1518:
1519: struct kmodel{
1520: int maintype; /* main type */
1521: int subtype; /* subtype */
1522: };
1523: struct kmodel modell[NCOVMAX];
1524:
1.143 brouard 1525: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1526: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1527:
1528: /**************** split *************************/
1529: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1530: {
1531: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1532: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1533: */
1534: char *ss; /* pointer */
1.186 brouard 1535: int l1=0, l2=0; /* length counters */
1.126 brouard 1536:
1537: l1 = strlen(path ); /* length of path */
1538: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1539: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1540: if ( ss == NULL ) { /* no directory, so determine current directory */
1541: strcpy( name, path ); /* we got the fullname name because no directory */
1542: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1543: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1544: /* get current working directory */
1545: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1546: #ifdef WIN32
1547: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1548: #else
1549: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1550: #endif
1.126 brouard 1551: return( GLOCK_ERROR_GETCWD );
1552: }
1553: /* got dirc from getcwd*/
1554: printf(" DIRC = %s \n",dirc);
1.205 brouard 1555: } else { /* strip directory from path */
1.126 brouard 1556: ss++; /* after this, the filename */
1557: l2 = strlen( ss ); /* length of filename */
1558: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1559: strcpy( name, ss ); /* save file name */
1560: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1561: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1562: printf(" DIRC2 = %s \n",dirc);
1563: }
1564: /* We add a separator at the end of dirc if not exists */
1565: l1 = strlen( dirc ); /* length of directory */
1566: if( dirc[l1-1] != DIRSEPARATOR ){
1567: dirc[l1] = DIRSEPARATOR;
1568: dirc[l1+1] = 0;
1569: printf(" DIRC3 = %s \n",dirc);
1570: }
1571: ss = strrchr( name, '.' ); /* find last / */
1572: if (ss >0){
1573: ss++;
1574: strcpy(ext,ss); /* save extension */
1575: l1= strlen( name);
1576: l2= strlen(ss)+1;
1577: strncpy( finame, name, l1-l2);
1578: finame[l1-l2]= 0;
1579: }
1580:
1581: return( 0 ); /* we're done */
1582: }
1583:
1584:
1585: /******************************************/
1586:
1587: void replace_back_to_slash(char *s, char*t)
1588: {
1589: int i;
1590: int lg=0;
1591: i=0;
1592: lg=strlen(t);
1593: for(i=0; i<= lg; i++) {
1594: (s[i] = t[i]);
1595: if (t[i]== '\\') s[i]='/';
1596: }
1597: }
1598:
1.132 brouard 1599: char *trimbb(char *out, char *in)
1.137 brouard 1600: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1601: char *s;
1602: s=out;
1603: while (*in != '\0'){
1.137 brouard 1604: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1605: in++;
1606: }
1607: *out++ = *in++;
1608: }
1609: *out='\0';
1610: return s;
1611: }
1612:
1.187 brouard 1613: /* char *substrchaine(char *out, char *in, char *chain) */
1614: /* { */
1615: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1616: /* char *s, *t; */
1617: /* t=in;s=out; */
1618: /* while ((*in != *chain) && (*in != '\0')){ */
1619: /* *out++ = *in++; */
1620: /* } */
1621:
1622: /* /\* *in matches *chain *\/ */
1623: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1624: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1625: /* } */
1626: /* in--; chain--; */
1627: /* while ( (*in != '\0')){ */
1628: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1629: /* *out++ = *in++; */
1630: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1631: /* } */
1632: /* *out='\0'; */
1633: /* out=s; */
1634: /* return out; */
1635: /* } */
1636: char *substrchaine(char *out, char *in, char *chain)
1637: {
1638: /* Substract chain 'chain' from 'in', return and output 'out' */
1639: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1640:
1641: char *strloc;
1642:
1643: strcpy (out, in);
1644: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1645: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1646: if(strloc != NULL){
1647: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1648: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1649: /* strcpy (strloc, strloc +strlen(chain));*/
1650: }
1651: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1652: return out;
1653: }
1654:
1655:
1.145 brouard 1656: char *cutl(char *blocc, char *alocc, char *in, char occ)
1657: {
1.187 brouard 1658: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1659: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1660: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1661: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1662: */
1.160 brouard 1663: char *s, *t;
1.145 brouard 1664: t=in;s=in;
1665: while ((*in != occ) && (*in != '\0')){
1666: *alocc++ = *in++;
1667: }
1668: if( *in == occ){
1669: *(alocc)='\0';
1670: s=++in;
1671: }
1672:
1673: if (s == t) {/* occ not found */
1674: *(alocc-(in-s))='\0';
1675: in=s;
1676: }
1677: while ( *in != '\0'){
1678: *blocc++ = *in++;
1679: }
1680:
1681: *blocc='\0';
1682: return t;
1683: }
1.137 brouard 1684: char *cutv(char *blocc, char *alocc, char *in, char occ)
1685: {
1.187 brouard 1686: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1687: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1688: gives blocc="abcdef2ghi" and alocc="j".
1689: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1690: */
1691: char *s, *t;
1692: t=in;s=in;
1693: while (*in != '\0'){
1694: while( *in == occ){
1695: *blocc++ = *in++;
1696: s=in;
1697: }
1698: *blocc++ = *in++;
1699: }
1700: if (s == t) /* occ not found */
1701: *(blocc-(in-s))='\0';
1702: else
1703: *(blocc-(in-s)-1)='\0';
1704: in=s;
1705: while ( *in != '\0'){
1706: *alocc++ = *in++;
1707: }
1708:
1709: *alocc='\0';
1710: return s;
1711: }
1712:
1.126 brouard 1713: int nbocc(char *s, char occ)
1714: {
1715: int i,j=0;
1716: int lg=20;
1717: i=0;
1718: lg=strlen(s);
1719: for(i=0; i<= lg; i++) {
1.234 brouard 1720: if (s[i] == occ ) j++;
1.126 brouard 1721: }
1722: return j;
1723: }
1724:
1.137 brouard 1725: /* void cutv(char *u,char *v, char*t, char occ) */
1726: /* { */
1727: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1728: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1729: /* gives u="abcdef2ghi" and v="j" *\/ */
1730: /* int i,lg,j,p=0; */
1731: /* i=0; */
1732: /* lg=strlen(t); */
1733: /* for(j=0; j<=lg-1; j++) { */
1734: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1735: /* } */
1.126 brouard 1736:
1.137 brouard 1737: /* for(j=0; j<p; j++) { */
1738: /* (u[j] = t[j]); */
1739: /* } */
1740: /* u[p]='\0'; */
1.126 brouard 1741:
1.137 brouard 1742: /* for(j=0; j<= lg; j++) { */
1743: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1744: /* } */
1745: /* } */
1.126 brouard 1746:
1.160 brouard 1747: #ifdef _WIN32
1748: char * strsep(char **pp, const char *delim)
1749: {
1750: char *p, *q;
1751:
1752: if ((p = *pp) == NULL)
1753: return 0;
1754: if ((q = strpbrk (p, delim)) != NULL)
1755: {
1756: *pp = q + 1;
1757: *q = '\0';
1758: }
1759: else
1760: *pp = 0;
1761: return p;
1762: }
1763: #endif
1764:
1.126 brouard 1765: /********************** nrerror ********************/
1766:
1767: void nrerror(char error_text[])
1768: {
1769: fprintf(stderr,"ERREUR ...\n");
1770: fprintf(stderr,"%s\n",error_text);
1771: exit(EXIT_FAILURE);
1772: }
1773: /*********************** vector *******************/
1774: double *vector(int nl, int nh)
1775: {
1776: double *v;
1777: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1778: if (!v) nrerror("allocation failure in vector");
1779: return v-nl+NR_END;
1780: }
1781:
1782: /************************ free vector ******************/
1783: void free_vector(double*v, int nl, int nh)
1784: {
1785: free((FREE_ARG)(v+nl-NR_END));
1786: }
1787:
1788: /************************ivector *******************************/
1789: int *ivector(long nl,long nh)
1790: {
1791: int *v;
1792: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1793: if (!v) nrerror("allocation failure in ivector");
1794: return v-nl+NR_END;
1795: }
1796:
1797: /******************free ivector **************************/
1798: void free_ivector(int *v, long nl, long nh)
1799: {
1800: free((FREE_ARG)(v+nl-NR_END));
1801: }
1802:
1803: /************************lvector *******************************/
1804: long *lvector(long nl,long nh)
1805: {
1806: long *v;
1807: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1808: if (!v) nrerror("allocation failure in ivector");
1809: return v-nl+NR_END;
1810: }
1811:
1812: /******************free lvector **************************/
1813: void free_lvector(long *v, long nl, long nh)
1814: {
1815: free((FREE_ARG)(v+nl-NR_END));
1816: }
1817:
1818: /******************* imatrix *******************************/
1819: int **imatrix(long nrl, long nrh, long ncl, long nch)
1820: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1821: {
1822: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1823: int **m;
1824:
1825: /* allocate pointers to rows */
1826: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1827: if (!m) nrerror("allocation failure 1 in matrix()");
1828: m += NR_END;
1829: m -= nrl;
1830:
1831:
1832: /* allocate rows and set pointers to them */
1833: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1834: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1835: m[nrl] += NR_END;
1836: m[nrl] -= ncl;
1837:
1838: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1839:
1840: /* return pointer to array of pointers to rows */
1841: return m;
1842: }
1843:
1844: /****************** free_imatrix *************************/
1845: void free_imatrix(m,nrl,nrh,ncl,nch)
1846: int **m;
1847: long nch,ncl,nrh,nrl;
1848: /* free an int matrix allocated by imatrix() */
1849: {
1850: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1851: free((FREE_ARG) (m+nrl-NR_END));
1852: }
1853:
1854: /******************* matrix *******************************/
1855: double **matrix(long nrl, long nrh, long ncl, long nch)
1856: {
1857: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1858: double **m;
1859:
1860: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1861: if (!m) nrerror("allocation failure 1 in matrix()");
1862: m += NR_END;
1863: m -= nrl;
1864:
1865: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1866: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1867: m[nrl] += NR_END;
1868: m[nrl] -= ncl;
1869:
1870: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1871: return m;
1.145 brouard 1872: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1873: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1874: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1875: */
1876: }
1877:
1878: /*************************free matrix ************************/
1879: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1880: {
1881: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1882: free((FREE_ARG)(m+nrl-NR_END));
1883: }
1884:
1885: /******************* ma3x *******************************/
1886: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1887: {
1888: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1889: double ***m;
1890:
1891: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1892: if (!m) nrerror("allocation failure 1 in matrix()");
1893: m += NR_END;
1894: m -= nrl;
1895:
1896: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1897: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1898: m[nrl] += NR_END;
1899: m[nrl] -= ncl;
1900:
1901: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1902:
1903: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1904: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1905: m[nrl][ncl] += NR_END;
1906: m[nrl][ncl] -= nll;
1907: for (j=ncl+1; j<=nch; j++)
1908: m[nrl][j]=m[nrl][j-1]+nlay;
1909:
1910: for (i=nrl+1; i<=nrh; i++) {
1911: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1912: for (j=ncl+1; j<=nch; j++)
1913: m[i][j]=m[i][j-1]+nlay;
1914: }
1915: return m;
1916: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1917: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1918: */
1919: }
1920:
1921: /*************************free ma3x ************************/
1922: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1923: {
1924: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1925: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1926: free((FREE_ARG)(m+nrl-NR_END));
1927: }
1928:
1929: /*************** function subdirf ***********/
1930: char *subdirf(char fileres[])
1931: {
1932: /* Caution optionfilefiname is hidden */
1933: strcpy(tmpout,optionfilefiname);
1934: strcat(tmpout,"/"); /* Add to the right */
1935: strcat(tmpout,fileres);
1936: return tmpout;
1937: }
1938:
1939: /*************** function subdirf2 ***********/
1940: char *subdirf2(char fileres[], char *preop)
1941: {
1.314 brouard 1942: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
1943: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 1944: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 1945: /* Caution optionfilefiname is hidden */
1946: strcpy(tmpout,optionfilefiname);
1947: strcat(tmpout,"/");
1948: strcat(tmpout,preop);
1949: strcat(tmpout,fileres);
1950: return tmpout;
1951: }
1952:
1953: /*************** function subdirf3 ***********/
1954: char *subdirf3(char fileres[], char *preop, char *preop2)
1955: {
1956:
1957: /* Caution optionfilefiname is hidden */
1958: strcpy(tmpout,optionfilefiname);
1959: strcat(tmpout,"/");
1960: strcat(tmpout,preop);
1961: strcat(tmpout,preop2);
1962: strcat(tmpout,fileres);
1963: return tmpout;
1964: }
1.213 brouard 1965:
1966: /*************** function subdirfext ***********/
1967: char *subdirfext(char fileres[], char *preop, char *postop)
1968: {
1969:
1970: strcpy(tmpout,preop);
1971: strcat(tmpout,fileres);
1972: strcat(tmpout,postop);
1973: return tmpout;
1974: }
1.126 brouard 1975:
1.213 brouard 1976: /*************** function subdirfext3 ***********/
1977: char *subdirfext3(char fileres[], char *preop, char *postop)
1978: {
1979:
1980: /* Caution optionfilefiname is hidden */
1981: strcpy(tmpout,optionfilefiname);
1982: strcat(tmpout,"/");
1983: strcat(tmpout,preop);
1984: strcat(tmpout,fileres);
1985: strcat(tmpout,postop);
1986: return tmpout;
1987: }
1988:
1.162 brouard 1989: char *asc_diff_time(long time_sec, char ascdiff[])
1990: {
1991: long sec_left, days, hours, minutes;
1992: days = (time_sec) / (60*60*24);
1993: sec_left = (time_sec) % (60*60*24);
1994: hours = (sec_left) / (60*60) ;
1995: sec_left = (sec_left) %(60*60);
1996: minutes = (sec_left) /60;
1997: sec_left = (sec_left) % (60);
1998: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1999: return ascdiff;
2000: }
2001:
1.126 brouard 2002: /***************** f1dim *************************/
2003: extern int ncom;
2004: extern double *pcom,*xicom;
2005: extern double (*nrfunc)(double []);
2006:
2007: double f1dim(double x)
2008: {
2009: int j;
2010: double f;
2011: double *xt;
2012:
2013: xt=vector(1,ncom);
2014: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2015: f=(*nrfunc)(xt);
2016: free_vector(xt,1,ncom);
2017: return f;
2018: }
2019:
2020: /*****************brent *************************/
2021: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2022: {
2023: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2024: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2025: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2026: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2027: * returned function value.
2028: */
1.126 brouard 2029: int iter;
2030: double a,b,d,etemp;
1.159 brouard 2031: double fu=0,fv,fw,fx;
1.164 brouard 2032: double ftemp=0.;
1.126 brouard 2033: double p,q,r,tol1,tol2,u,v,w,x,xm;
2034: double e=0.0;
2035:
2036: a=(ax < cx ? ax : cx);
2037: b=(ax > cx ? ax : cx);
2038: x=w=v=bx;
2039: fw=fv=fx=(*f)(x);
2040: for (iter=1;iter<=ITMAX;iter++) {
2041: xm=0.5*(a+b);
2042: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2043: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2044: printf(".");fflush(stdout);
2045: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2046: #ifdef DEBUGBRENT
1.126 brouard 2047: 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);
2048: 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);
2049: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2050: #endif
2051: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2052: *xmin=x;
2053: return fx;
2054: }
2055: ftemp=fu;
2056: if (fabs(e) > tol1) {
2057: r=(x-w)*(fx-fv);
2058: q=(x-v)*(fx-fw);
2059: p=(x-v)*q-(x-w)*r;
2060: q=2.0*(q-r);
2061: if (q > 0.0) p = -p;
2062: q=fabs(q);
2063: etemp=e;
2064: e=d;
2065: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2066: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2067: else {
1.224 brouard 2068: d=p/q;
2069: u=x+d;
2070: if (u-a < tol2 || b-u < tol2)
2071: d=SIGN(tol1,xm-x);
1.126 brouard 2072: }
2073: } else {
2074: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2075: }
2076: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2077: fu=(*f)(u);
2078: if (fu <= fx) {
2079: if (u >= x) a=x; else b=x;
2080: SHFT(v,w,x,u)
1.183 brouard 2081: SHFT(fv,fw,fx,fu)
2082: } else {
2083: if (u < x) a=u; else b=u;
2084: if (fu <= fw || w == x) {
1.224 brouard 2085: v=w;
2086: w=u;
2087: fv=fw;
2088: fw=fu;
1.183 brouard 2089: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2090: v=u;
2091: fv=fu;
1.183 brouard 2092: }
2093: }
1.126 brouard 2094: }
2095: nrerror("Too many iterations in brent");
2096: *xmin=x;
2097: return fx;
2098: }
2099:
2100: /****************** mnbrak ***********************/
2101:
2102: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2103: double (*func)(double))
1.183 brouard 2104: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2105: the downhill direction (defined by the function as evaluated at the initial points) and returns
2106: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2107: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2108: */
1.126 brouard 2109: double ulim,u,r,q, dum;
2110: double fu;
1.187 brouard 2111:
2112: double scale=10.;
2113: int iterscale=0;
2114:
2115: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2116: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2117:
2118:
2119: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2120: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2121: /* *bx = *ax - (*ax - *bx)/scale; */
2122: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2123: /* } */
2124:
1.126 brouard 2125: if (*fb > *fa) {
2126: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2127: SHFT(dum,*fb,*fa,dum)
2128: }
1.126 brouard 2129: *cx=(*bx)+GOLD*(*bx-*ax);
2130: *fc=(*func)(*cx);
1.183 brouard 2131: #ifdef DEBUG
1.224 brouard 2132: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2133: 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 2134: #endif
1.224 brouard 2135: 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 2136: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2137: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2138: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2139: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2140: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2141: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2142: fu=(*func)(u);
1.163 brouard 2143: #ifdef DEBUG
2144: /* f(x)=A(x-u)**2+f(u) */
2145: double A, fparabu;
2146: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2147: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2148: 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);
2149: 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 2150: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2151: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2152: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2153: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2154: #endif
1.184 brouard 2155: #ifdef MNBRAKORIGINAL
1.183 brouard 2156: #else
1.191 brouard 2157: /* if (fu > *fc) { */
2158: /* #ifdef DEBUG */
2159: /* printf("mnbrak4 fu > fc \n"); */
2160: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2161: /* #endif */
2162: /* /\* 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 *\\/ *\/ */
2163: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2164: /* dum=u; /\* Shifting c and u *\/ */
2165: /* u = *cx; */
2166: /* *cx = dum; */
2167: /* dum = fu; */
2168: /* fu = *fc; */
2169: /* *fc =dum; */
2170: /* } else { /\* end *\/ */
2171: /* #ifdef DEBUG */
2172: /* printf("mnbrak3 fu < fc \n"); */
2173: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2174: /* #endif */
2175: /* dum=u; /\* Shifting c and u *\/ */
2176: /* u = *cx; */
2177: /* *cx = dum; */
2178: /* dum = fu; */
2179: /* fu = *fc; */
2180: /* *fc =dum; */
2181: /* } */
1.224 brouard 2182: #ifdef DEBUGMNBRAK
2183: double A, fparabu;
2184: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2185: fparabu= *fa - A*(*ax-u)*(*ax-u);
2186: 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);
2187: 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 2188: #endif
1.191 brouard 2189: dum=u; /* Shifting c and u */
2190: u = *cx;
2191: *cx = dum;
2192: dum = fu;
2193: fu = *fc;
2194: *fc =dum;
1.183 brouard 2195: #endif
1.162 brouard 2196: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2197: #ifdef DEBUG
1.224 brouard 2198: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2199: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2200: #endif
1.126 brouard 2201: fu=(*func)(u);
2202: if (fu < *fc) {
1.183 brouard 2203: #ifdef DEBUG
1.224 brouard 2204: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2205: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2206: #endif
2207: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2208: SHFT(*fb,*fc,fu,(*func)(u))
2209: #ifdef DEBUG
2210: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2211: #endif
2212: }
1.162 brouard 2213: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2214: #ifdef DEBUG
1.224 brouard 2215: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2216: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2217: #endif
1.126 brouard 2218: u=ulim;
2219: fu=(*func)(u);
1.183 brouard 2220: } else { /* u could be left to b (if r > q parabola has a maximum) */
2221: #ifdef DEBUG
1.224 brouard 2222: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2223: 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 2224: #endif
1.126 brouard 2225: u=(*cx)+GOLD*(*cx-*bx);
2226: fu=(*func)(u);
1.224 brouard 2227: #ifdef DEBUG
2228: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2229: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2230: #endif
1.183 brouard 2231: } /* end tests */
1.126 brouard 2232: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2233: SHFT(*fa,*fb,*fc,fu)
2234: #ifdef DEBUG
1.224 brouard 2235: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2236: 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 2237: #endif
2238: } /* 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 2239: }
2240:
2241: /*************** linmin ************************/
1.162 brouard 2242: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2243: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2244: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2245: the value of func at the returned location p . This is actually all accomplished by calling the
2246: routines mnbrak and brent .*/
1.126 brouard 2247: int ncom;
2248: double *pcom,*xicom;
2249: double (*nrfunc)(double []);
2250:
1.224 brouard 2251: #ifdef LINMINORIGINAL
1.126 brouard 2252: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2253: #else
2254: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2255: #endif
1.126 brouard 2256: {
2257: double brent(double ax, double bx, double cx,
2258: double (*f)(double), double tol, double *xmin);
2259: double f1dim(double x);
2260: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2261: double *fc, double (*func)(double));
2262: int j;
2263: double xx,xmin,bx,ax;
2264: double fx,fb,fa;
1.187 brouard 2265:
1.203 brouard 2266: #ifdef LINMINORIGINAL
2267: #else
2268: double scale=10., axs, xxs; /* Scale added for infinity */
2269: #endif
2270:
1.126 brouard 2271: ncom=n;
2272: pcom=vector(1,n);
2273: xicom=vector(1,n);
2274: nrfunc=func;
2275: for (j=1;j<=n;j++) {
2276: pcom[j]=p[j];
1.202 brouard 2277: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2278: }
1.187 brouard 2279:
1.203 brouard 2280: #ifdef LINMINORIGINAL
2281: xx=1.;
2282: #else
2283: axs=0.0;
2284: xxs=1.;
2285: do{
2286: xx= xxs;
2287: #endif
1.187 brouard 2288: ax=0.;
2289: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2290: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2291: /* 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)) */
2292: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2293: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2294: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2295: /* 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 2296: #ifdef LINMINORIGINAL
2297: #else
2298: if (fx != fx){
1.224 brouard 2299: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2300: printf("|");
2301: fprintf(ficlog,"|");
1.203 brouard 2302: #ifdef DEBUGLINMIN
1.224 brouard 2303: 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 2304: #endif
2305: }
1.224 brouard 2306: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2307: #endif
2308:
1.191 brouard 2309: #ifdef DEBUGLINMIN
2310: 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 2311: 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 2312: #endif
1.224 brouard 2313: #ifdef LINMINORIGINAL
2314: #else
1.317 brouard 2315: if(fb == fx){ /* Flat function in the direction */
2316: xmin=xx;
1.224 brouard 2317: *flat=1;
1.317 brouard 2318: }else{
1.224 brouard 2319: *flat=0;
2320: #endif
2321: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2322: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2323: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2324: /* fmin = f(p[j] + xmin * xi[j]) */
2325: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2326: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2327: #ifdef DEBUG
1.224 brouard 2328: 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);
2329: 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);
2330: #endif
2331: #ifdef LINMINORIGINAL
2332: #else
2333: }
1.126 brouard 2334: #endif
1.191 brouard 2335: #ifdef DEBUGLINMIN
2336: printf("linmin end ");
1.202 brouard 2337: fprintf(ficlog,"linmin end ");
1.191 brouard 2338: #endif
1.126 brouard 2339: for (j=1;j<=n;j++) {
1.203 brouard 2340: #ifdef LINMINORIGINAL
2341: xi[j] *= xmin;
2342: #else
2343: #ifdef DEBUGLINMIN
2344: if(xxs <1.0)
2345: printf(" before xi[%d]=%12.8f", j,xi[j]);
2346: #endif
2347: 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) */
2348: #ifdef DEBUGLINMIN
2349: if(xxs <1.0)
2350: 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 );
2351: #endif
2352: #endif
1.187 brouard 2353: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2354: }
1.191 brouard 2355: #ifdef DEBUGLINMIN
1.203 brouard 2356: printf("\n");
1.191 brouard 2357: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2358: 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 2359: for (j=1;j<=n;j++) {
1.202 brouard 2360: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2361: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2362: if(j % ncovmodel == 0){
1.191 brouard 2363: printf("\n");
1.202 brouard 2364: fprintf(ficlog,"\n");
2365: }
1.191 brouard 2366: }
1.203 brouard 2367: #else
1.191 brouard 2368: #endif
1.126 brouard 2369: free_vector(xicom,1,n);
2370: free_vector(pcom,1,n);
2371: }
2372:
2373:
2374: /*************** powell ************************/
1.162 brouard 2375: /*
1.317 brouard 2376: Minimization of a function func of n variables. Input consists in an initial starting point
2377: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2378: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2379: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2380: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2381: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2382: */
1.224 brouard 2383: #ifdef LINMINORIGINAL
2384: #else
2385: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2386: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2387: #endif
1.126 brouard 2388: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2389: double (*func)(double []))
2390: {
1.224 brouard 2391: #ifdef LINMINORIGINAL
2392: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2393: double (*func)(double []));
1.224 brouard 2394: #else
1.241 brouard 2395: void linmin(double p[], double xi[], int n, double *fret,
2396: double (*func)(double []),int *flat);
1.224 brouard 2397: #endif
1.239 brouard 2398: int i,ibig,j,jk,k;
1.126 brouard 2399: double del,t,*pt,*ptt,*xit;
1.181 brouard 2400: double directest;
1.126 brouard 2401: double fp,fptt;
2402: double *xits;
2403: int niterf, itmp;
2404:
2405: pt=vector(1,n);
2406: ptt=vector(1,n);
2407: xit=vector(1,n);
2408: xits=vector(1,n);
2409: *fret=(*func)(p);
2410: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2411: rcurr_time = time(NULL);
1.126 brouard 2412: for (*iter=1;;++(*iter)) {
1.187 brouard 2413: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2414: ibig=0;
2415: del=0.0;
1.157 brouard 2416: rlast_time=rcurr_time;
2417: /* (void) gettimeofday(&curr_time,&tzp); */
2418: rcurr_time = time(NULL);
2419: curr_time = *localtime(&rcurr_time);
2420: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2421: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2422: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2423: for (i=1;i<=n;i++) {
1.126 brouard 2424: fprintf(ficrespow," %.12lf", p[i]);
2425: }
1.239 brouard 2426: fprintf(ficrespow,"\n");fflush(ficrespow);
2427: printf("\n#model= 1 + age ");
2428: fprintf(ficlog,"\n#model= 1 + age ");
2429: if(nagesqr==1){
1.241 brouard 2430: printf(" + age*age ");
2431: fprintf(ficlog," + age*age ");
1.239 brouard 2432: }
2433: for(j=1;j <=ncovmodel-2;j++){
2434: if(Typevar[j]==0) {
2435: printf(" + V%d ",Tvar[j]);
2436: fprintf(ficlog," + V%d ",Tvar[j]);
2437: }else if(Typevar[j]==1) {
2438: printf(" + V%d*age ",Tvar[j]);
2439: fprintf(ficlog," + V%d*age ",Tvar[j]);
2440: }else if(Typevar[j]==2) {
2441: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2442: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2443: }
2444: }
1.126 brouard 2445: printf("\n");
1.239 brouard 2446: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2447: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2448: fprintf(ficlog,"\n");
1.239 brouard 2449: for(i=1,jk=1; i <=nlstate; i++){
2450: for(k=1; k <=(nlstate+ndeath); k++){
2451: if (k != i) {
2452: printf("%d%d ",i,k);
2453: fprintf(ficlog,"%d%d ",i,k);
2454: for(j=1; j <=ncovmodel; j++){
2455: printf("%12.7f ",p[jk]);
2456: fprintf(ficlog,"%12.7f ",p[jk]);
2457: jk++;
2458: }
2459: printf("\n");
2460: fprintf(ficlog,"\n");
2461: }
2462: }
2463: }
1.241 brouard 2464: if(*iter <=3 && *iter >1){
1.157 brouard 2465: tml = *localtime(&rcurr_time);
2466: strcpy(strcurr,asctime(&tml));
2467: rforecast_time=rcurr_time;
1.126 brouard 2468: itmp = strlen(strcurr);
2469: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2470: strcurr[itmp-1]='\0';
1.162 brouard 2471: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2472: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2473: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2474: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2475: forecast_time = *localtime(&rforecast_time);
2476: strcpy(strfor,asctime(&forecast_time));
2477: itmp = strlen(strfor);
2478: if(strfor[itmp-1]=='\n')
2479: strfor[itmp-1]='\0';
2480: 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);
2481: 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 2482: }
2483: }
1.187 brouard 2484: for (i=1;i<=n;i++) { /* For each direction i */
2485: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2486: fptt=(*fret);
2487: #ifdef DEBUG
1.203 brouard 2488: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2489: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2490: #endif
1.203 brouard 2491: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2492: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2493: #ifdef LINMINORIGINAL
1.188 brouard 2494: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2495: #else
2496: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2497: flatdir[i]=flat; /* Function is vanishing in that direction i */
2498: #endif
2499: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2500: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2501: /* because that direction will be replaced unless the gain del is small */
2502: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2503: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2504: /* with the new direction. */
2505: del=fabs(fptt-(*fret));
2506: ibig=i;
1.126 brouard 2507: }
2508: #ifdef DEBUG
2509: printf("%d %.12e",i,(*fret));
2510: fprintf(ficlog,"%d %.12e",i,(*fret));
2511: for (j=1;j<=n;j++) {
1.224 brouard 2512: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2513: printf(" x(%d)=%.12e",j,xit[j]);
2514: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2515: }
2516: for(j=1;j<=n;j++) {
1.225 brouard 2517: printf(" p(%d)=%.12e",j,p[j]);
2518: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2519: }
2520: printf("\n");
2521: fprintf(ficlog,"\n");
2522: #endif
1.187 brouard 2523: } /* end loop on each direction i */
2524: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2525: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2526: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2527: for(j=1;j<=n;j++) {
2528: if(flatdir[j] >0){
2529: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2530: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2531: }
1.319 brouard 2532: /* printf("\n"); */
2533: /* fprintf(ficlog,"\n"); */
2534: }
1.243 brouard 2535: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2536: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2537: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2538: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2539: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2540: /* decreased of more than 3.84 */
2541: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2542: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2543: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2544:
1.188 brouard 2545: /* Starting the program with initial values given by a former maximization will simply change */
2546: /* the scales of the directions and the directions, because the are reset to canonical directions */
2547: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2548: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2549: #ifdef DEBUG
2550: int k[2],l;
2551: k[0]=1;
2552: k[1]=-1;
2553: printf("Max: %.12e",(*func)(p));
2554: fprintf(ficlog,"Max: %.12e",(*func)(p));
2555: for (j=1;j<=n;j++) {
2556: printf(" %.12e",p[j]);
2557: fprintf(ficlog," %.12e",p[j]);
2558: }
2559: printf("\n");
2560: fprintf(ficlog,"\n");
2561: for(l=0;l<=1;l++) {
2562: for (j=1;j<=n;j++) {
2563: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2564: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2565: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2566: }
2567: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2568: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2569: }
2570: #endif
2571:
2572: free_vector(xit,1,n);
2573: free_vector(xits,1,n);
2574: free_vector(ptt,1,n);
2575: free_vector(pt,1,n);
2576: return;
1.192 brouard 2577: } /* enough precision */
1.240 brouard 2578: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2579: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2580: ptt[j]=2.0*p[j]-pt[j];
2581: xit[j]=p[j]-pt[j];
2582: pt[j]=p[j];
2583: }
1.181 brouard 2584: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2585: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2586: if (*iter <=4) {
1.225 brouard 2587: #else
2588: #endif
1.224 brouard 2589: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2590: #else
1.161 brouard 2591: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2592: #endif
1.162 brouard 2593: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2594: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2595: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2596: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2597: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2598: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2599: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2600: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2601: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2602: /* Even if f3 <f1, directest can be negative and t >0 */
2603: /* mu² and del² are equal when f3=f1 */
2604: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2605: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2606: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2607: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2608: #ifdef NRCORIGINAL
2609: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2610: #else
2611: 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 2612: t= t- del*SQR(fp-fptt);
1.183 brouard 2613: #endif
1.202 brouard 2614: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2615: #ifdef DEBUG
1.181 brouard 2616: 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);
2617: 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 2618: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2619: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2620: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2621: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2622: 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);
2623: 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);
2624: #endif
1.183 brouard 2625: #ifdef POWELLORIGINAL
2626: if (t < 0.0) { /* Then we use it for new direction */
2627: #else
1.182 brouard 2628: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2629: 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 2630: 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 2631: 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 2632: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2633: }
1.181 brouard 2634: if (directest < 0.0) { /* Then we use it for new direction */
2635: #endif
1.191 brouard 2636: #ifdef DEBUGLINMIN
1.234 brouard 2637: printf("Before linmin in direction P%d-P0\n",n);
2638: for (j=1;j<=n;j++) {
2639: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2640: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2641: if(j % ncovmodel == 0){
2642: printf("\n");
2643: fprintf(ficlog,"\n");
2644: }
2645: }
1.224 brouard 2646: #endif
2647: #ifdef LINMINORIGINAL
1.234 brouard 2648: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2649: #else
1.234 brouard 2650: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2651: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2652: #endif
1.234 brouard 2653:
1.191 brouard 2654: #ifdef DEBUGLINMIN
1.234 brouard 2655: for (j=1;j<=n;j++) {
2656: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2657: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2658: if(j % ncovmodel == 0){
2659: printf("\n");
2660: fprintf(ficlog,"\n");
2661: }
2662: }
1.224 brouard 2663: #endif
1.234 brouard 2664: for (j=1;j<=n;j++) {
2665: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2666: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2667: }
1.224 brouard 2668: #ifdef LINMINORIGINAL
2669: #else
1.234 brouard 2670: for (j=1, flatd=0;j<=n;j++) {
2671: if(flatdir[j]>0)
2672: flatd++;
2673: }
2674: if(flatd >0){
1.255 brouard 2675: printf("%d flat directions: ",flatd);
2676: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2677: for (j=1;j<=n;j++) {
2678: if(flatdir[j]>0){
2679: printf("%d ",j);
2680: fprintf(ficlog,"%d ",j);
2681: }
2682: }
2683: printf("\n");
2684: fprintf(ficlog,"\n");
1.319 brouard 2685: #ifdef FLATSUP
2686: free_vector(xit,1,n);
2687: free_vector(xits,1,n);
2688: free_vector(ptt,1,n);
2689: free_vector(pt,1,n);
2690: return;
2691: #endif
1.234 brouard 2692: }
1.191 brouard 2693: #endif
1.234 brouard 2694: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2695: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2696:
1.126 brouard 2697: #ifdef DEBUG
1.234 brouard 2698: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2699: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2700: for(j=1;j<=n;j++){
2701: printf(" %lf",xit[j]);
2702: fprintf(ficlog," %lf",xit[j]);
2703: }
2704: printf("\n");
2705: fprintf(ficlog,"\n");
1.126 brouard 2706: #endif
1.192 brouard 2707: } /* end of t or directest negative */
1.224 brouard 2708: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2709: #else
1.234 brouard 2710: } /* end if (fptt < fp) */
1.192 brouard 2711: #endif
1.225 brouard 2712: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2713: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2714: #else
1.224 brouard 2715: #endif
1.234 brouard 2716: } /* loop iteration */
1.126 brouard 2717: }
1.234 brouard 2718:
1.126 brouard 2719: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2720:
1.235 brouard 2721: 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 2722: {
1.279 brouard 2723: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2724: * (and selected quantitative values in nres)
2725: * by left multiplying the unit
2726: * matrix by transitions matrix until convergence is reached with precision ftolpl
2727: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2728: * Wx is row vector: population in state 1, population in state 2, population dead
2729: * or prevalence in state 1, prevalence in state 2, 0
2730: * newm is the matrix after multiplications, its rows are identical at a factor.
2731: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2732: * Output is prlim.
2733: * Initial matrix pimij
2734: */
1.206 brouard 2735: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2736: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2737: /* 0, 0 , 1} */
2738: /*
2739: * and after some iteration: */
2740: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2741: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2742: /* 0, 0 , 1} */
2743: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2744: /* {0.51571254859325999, 0.4842874514067399, */
2745: /* 0.51326036147820708, 0.48673963852179264} */
2746: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2747:
1.126 brouard 2748: int i, ii,j,k;
1.209 brouard 2749: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2750: /* double **matprod2(); */ /* test */
1.218 brouard 2751: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2752: double **newm;
1.209 brouard 2753: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2754: int ncvloop=0;
1.288 brouard 2755: int first=0;
1.169 brouard 2756:
1.209 brouard 2757: min=vector(1,nlstate);
2758: max=vector(1,nlstate);
2759: meandiff=vector(1,nlstate);
2760:
1.218 brouard 2761: /* Starting with matrix unity */
1.126 brouard 2762: for (ii=1;ii<=nlstate+ndeath;ii++)
2763: for (j=1;j<=nlstate+ndeath;j++){
2764: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2765: }
1.169 brouard 2766:
2767: cov[1]=1.;
2768:
2769: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2770: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2771: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2772: ncvloop++;
1.126 brouard 2773: newm=savm;
2774: /* Covariates have to be included here again */
1.138 brouard 2775: cov[2]=agefin;
1.319 brouard 2776: if(nagesqr==1){
2777: cov[3]= agefin*agefin;
2778: }
1.234 brouard 2779: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2780: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2781: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.319 brouard 2782: /* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; */
1.235 brouard 2783: /* 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)); */
1.234 brouard 2784: }
2785: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2786: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.319 brouard 2787: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2788: /* cov[++k1]=Tqresult[nres][k]; */
1.235 brouard 2789: /* 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]); */
1.138 brouard 2790: }
1.237 brouard 2791: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.319 brouard 2792: if(Dummy[Tage[k]]==2){ /* dummy with age */
1.234 brouard 2793: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 2794: /* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
2795: } else if(Dummy[Tage[k]]==3){ /* quantitative with age */
2796: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2797: /* cov[++k1]=Tqresult[nres][k]; */
1.234 brouard 2798: }
1.235 brouard 2799: /* 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]); */
1.234 brouard 2800: }
1.237 brouard 2801: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2802: /* 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]); */
1.237 brouard 2803: if(Dummy[Tvard[k][1]==0]){
2804: if(Dummy[Tvard[k][2]==0]){
2805: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
1.319 brouard 2806: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.237 brouard 2807: }else{
2808: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
1.319 brouard 2809: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; */
1.237 brouard 2810: }
2811: }else{
2812: if(Dummy[Tvard[k][2]==0]){
2813: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
1.319 brouard 2814: /* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; */
1.237 brouard 2815: }else{
2816: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
1.319 brouard 2817: /* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
1.237 brouard 2818: }
2819: }
1.234 brouard 2820: }
1.138 brouard 2821: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2822: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2823: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2824: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2825: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 2826: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2827: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2828:
1.126 brouard 2829: savm=oldm;
2830: oldm=newm;
1.209 brouard 2831:
2832: for(j=1; j<=nlstate; j++){
2833: max[j]=0.;
2834: min[j]=1.;
2835: }
2836: for(i=1;i<=nlstate;i++){
2837: sumnew=0;
2838: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2839: for(j=1; j<=nlstate; j++){
2840: prlim[i][j]= newm[i][j]/(1-sumnew);
2841: max[j]=FMAX(max[j],prlim[i][j]);
2842: min[j]=FMIN(min[j],prlim[i][j]);
2843: }
2844: }
2845:
1.126 brouard 2846: maxmax=0.;
1.209 brouard 2847: for(j=1; j<=nlstate; j++){
2848: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2849: maxmax=FMAX(maxmax,meandiff[j]);
2850: /* 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 2851: } /* j loop */
1.203 brouard 2852: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2853: /* 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 2854: if(maxmax < ftolpl){
1.209 brouard 2855: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2856: free_vector(min,1,nlstate);
2857: free_vector(max,1,nlstate);
2858: free_vector(meandiff,1,nlstate);
1.126 brouard 2859: return prlim;
2860: }
1.288 brouard 2861: } /* agefin loop */
1.208 brouard 2862: /* After some age loop it doesn't converge */
1.288 brouard 2863: if(!first){
2864: first=1;
2865: 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 2866: 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);
2867: }else if (first >=1 && first <10){
2868: 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);
2869: first++;
2870: }else if (first ==10){
2871: 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);
2872: 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");
2873: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
2874: first++;
1.288 brouard 2875: }
2876:
1.209 brouard 2877: /* 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); */
2878: free_vector(min,1,nlstate);
2879: free_vector(max,1,nlstate);
2880: free_vector(meandiff,1,nlstate);
1.208 brouard 2881:
1.169 brouard 2882: return prlim; /* should not reach here */
1.126 brouard 2883: }
2884:
1.217 brouard 2885:
2886: /**** Back Prevalence limit (stable or period prevalence) ****************/
2887:
1.218 brouard 2888: /* 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) */
2889: /* 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 2890: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2891: {
1.264 brouard 2892: /* 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 2893: matrix by transitions matrix until convergence is reached with precision ftolpl */
2894: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2895: /* Wx is row vector: population in state 1, population in state 2, population dead */
2896: /* or prevalence in state 1, prevalence in state 2, 0 */
2897: /* newm is the matrix after multiplications, its rows are identical at a factor */
2898: /* Initial matrix pimij */
2899: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2900: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2901: /* 0, 0 , 1} */
2902: /*
2903: * and after some iteration: */
2904: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2905: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2906: /* 0, 0 , 1} */
2907: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2908: /* {0.51571254859325999, 0.4842874514067399, */
2909: /* 0.51326036147820708, 0.48673963852179264} */
2910: /* If we start from prlim again, prlim tends to a constant matrix */
2911:
2912: int i, ii,j,k;
1.247 brouard 2913: int first=0;
1.217 brouard 2914: double *min, *max, *meandiff, maxmax,sumnew=0.;
2915: /* double **matprod2(); */ /* test */
2916: double **out, cov[NCOVMAX+1], **bmij();
2917: double **newm;
1.218 brouard 2918: double **dnewm, **doldm, **dsavm; /* for use */
2919: double **oldm, **savm; /* for use */
2920:
1.217 brouard 2921: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2922: int ncvloop=0;
2923:
2924: min=vector(1,nlstate);
2925: max=vector(1,nlstate);
2926: meandiff=vector(1,nlstate);
2927:
1.266 brouard 2928: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2929: oldm=oldms; savm=savms;
2930:
2931: /* Starting with matrix unity */
2932: for (ii=1;ii<=nlstate+ndeath;ii++)
2933: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2934: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2935: }
2936:
2937: cov[1]=1.;
2938:
2939: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2940: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2941: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2942: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2943: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2944: ncvloop++;
1.218 brouard 2945: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2946: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2947: /* Covariates have to be included here again */
2948: cov[2]=agefin;
1.319 brouard 2949: if(nagesqr==1){
1.217 brouard 2950: cov[3]= agefin*agefin;;
1.319 brouard 2951: }
1.242 brouard 2952: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2953: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2954: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2955: /* 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)); */
1.242 brouard 2956: }
2957: /* for (k=1; k<=cptcovn;k++) { */
2958: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2959: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2960: /* /\* 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])]); *\/ */
2961: /* } */
2962: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2963: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2964: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2965: /* 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]); */
2966: }
2967: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2968: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2969: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2970: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2971: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.319 brouard 2972: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age *\/ ERROR ???*/
2973: if(Dummy[Tage[k]]== 2){ /* dummy with age */
1.242 brouard 2974: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 2975: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
2976: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.242 brouard 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,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2985: }else{
2986: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2987: }
2988: }else{
2989: if(Dummy[Tvard[k][2]==0]){
2990: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2991: }else{
2992: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2993: }
2994: }
1.217 brouard 2995: }
2996:
2997: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2998: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2999: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3000: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3001: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3002: /* ij should be linked to the correct index of cov */
3003: /* age and covariate values ij are in 'cov', but we need to pass
3004: * ij for the observed prevalence at age and status and covariate
3005: * number: prevacurrent[(int)agefin][ii][ij]
3006: */
3007: /* 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 *\/ */
3008: /* 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 *\/ */
3009: 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 3010: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3011: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3012: /* for(i=1; i<=nlstate+ndeath; i++) { */
3013: /* printf("%d newm= ",i); */
3014: /* for(j=1;j<=nlstate+ndeath;j++) { */
3015: /* printf("%f ",newm[i][j]); */
3016: /* } */
3017: /* printf("oldm * "); */
3018: /* for(j=1;j<=nlstate+ndeath;j++) { */
3019: /* printf("%f ",oldm[i][j]); */
3020: /* } */
1.268 brouard 3021: /* printf(" bmmij "); */
1.266 brouard 3022: /* for(j=1;j<=nlstate+ndeath;j++) { */
3023: /* printf("%f ",pmmij[i][j]); */
3024: /* } */
3025: /* printf("\n"); */
3026: /* } */
3027: /* } */
1.217 brouard 3028: savm=oldm;
3029: oldm=newm;
1.266 brouard 3030:
1.217 brouard 3031: for(j=1; j<=nlstate; j++){
3032: max[j]=0.;
3033: min[j]=1.;
3034: }
3035: for(j=1; j<=nlstate; j++){
3036: for(i=1;i<=nlstate;i++){
1.234 brouard 3037: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3038: bprlim[i][j]= newm[i][j];
3039: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3040: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3041: }
3042: }
1.218 brouard 3043:
1.217 brouard 3044: maxmax=0.;
3045: for(i=1; i<=nlstate; i++){
1.318 brouard 3046: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3047: maxmax=FMAX(maxmax,meandiff[i]);
3048: /* 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 3049: } /* i loop */
1.217 brouard 3050: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3051: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3052: if(maxmax < ftolpl){
1.220 brouard 3053: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3054: free_vector(min,1,nlstate);
3055: free_vector(max,1,nlstate);
3056: free_vector(meandiff,1,nlstate);
3057: return bprlim;
3058: }
1.288 brouard 3059: } /* agefin loop */
1.217 brouard 3060: /* After some age loop it doesn't converge */
1.288 brouard 3061: if(!first){
1.247 brouard 3062: first=1;
3063: 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\
3064: 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);
3065: }
3066: 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 3067: 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);
3068: /* Try to lower 'ftol', for example from 1.e-8 to 6.e-9.\n", ftolpl, (int)age, (int)delaymax, (int)agefin, ncvloop, (int)age-(int)agefin); */
3069: free_vector(min,1,nlstate);
3070: free_vector(max,1,nlstate);
3071: free_vector(meandiff,1,nlstate);
3072:
3073: return bprlim; /* should not reach here */
3074: }
3075:
1.126 brouard 3076: /*************** transition probabilities ***************/
3077:
3078: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3079: {
1.138 brouard 3080: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3081: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3082: model to the ncovmodel covariates (including constant and age).
3083: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3084: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3085: ncth covariate in the global vector x is given by the formula:
3086: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3087: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3088: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3089: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3090: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3091: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3092: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3093: */
3094: double s1, lnpijopii;
1.126 brouard 3095: /*double t34;*/
1.164 brouard 3096: int i,j, nc, ii, jj;
1.126 brouard 3097:
1.223 brouard 3098: for(i=1; i<= nlstate; i++){
3099: for(j=1; j<i;j++){
3100: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3101: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3102: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3103: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3104: }
3105: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3106: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3107: }
3108: for(j=i+1; j<=nlstate+ndeath;j++){
3109: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3110: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3111: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3112: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3113: }
3114: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3115: }
3116: }
1.218 brouard 3117:
1.223 brouard 3118: for(i=1; i<= nlstate; i++){
3119: s1=0;
3120: for(j=1; j<i; j++){
3121: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3122: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3123: }
3124: for(j=i+1; j<=nlstate+ndeath; j++){
3125: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3126: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3127: }
3128: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3129: ps[i][i]=1./(s1+1.);
3130: /* Computing other pijs */
3131: for(j=1; j<i; j++)
3132: ps[i][j]= exp(ps[i][j])*ps[i][i];
3133: for(j=i+1; j<=nlstate+ndeath; j++)
3134: ps[i][j]= exp(ps[i][j])*ps[i][i];
3135: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3136: } /* end i */
1.218 brouard 3137:
1.223 brouard 3138: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3139: for(jj=1; jj<= nlstate+ndeath; jj++){
3140: ps[ii][jj]=0;
3141: ps[ii][ii]=1;
3142: }
3143: }
1.294 brouard 3144:
3145:
1.223 brouard 3146: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3147: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3148: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3149: /* } */
3150: /* printf("\n "); */
3151: /* } */
3152: /* printf("\n ");printf("%lf ",cov[2]);*/
3153: /*
3154: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3155: goto end;*/
1.266 brouard 3156: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3157: }
3158:
1.218 brouard 3159: /*************** backward transition probabilities ***************/
3160:
3161: /* 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 ) */
3162: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3163: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3164: {
1.302 brouard 3165: /* 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 3166: * 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 3167: */
1.218 brouard 3168: int i, ii, j,k;
1.222 brouard 3169:
3170: double **out, **pmij();
3171: double sumnew=0.;
1.218 brouard 3172: double agefin;
1.292 brouard 3173: 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 3174: double **dnewm, **dsavm, **doldm;
3175: double **bbmij;
3176:
1.218 brouard 3177: doldm=ddoldms; /* global pointers */
1.222 brouard 3178: dnewm=ddnewms;
3179: dsavm=ddsavms;
1.318 brouard 3180:
3181: /* Debug */
3182: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3183: agefin=cov[2];
1.268 brouard 3184: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3185: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3186: the observed prevalence (with this covariate ij) at beginning of transition */
3187: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3188:
3189: /* P_x */
1.266 brouard 3190: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 3191: /* outputs pmmij which is a stochastic matrix in row */
3192:
3193: /* Diag(w_x) */
1.292 brouard 3194: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3195: sumnew=0.;
1.269 brouard 3196: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3197: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3198: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3199: sumnew+=prevacurrent[(int)agefin][ii][ij];
3200: }
3201: if(sumnew >0.01){ /* At least some value in the prevalence */
3202: for (ii=1;ii<=nlstate+ndeath;ii++){
3203: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3204: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3205: }
3206: }else{
3207: for (ii=1;ii<=nlstate+ndeath;ii++){
3208: for (j=1;j<=nlstate+ndeath;j++)
3209: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3210: }
3211: /* if(sumnew <0.9){ */
3212: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3213: /* } */
3214: }
3215: k3=0.0; /* We put the last diagonal to 0 */
3216: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3217: doldm[ii][ii]= k3;
3218: }
3219: /* End doldm, At the end doldm is diag[(w_i)] */
3220:
1.292 brouard 3221: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3222: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3223:
1.292 brouard 3224: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3225: /* 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 3226: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3227: sumnew=0.;
1.222 brouard 3228: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3229: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3230: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3231: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3232: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3233: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3234: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3235: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3236: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3237: /* }else */
1.268 brouard 3238: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3239: } /*End ii */
3240: } /* 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 */
3241:
1.292 brouard 3242: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3243: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3244: /* end bmij */
1.266 brouard 3245: return ps; /*pointer is unchanged */
1.218 brouard 3246: }
1.217 brouard 3247: /*************** transition probabilities ***************/
3248:
1.218 brouard 3249: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3250: {
3251: /* According to parameters values stored in x and the covariate's values stored in cov,
3252: computes the probability to be observed in state j being in state i by appying the
3253: model to the ncovmodel covariates (including constant and age).
3254: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3255: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3256: ncth covariate in the global vector x is given by the formula:
3257: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3258: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3259: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3260: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3261: Outputs ps[i][j] the probability to be observed in j being in j according to
3262: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3263: */
3264: double s1, lnpijopii;
3265: /*double t34;*/
3266: int i,j, nc, ii, jj;
3267:
1.234 brouard 3268: for(i=1; i<= nlstate; i++){
3269: for(j=1; j<i;j++){
3270: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3271: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3272: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3273: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3274: }
3275: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3276: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3277: }
3278: for(j=i+1; j<=nlstate+ndeath;j++){
3279: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3280: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3281: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3282: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3283: }
3284: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3285: }
3286: }
3287:
3288: for(i=1; i<= nlstate; i++){
3289: s1=0;
3290: for(j=1; j<i; j++){
3291: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3292: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3293: }
3294: for(j=i+1; j<=nlstate+ndeath; j++){
3295: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3296: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3297: }
3298: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3299: ps[i][i]=1./(s1+1.);
3300: /* Computing other pijs */
3301: for(j=1; j<i; j++)
3302: ps[i][j]= exp(ps[i][j])*ps[i][i];
3303: for(j=i+1; j<=nlstate+ndeath; j++)
3304: ps[i][j]= exp(ps[i][j])*ps[i][i];
3305: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3306: } /* end i */
3307:
3308: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3309: for(jj=1; jj<= nlstate+ndeath; jj++){
3310: ps[ii][jj]=0;
3311: ps[ii][ii]=1;
3312: }
3313: }
1.296 brouard 3314: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3315: for(jj=1; jj<= nlstate+ndeath; jj++){
3316: s1=0.;
3317: for(ii=1; ii<= nlstate+ndeath; ii++){
3318: s1+=ps[ii][jj];
3319: }
3320: for(ii=1; ii<= nlstate; ii++){
3321: ps[ii][jj]=ps[ii][jj]/s1;
3322: }
3323: }
3324: /* Transposition */
3325: for(jj=1; jj<= nlstate+ndeath; jj++){
3326: for(ii=jj; ii<= nlstate+ndeath; ii++){
3327: s1=ps[ii][jj];
3328: ps[ii][jj]=ps[jj][ii];
3329: ps[jj][ii]=s1;
3330: }
3331: }
3332: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3333: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3334: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3335: /* } */
3336: /* printf("\n "); */
3337: /* } */
3338: /* printf("\n ");printf("%lf ",cov[2]);*/
3339: /*
3340: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3341: goto end;*/
3342: return ps;
1.217 brouard 3343: }
3344:
3345:
1.126 brouard 3346: /**************** Product of 2 matrices ******************/
3347:
1.145 brouard 3348: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3349: {
3350: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3351: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3352: /* in, b, out are matrice of pointers which should have been initialized
3353: before: only the contents of out is modified. The function returns
3354: a pointer to pointers identical to out */
1.145 brouard 3355: int i, j, k;
1.126 brouard 3356: for(i=nrl; i<= nrh; i++)
1.145 brouard 3357: for(k=ncolol; k<=ncoloh; k++){
3358: out[i][k]=0.;
3359: for(j=ncl; j<=nch; j++)
3360: out[i][k] +=in[i][j]*b[j][k];
3361: }
1.126 brouard 3362: return out;
3363: }
3364:
3365:
3366: /************* Higher Matrix Product ***************/
3367:
1.235 brouard 3368: 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 3369: {
1.218 brouard 3370: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3371: 'nhstepm*hstepm*stepm' months (i.e. until
3372: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3373: nhstepm*hstepm matrices.
3374: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3375: (typically every 2 years instead of every month which is too big
3376: for the memory).
3377: Model is determined by parameters x and covariates have to be
3378: included manually here.
3379:
3380: */
3381:
3382: int i, j, d, h, k;
1.131 brouard 3383: double **out, cov[NCOVMAX+1];
1.126 brouard 3384: double **newm;
1.187 brouard 3385: double agexact;
1.214 brouard 3386: double agebegin, ageend;
1.126 brouard 3387:
3388: /* Hstepm could be zero and should return the unit matrix */
3389: for (i=1;i<=nlstate+ndeath;i++)
3390: for (j=1;j<=nlstate+ndeath;j++){
3391: oldm[i][j]=(i==j ? 1.0 : 0.0);
3392: po[i][j][0]=(i==j ? 1.0 : 0.0);
3393: }
3394: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3395: for(h=1; h <=nhstepm; h++){
3396: for(d=1; d <=hstepm; d++){
3397: newm=savm;
3398: /* Covariates have to be included here again */
3399: cov[1]=1.;
1.214 brouard 3400: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3401: cov[2]=agexact;
1.319 brouard 3402: if(nagesqr==1){
1.227 brouard 3403: cov[3]= agexact*agexact;
1.319 brouard 3404: }
1.235 brouard 3405: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
1.319 brouard 3406: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3407: /* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 */
3408: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3409: /* k 1 2 3 4 5 6 7 8 9 */
3410: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
3411: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
3412: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
3413: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1.235 brouard 3414: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3415: /* printf("hpxij 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)); */
3416: }
3417: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3418: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.319 brouard 3419: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
1.235 brouard 3420: /* 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]); */
3421: }
1.319 brouard 3422: for (k=1; k<=cptcovage;k++){ /* For product with age V1+V1*age +V4 +age*V3 */
3423: /* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*/
3424: /* */
3425: if(Dummy[Tage[k]]== 2){ /* dummy with age */
3426: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age *\/ */
1.235 brouard 3427: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 3428: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
3429: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.235 brouard 3430: }
3431: /* 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]); */
3432: }
1.319 brouard 3433: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 3434: /* printf("hPxij 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]); */
1.319 brouard 3435: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
3436: if(Dummy[Tvard[k][1]==0]){
3437: if(Dummy[Tvard[k][2]==0]){
3438: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3439: }else{
3440: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
3441: }
3442: }else{
3443: if(Dummy[Tvard[k][2]==0]){
3444: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
3445: }else{
3446: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
3447: }
3448: }
1.235 brouard 3449: }
3450: /* for (k=1; k<=cptcovn;k++) */
3451: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3452: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3453: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3454: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3455: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3456:
3457:
1.126 brouard 3458: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3459: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3460: /* right multiplication of oldm by the current matrix */
1.126 brouard 3461: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3462: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3463: /* if((int)age == 70){ */
3464: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3465: /* for(i=1; i<=nlstate+ndeath; i++) { */
3466: /* printf("%d pmmij ",i); */
3467: /* for(j=1;j<=nlstate+ndeath;j++) { */
3468: /* printf("%f ",pmmij[i][j]); */
3469: /* } */
3470: /* printf(" oldm "); */
3471: /* for(j=1;j<=nlstate+ndeath;j++) { */
3472: /* printf("%f ",oldm[i][j]); */
3473: /* } */
3474: /* printf("\n"); */
3475: /* } */
3476: /* } */
1.126 brouard 3477: savm=oldm;
3478: oldm=newm;
3479: }
3480: for(i=1; i<=nlstate+ndeath; i++)
3481: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3482: po[i][j][h]=newm[i][j];
3483: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3484: }
1.128 brouard 3485: /*printf("h=%d ",h);*/
1.126 brouard 3486: } /* end h */
1.267 brouard 3487: /* printf("\n H=%d \n",h); */
1.126 brouard 3488: return po;
3489: }
3490:
1.217 brouard 3491: /************* Higher Back Matrix Product ***************/
1.218 brouard 3492: /* 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 3493: 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 3494: {
1.266 brouard 3495: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3496: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3497: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3498: nhstepm*hstepm matrices.
3499: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3500: (typically every 2 years instead of every month which is too big
1.217 brouard 3501: for the memory).
1.218 brouard 3502: Model is determined by parameters x and covariates have to be
1.266 brouard 3503: included manually here. Then we use a call to bmij(x and cov)
3504: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3505: */
1.217 brouard 3506:
3507: int i, j, d, h, k;
1.266 brouard 3508: double **out, cov[NCOVMAX+1], **bmij();
3509: double **newm, ***newmm;
1.217 brouard 3510: double agexact;
3511: double agebegin, ageend;
1.222 brouard 3512: double **oldm, **savm;
1.217 brouard 3513:
1.266 brouard 3514: newmm=po; /* To be saved */
3515: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3516: /* Hstepm could be zero and should return the unit matrix */
3517: for (i=1;i<=nlstate+ndeath;i++)
3518: for (j=1;j<=nlstate+ndeath;j++){
3519: oldm[i][j]=(i==j ? 1.0 : 0.0);
3520: po[i][j][0]=(i==j ? 1.0 : 0.0);
3521: }
3522: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3523: for(h=1; h <=nhstepm; h++){
3524: for(d=1; d <=hstepm; d++){
3525: newm=savm;
3526: /* Covariates have to be included here again */
3527: cov[1]=1.;
1.271 brouard 3528: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3529: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3530: /* Debug */
3531: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3532: cov[2]=agexact;
3533: if(nagesqr==1)
1.222 brouard 3534: cov[3]= agexact*agexact;
1.266 brouard 3535: for (k=1; k<=cptcovn;k++){
3536: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3537: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3538: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3539: /* 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)); */
3540: }
1.267 brouard 3541: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3542: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3543: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3544: /* 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]); */
3545: }
1.319 brouard 3546: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 *//* For product with age */
3547: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age error!!!*\/ */
3548: if(Dummy[Tage[k]]== 2){ /* dummy with age */
1.267 brouard 3549: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 3550: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
1.267 brouard 3551: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3552: }
3553: /* 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]); */
3554: }
3555: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3556: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3557: }
1.217 brouard 3558: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3559: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3560:
1.218 brouard 3561: /* Careful transposed matrix */
1.266 brouard 3562: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3563: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3564: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3565: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3566: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3567: /* if((int)age == 70){ */
3568: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3569: /* for(i=1; i<=nlstate+ndeath; i++) { */
3570: /* printf("%d pmmij ",i); */
3571: /* for(j=1;j<=nlstate+ndeath;j++) { */
3572: /* printf("%f ",pmmij[i][j]); */
3573: /* } */
3574: /* printf(" oldm "); */
3575: /* for(j=1;j<=nlstate+ndeath;j++) { */
3576: /* printf("%f ",oldm[i][j]); */
3577: /* } */
3578: /* printf("\n"); */
3579: /* } */
3580: /* } */
3581: savm=oldm;
3582: oldm=newm;
3583: }
3584: for(i=1; i<=nlstate+ndeath; i++)
3585: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3586: po[i][j][h]=newm[i][j];
1.268 brouard 3587: /* if(h==nhstepm) */
3588: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3589: }
1.268 brouard 3590: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3591: } /* end h */
1.268 brouard 3592: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3593: return po;
3594: }
3595:
3596:
1.162 brouard 3597: #ifdef NLOPT
3598: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3599: double fret;
3600: double *xt;
3601: int j;
3602: myfunc_data *d2 = (myfunc_data *) pd;
3603: /* xt = (p1-1); */
3604: xt=vector(1,n);
3605: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3606:
3607: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3608: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3609: printf("Function = %.12lf ",fret);
3610: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3611: printf("\n");
3612: free_vector(xt,1,n);
3613: return fret;
3614: }
3615: #endif
1.126 brouard 3616:
3617: /*************** log-likelihood *************/
3618: double func( double *x)
3619: {
1.226 brouard 3620: int i, ii, j, k, mi, d, kk;
3621: int ioffset=0;
3622: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3623: double **out;
3624: double lli; /* Individual log likelihood */
3625: int s1, s2;
1.228 brouard 3626: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
1.226 brouard 3627: double bbh, survp;
3628: long ipmx;
3629: double agexact;
3630: /*extern weight */
3631: /* We are differentiating ll according to initial status */
3632: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3633: /*for(i=1;i<imx;i++)
3634: printf(" %d\n",s[4][i]);
3635: */
1.162 brouard 3636:
1.226 brouard 3637: ++countcallfunc;
1.162 brouard 3638:
1.226 brouard 3639: cov[1]=1.;
1.126 brouard 3640:
1.226 brouard 3641: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3642: ioffset=0;
1.226 brouard 3643: if(mle==1){
3644: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3645: /* Computes the values of the ncovmodel covariates of the model
3646: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3647: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3648: to be observed in j being in i according to the model.
3649: */
1.243 brouard 3650: ioffset=2+nagesqr ;
1.233 brouard 3651: /* Fixed */
1.319 brouard 3652: for (k=1; k<=ncovf;k++){ /* For each fixed covariate dummu or quant or prod */
3653: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3654: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3655: /* 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 3656: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.319 brouard 3657: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V1 is fixed (TvarFind[1]=6)*/
3658: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3659: }
1.226 brouard 3660: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3661: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3662: has been calculated etc */
3663: /* For an individual i, wav[i] gives the number of effective waves */
3664: /* We compute the contribution to Likelihood of each effective transition
3665: mw[mi][i] is real wave of the mi th effectve wave */
3666: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3667: s2=s[mw[mi+1][i]][i];
3668: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3669: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3670: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3671: */
3672: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3673: 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*/
3674: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? */
1.242 brouard 3675: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3676: }
3677: for (ii=1;ii<=nlstate+ndeath;ii++)
3678: for (j=1;j<=nlstate+ndeath;j++){
3679: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3680: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3681: }
3682: for(d=0; d<dh[mi][i]; d++){
3683: newm=savm;
3684: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3685: cov[2]=agexact;
3686: if(nagesqr==1)
3687: cov[3]= agexact*agexact; /* Should be changed here */
3688: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 3689: if(!FixedV[Tvar[Tage[kk]]])
3690: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
3691: else
3692: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3693: }
3694: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3695: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3696: savm=oldm;
3697: oldm=newm;
3698: } /* end mult */
3699:
3700: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3701: /* But now since version 0.9 we anticipate for bias at large stepm.
3702: * If stepm is larger than one month (smallest stepm) and if the exact delay
3703: * (in months) between two waves is not a multiple of stepm, we rounded to
3704: * the nearest (and in case of equal distance, to the lowest) interval but now
3705: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3706: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3707: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3708: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3709: * -stepm/2 to stepm/2 .
3710: * For stepm=1 the results are the same as for previous versions of Imach.
3711: * For stepm > 1 the results are less biased than in previous versions.
3712: */
1.234 brouard 3713: s1=s[mw[mi][i]][i];
3714: s2=s[mw[mi+1][i]][i];
3715: bbh=(double)bh[mi][i]/(double)stepm;
3716: /* bias bh is positive if real duration
3717: * is higher than the multiple of stepm and negative otherwise.
3718: */
3719: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3720: if( s2 > nlstate){
3721: /* i.e. if s2 is a death state and if the date of death is known
3722: then the contribution to the likelihood is the probability to
3723: die between last step unit time and current step unit time,
3724: which is also equal to probability to die before dh
3725: minus probability to die before dh-stepm .
3726: In version up to 0.92 likelihood was computed
3727: as if date of death was unknown. Death was treated as any other
3728: health state: the date of the interview describes the actual state
3729: and not the date of a change in health state. The former idea was
3730: to consider that at each interview the state was recorded
3731: (healthy, disable or death) and IMaCh was corrected; but when we
3732: introduced the exact date of death then we should have modified
3733: the contribution of an exact death to the likelihood. This new
3734: contribution is smaller and very dependent of the step unit
3735: stepm. It is no more the probability to die between last interview
3736: and month of death but the probability to survive from last
3737: interview up to one month before death multiplied by the
3738: probability to die within a month. Thanks to Chris
3739: Jackson for correcting this bug. Former versions increased
3740: mortality artificially. The bad side is that we add another loop
3741: which slows down the processing. The difference can be up to 10%
3742: lower mortality.
3743: */
3744: /* If, at the beginning of the maximization mostly, the
3745: cumulative probability or probability to be dead is
3746: constant (ie = 1) over time d, the difference is equal to
3747: 0. out[s1][3] = savm[s1][3]: probability, being at state
3748: s1 at precedent wave, to be dead a month before current
3749: wave is equal to probability, being at state s1 at
3750: precedent wave, to be dead at mont of the current
3751: wave. Then the observed probability (that this person died)
3752: is null according to current estimated parameter. In fact,
3753: it should be very low but not zero otherwise the log go to
3754: infinity.
3755: */
1.183 brouard 3756: /* #ifdef INFINITYORIGINAL */
3757: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3758: /* #else */
3759: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3760: /* lli=log(mytinydouble); */
3761: /* else */
3762: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3763: /* #endif */
1.226 brouard 3764: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3765:
1.226 brouard 3766: } else if ( s2==-1 ) { /* alive */
3767: for (j=1,survp=0. ; j<=nlstate; j++)
3768: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3769: /*survp += out[s1][j]; */
3770: lli= log(survp);
3771: }
3772: else if (s2==-4) {
3773: for (j=3,survp=0. ; j<=nlstate; j++)
3774: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3775: lli= log(survp);
3776: }
3777: else if (s2==-5) {
3778: for (j=1,survp=0. ; j<=2; j++)
3779: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3780: lli= log(survp);
3781: }
3782: else{
3783: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3784: /* 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 */
3785: }
3786: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3787: /*if(lli ==000.0)*/
3788: /*printf("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); */
3789: ipmx +=1;
3790: sw += weight[i];
3791: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3792: /* if (lli < log(mytinydouble)){ */
3793: /* 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); */
3794: /* 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]); */
3795: /* } */
3796: } /* end of wave */
3797: } /* end of individual */
3798: } else if(mle==2){
3799: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 3800: ioffset=2+nagesqr ;
3801: for (k=1; k<=ncovf;k++)
3802: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 3803: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3804: for(k=1; k <= ncovv ; k++){
3805: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3806: }
1.226 brouard 3807: for (ii=1;ii<=nlstate+ndeath;ii++)
3808: for (j=1;j<=nlstate+ndeath;j++){
3809: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3810: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3811: }
3812: for(d=0; d<=dh[mi][i]; d++){
3813: newm=savm;
3814: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3815: cov[2]=agexact;
3816: if(nagesqr==1)
3817: cov[3]= agexact*agexact;
3818: for (kk=1; kk<=cptcovage;kk++) {
3819: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3820: }
3821: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3822: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3823: savm=oldm;
3824: oldm=newm;
3825: } /* end mult */
3826:
3827: s1=s[mw[mi][i]][i];
3828: s2=s[mw[mi+1][i]][i];
3829: bbh=(double)bh[mi][i]/(double)stepm;
3830: 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 */
3831: ipmx +=1;
3832: sw += weight[i];
3833: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3834: } /* end of wave */
3835: } /* end of individual */
3836: } else if(mle==3){ /* exponential inter-extrapolation */
3837: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3838: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3839: for(mi=1; mi<= wav[i]-1; mi++){
3840: for (ii=1;ii<=nlstate+ndeath;ii++)
3841: for (j=1;j<=nlstate+ndeath;j++){
3842: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3843: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3844: }
3845: for(d=0; d<dh[mi][i]; d++){
3846: newm=savm;
3847: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3848: cov[2]=agexact;
3849: if(nagesqr==1)
3850: cov[3]= agexact*agexact;
3851: for (kk=1; kk<=cptcovage;kk++) {
3852: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3853: }
3854: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3855: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3856: savm=oldm;
3857: oldm=newm;
3858: } /* end mult */
3859:
3860: s1=s[mw[mi][i]][i];
3861: s2=s[mw[mi+1][i]][i];
3862: bbh=(double)bh[mi][i]/(double)stepm;
3863: 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 */
3864: ipmx +=1;
3865: sw += weight[i];
3866: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3867: } /* end of wave */
3868: } /* end of individual */
3869: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3870: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3871: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3872: for(mi=1; mi<= wav[i]-1; mi++){
3873: for (ii=1;ii<=nlstate+ndeath;ii++)
3874: for (j=1;j<=nlstate+ndeath;j++){
3875: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3876: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3877: }
3878: for(d=0; d<dh[mi][i]; d++){
3879: newm=savm;
3880: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3881: cov[2]=agexact;
3882: if(nagesqr==1)
3883: cov[3]= agexact*agexact;
3884: for (kk=1; kk<=cptcovage;kk++) {
3885: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3886: }
1.126 brouard 3887:
1.226 brouard 3888: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3889: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3890: savm=oldm;
3891: oldm=newm;
3892: } /* end mult */
3893:
3894: s1=s[mw[mi][i]][i];
3895: s2=s[mw[mi+1][i]][i];
3896: if( s2 > nlstate){
3897: lli=log(out[s1][s2] - savm[s1][s2]);
3898: } else if ( s2==-1 ) { /* alive */
3899: for (j=1,survp=0. ; j<=nlstate; j++)
3900: survp += out[s1][j];
3901: lli= log(survp);
3902: }else{
3903: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3904: }
3905: ipmx +=1;
3906: sw += weight[i];
3907: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3908: /* 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 3909: } /* end of wave */
3910: } /* end of individual */
3911: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3912: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3913: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3914: for(mi=1; mi<= wav[i]-1; mi++){
3915: for (ii=1;ii<=nlstate+ndeath;ii++)
3916: for (j=1;j<=nlstate+ndeath;j++){
3917: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3918: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3919: }
3920: for(d=0; d<dh[mi][i]; d++){
3921: newm=savm;
3922: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3923: cov[2]=agexact;
3924: if(nagesqr==1)
3925: cov[3]= agexact*agexact;
3926: for (kk=1; kk<=cptcovage;kk++) {
3927: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3928: }
1.126 brouard 3929:
1.226 brouard 3930: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3931: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3932: savm=oldm;
3933: oldm=newm;
3934: } /* end mult */
3935:
3936: s1=s[mw[mi][i]][i];
3937: s2=s[mw[mi+1][i]][i];
3938: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3939: ipmx +=1;
3940: sw += weight[i];
3941: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3942: /*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]);*/
3943: } /* end of wave */
3944: } /* end of individual */
3945: } /* End of if */
3946: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3947: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3948: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3949: return -l;
1.126 brouard 3950: }
3951:
3952: /*************** log-likelihood *************/
3953: double funcone( double *x)
3954: {
1.228 brouard 3955: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3956: int i, ii, j, k, mi, d, kk;
1.228 brouard 3957: int ioffset=0;
1.131 brouard 3958: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3959: double **out;
3960: double lli; /* Individual log likelihood */
3961: double llt;
3962: int s1, s2;
1.228 brouard 3963: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3964:
1.126 brouard 3965: double bbh, survp;
1.187 brouard 3966: double agexact;
1.214 brouard 3967: double agebegin, ageend;
1.126 brouard 3968: /*extern weight */
3969: /* We are differentiating ll according to initial status */
3970: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3971: /*for(i=1;i<imx;i++)
3972: printf(" %d\n",s[4][i]);
3973: */
3974: cov[1]=1.;
3975:
3976: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3977: ioffset=0;
3978: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3979: /* ioffset=2+nagesqr+cptcovage; */
3980: ioffset=2+nagesqr;
1.232 brouard 3981: /* Fixed */
1.224 brouard 3982: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3983: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.311 brouard 3984: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
1.232 brouard 3985: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V1 is fixed (k=6)*/
3986: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3987: /* cov[2+6]=covar[Tvar[6]][i]; */
3988: /* cov[2+6]=covar[2][i]; V2 */
3989: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3990: /* cov[2+7]=covar[Tvar[7]][i]; */
3991: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3992: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3993: /* cov[2+9]=covar[Tvar[9]][i]; */
3994: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3995: }
1.232 brouard 3996: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3997: /* 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?)*\/ */
3998: /* } */
1.231 brouard 3999: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4000: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4001: /* } */
1.225 brouard 4002:
1.233 brouard 4003:
4004: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 4005: /* Wave varying (but not age varying) */
4006: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 4007: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
4008: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4009: }
1.232 brouard 4010: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 4011: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4012: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
4013: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
4014: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
4015: /* 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]); */
1.232 brouard 4016: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4017: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4018: /* /\* 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]); *\/ */
4019: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4020: /* } */
1.126 brouard 4021: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4022: for (j=1;j<=nlstate+ndeath;j++){
4023: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4024: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4025: }
1.214 brouard 4026:
4027: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4028: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4029: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4030: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4031: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4032: and mw[mi+1][i]. dh depends on stepm.*/
4033: newm=savm;
1.247 brouard 4034: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4035: cov[2]=agexact;
4036: if(nagesqr==1)
4037: cov[3]= agexact*agexact;
4038: for (kk=1; kk<=cptcovage;kk++) {
4039: if(!FixedV[Tvar[Tage[kk]]])
4040: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4041: else
4042: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
4043: }
4044: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4045: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4046: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4047: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4048: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4049: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4050: savm=oldm;
4051: oldm=newm;
1.126 brouard 4052: } /* end mult */
4053:
4054: s1=s[mw[mi][i]][i];
4055: s2=s[mw[mi+1][i]][i];
1.217 brouard 4056: /* if(s2==-1){ */
1.268 brouard 4057: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4058: /* /\* exit(1); *\/ */
4059: /* } */
1.126 brouard 4060: bbh=(double)bh[mi][i]/(double)stepm;
4061: /* bias is positive if real duration
4062: * is higher than the multiple of stepm and negative otherwise.
4063: */
4064: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4065: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4066: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4067: for (j=1,survp=0. ; j<=nlstate; j++)
4068: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4069: lli= log(survp);
1.126 brouard 4070: }else if (mle==1){
1.242 brouard 4071: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4072: } else if(mle==2){
1.242 brouard 4073: 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 4074: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4075: 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 4076: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4077: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4078: } else{ /* mle=0 back to 1 */
1.242 brouard 4079: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4080: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4081: } /* End of if */
4082: ipmx +=1;
4083: sw += weight[i];
4084: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 4085: /*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]); */
1.126 brouard 4086: if(globpr){
1.246 brouard 4087: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4088: %11.6f %11.6f %11.6f ", \
1.242 brouard 4089: 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 4090: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 4091: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4092: llt +=ll[k]*gipmx/gsw;
4093: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
4094: }
4095: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 4096: }
1.232 brouard 4097: } /* end of wave */
4098: } /* end of individual */
4099: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4100: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4101: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4102: if(globpr==0){ /* First time we count the contributions and weights */
4103: gipmx=ipmx;
4104: gsw=sw;
4105: }
4106: return -l;
1.126 brouard 4107: }
4108:
4109:
4110: /*************** function likelione ***********/
1.292 brouard 4111: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4112: {
4113: /* This routine should help understanding what is done with
4114: the selection of individuals/waves and
4115: to check the exact contribution to the likelihood.
4116: Plotting could be done.
4117: */
4118: int k;
4119:
4120: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4121: strcpy(fileresilk,"ILK_");
1.202 brouard 4122: strcat(fileresilk,fileresu);
1.126 brouard 4123: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4124: printf("Problem with resultfile: %s\n", fileresilk);
4125: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4126: }
1.214 brouard 4127: 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");
4128: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4129: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4130: for(k=1; k<=nlstate; k++)
4131: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4132: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4133: }
4134:
1.292 brouard 4135: *fretone=(*func)(p);
1.126 brouard 4136: if(*globpri !=0){
4137: fclose(ficresilk);
1.205 brouard 4138: if (mle ==0)
4139: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4140: else if(mle >=1)
4141: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4142: 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 4143: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4144:
4145: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4146: 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 4147: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4148: }
1.207 brouard 4149: 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 4150: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4151: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4152: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4153: fflush(fichtm);
1.205 brouard 4154: }
1.126 brouard 4155: return;
4156: }
4157:
4158:
4159: /*********** Maximum Likelihood Estimation ***************/
4160:
4161: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4162: {
1.319 brouard 4163: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4164: double **xi;
4165: double fret;
4166: double fretone; /* Only one call to likelihood */
4167: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4168:
4169: #ifdef NLOPT
4170: int creturn;
4171: nlopt_opt opt;
4172: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4173: double *lb;
4174: double minf; /* the minimum objective value, upon return */
4175: double * p1; /* Shifted parameters from 0 instead of 1 */
4176: myfunc_data dinst, *d = &dinst;
4177: #endif
4178:
4179:
1.126 brouard 4180: xi=matrix(1,npar,1,npar);
4181: for (i=1;i<=npar;i++)
4182: for (j=1;j<=npar;j++)
4183: xi[i][j]=(i==j ? 1.0 : 0.0);
4184: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4185: strcpy(filerespow,"POW_");
1.126 brouard 4186: strcat(filerespow,fileres);
4187: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4188: printf("Problem with resultfile: %s\n", filerespow);
4189: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4190: }
4191: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4192: for (i=1;i<=nlstate;i++)
4193: for(j=1;j<=nlstate+ndeath;j++)
4194: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4195: fprintf(ficrespow,"\n");
1.162 brouard 4196: #ifdef POWELL
1.319 brouard 4197: #ifdef LINMINORIGINAL
4198: #else /* LINMINORIGINAL */
4199:
4200: flatdir=ivector(1,npar);
4201: for (j=1;j<=npar;j++) flatdir[j]=0;
4202: #endif /*LINMINORIGINAL */
4203:
4204: #ifdef FLATSUP
4205: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4206: /* reorganizing p by suppressing flat directions */
4207: for(i=1, jk=1; i <=nlstate; i++){
4208: for(k=1; k <=(nlstate+ndeath); k++){
4209: if (k != i) {
4210: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4211: if(flatdir[jk]==1){
4212: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4213: }
4214: for(j=1; j <=ncovmodel; j++){
4215: printf("%12.7f ",p[jk]);
4216: jk++;
4217: }
4218: printf("\n");
4219: }
4220: }
4221: }
4222: /* skipping */
4223: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4224: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4225: for(k=1; k <=(nlstate+ndeath); k++){
4226: if (k != i) {
4227: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4228: if(flatdir[jk]==1){
4229: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4230: for(j=1; j <=ncovmodel; jk++,j++){
4231: printf(" p[%d]=%12.7f",jk, p[jk]);
4232: /*q[jjk]=p[jk];*/
4233: }
4234: }else{
4235: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4236: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4237: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4238: /*q[jjk]=p[jk];*/
4239: }
4240: }
4241: printf("\n");
4242: }
4243: fflush(stdout);
4244: }
4245: }
4246: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4247: #else /* FLATSUP */
1.126 brouard 4248: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4249: #endif /* FLATSUP */
4250:
4251: #ifdef LINMINORIGINAL
4252: #else
4253: free_ivector(flatdir,1,npar);
4254: #endif /* LINMINORIGINAL*/
4255: #endif /* POWELL */
1.126 brouard 4256:
1.162 brouard 4257: #ifdef NLOPT
4258: #ifdef NEWUOA
4259: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4260: #else
4261: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4262: #endif
4263: lb=vector(0,npar-1);
4264: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4265: nlopt_set_lower_bounds(opt, lb);
4266: nlopt_set_initial_step1(opt, 0.1);
4267:
4268: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4269: d->function = func;
4270: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4271: nlopt_set_min_objective(opt, myfunc, d);
4272: nlopt_set_xtol_rel(opt, ftol);
4273: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4274: printf("nlopt failed! %d\n",creturn);
4275: }
4276: else {
4277: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4278: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4279: iter=1; /* not equal */
4280: }
4281: nlopt_destroy(opt);
4282: #endif
1.319 brouard 4283: #ifdef FLATSUP
4284: /* npared = npar -flatd/ncovmodel; */
4285: /* xired= matrix(1,npared,1,npared); */
4286: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4287: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4288: /* free_matrix(xire,1,npared,1,npared); */
4289: #else /* FLATSUP */
4290: #endif /* FLATSUP */
1.126 brouard 4291: free_matrix(xi,1,npar,1,npar);
4292: fclose(ficrespow);
1.203 brouard 4293: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4294: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4295: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4296:
4297: }
4298:
4299: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4300: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4301: {
4302: double **a,**y,*x,pd;
1.203 brouard 4303: /* double **hess; */
1.164 brouard 4304: int i, j;
1.126 brouard 4305: int *indx;
4306:
4307: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4308: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4309: void lubksb(double **a, int npar, int *indx, double b[]) ;
4310: void ludcmp(double **a, int npar, int *indx, double *d) ;
4311: double gompertz(double p[]);
1.203 brouard 4312: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4313:
4314: printf("\nCalculation of the hessian matrix. Wait...\n");
4315: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4316: for (i=1;i<=npar;i++){
1.203 brouard 4317: printf("%d-",i);fflush(stdout);
4318: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4319:
4320: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4321:
4322: /* printf(" %f ",p[i]);
4323: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4324: }
4325:
4326: for (i=1;i<=npar;i++) {
4327: for (j=1;j<=npar;j++) {
4328: if (j>i) {
1.203 brouard 4329: printf(".%d-%d",i,j);fflush(stdout);
4330: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4331: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4332:
4333: hess[j][i]=hess[i][j];
4334: /*printf(" %lf ",hess[i][j]);*/
4335: }
4336: }
4337: }
4338: printf("\n");
4339: fprintf(ficlog,"\n");
4340:
4341: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4342: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4343:
4344: a=matrix(1,npar,1,npar);
4345: y=matrix(1,npar,1,npar);
4346: x=vector(1,npar);
4347: indx=ivector(1,npar);
4348: for (i=1;i<=npar;i++)
4349: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4350: ludcmp(a,npar,indx,&pd);
4351:
4352: for (j=1;j<=npar;j++) {
4353: for (i=1;i<=npar;i++) x[i]=0;
4354: x[j]=1;
4355: lubksb(a,npar,indx,x);
4356: for (i=1;i<=npar;i++){
4357: matcov[i][j]=x[i];
4358: }
4359: }
4360:
4361: printf("\n#Hessian matrix#\n");
4362: fprintf(ficlog,"\n#Hessian matrix#\n");
4363: for (i=1;i<=npar;i++) {
4364: for (j=1;j<=npar;j++) {
1.203 brouard 4365: printf("%.6e ",hess[i][j]);
4366: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4367: }
4368: printf("\n");
4369: fprintf(ficlog,"\n");
4370: }
4371:
1.203 brouard 4372: /* printf("\n#Covariance matrix#\n"); */
4373: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4374: /* for (i=1;i<=npar;i++) { */
4375: /* for (j=1;j<=npar;j++) { */
4376: /* printf("%.6e ",matcov[i][j]); */
4377: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4378: /* } */
4379: /* printf("\n"); */
4380: /* fprintf(ficlog,"\n"); */
4381: /* } */
4382:
1.126 brouard 4383: /* Recompute Inverse */
1.203 brouard 4384: /* for (i=1;i<=npar;i++) */
4385: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4386: /* ludcmp(a,npar,indx,&pd); */
4387:
4388: /* printf("\n#Hessian matrix recomputed#\n"); */
4389:
4390: /* for (j=1;j<=npar;j++) { */
4391: /* for (i=1;i<=npar;i++) x[i]=0; */
4392: /* x[j]=1; */
4393: /* lubksb(a,npar,indx,x); */
4394: /* for (i=1;i<=npar;i++){ */
4395: /* y[i][j]=x[i]; */
4396: /* printf("%.3e ",y[i][j]); */
4397: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4398: /* } */
4399: /* printf("\n"); */
4400: /* fprintf(ficlog,"\n"); */
4401: /* } */
4402:
4403: /* Verifying the inverse matrix */
4404: #ifdef DEBUGHESS
4405: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4406:
1.203 brouard 4407: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4408: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4409:
4410: for (j=1;j<=npar;j++) {
4411: for (i=1;i<=npar;i++){
1.203 brouard 4412: printf("%.2f ",y[i][j]);
4413: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4414: }
4415: printf("\n");
4416: fprintf(ficlog,"\n");
4417: }
1.203 brouard 4418: #endif
1.126 brouard 4419:
4420: free_matrix(a,1,npar,1,npar);
4421: free_matrix(y,1,npar,1,npar);
4422: free_vector(x,1,npar);
4423: free_ivector(indx,1,npar);
1.203 brouard 4424: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4425:
4426:
4427: }
4428:
4429: /*************** hessian matrix ****************/
4430: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4431: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4432: int i;
4433: int l=1, lmax=20;
1.203 brouard 4434: double k1,k2, res, fx;
1.132 brouard 4435: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4436: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4437: int k=0,kmax=10;
4438: double l1;
4439:
4440: fx=func(x);
4441: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4442: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4443: l1=pow(10,l);
4444: delts=delt;
4445: for(k=1 ; k <kmax; k=k+1){
4446: delt = delta*(l1*k);
4447: p2[theta]=x[theta] +delt;
1.145 brouard 4448: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4449: p2[theta]=x[theta]-delt;
4450: k2=func(p2)-fx;
4451: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4452: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4453:
1.203 brouard 4454: #ifdef DEBUGHESSII
1.126 brouard 4455: 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);
4456: 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);
4457: #endif
4458: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4459: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4460: k=kmax;
4461: }
4462: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4463: k=kmax; l=lmax*10;
1.126 brouard 4464: }
4465: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4466: delts=delt;
4467: }
1.203 brouard 4468: } /* End loop k */
1.126 brouard 4469: }
4470: delti[theta]=delts;
4471: return res;
4472:
4473: }
4474:
1.203 brouard 4475: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4476: {
4477: int i;
1.164 brouard 4478: int l=1, lmax=20;
1.126 brouard 4479: double k1,k2,k3,k4,res,fx;
1.132 brouard 4480: double p2[MAXPARM+1];
1.203 brouard 4481: int k, kmax=1;
4482: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4483:
4484: int firstime=0;
1.203 brouard 4485:
1.126 brouard 4486: fx=func(x);
1.203 brouard 4487: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4488: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4489: p2[thetai]=x[thetai]+delti[thetai]*k;
4490: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4491: k1=func(p2)-fx;
4492:
1.203 brouard 4493: p2[thetai]=x[thetai]+delti[thetai]*k;
4494: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4495: k2=func(p2)-fx;
4496:
1.203 brouard 4497: p2[thetai]=x[thetai]-delti[thetai]*k;
4498: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4499: k3=func(p2)-fx;
4500:
1.203 brouard 4501: p2[thetai]=x[thetai]-delti[thetai]*k;
4502: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4503: k4=func(p2)-fx;
1.203 brouard 4504: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4505: if(k1*k2*k3*k4 <0.){
1.208 brouard 4506: firstime=1;
1.203 brouard 4507: kmax=kmax+10;
1.208 brouard 4508: }
4509: if(kmax >=10 || firstime ==1){
1.246 brouard 4510: 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);
4511: 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 4512: 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);
4513: 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);
4514: }
4515: #ifdef DEBUGHESSIJ
4516: v1=hess[thetai][thetai];
4517: v2=hess[thetaj][thetaj];
4518: cv12=res;
4519: /* Computing eigen value of Hessian matrix */
4520: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4521: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4522: if ((lc2 <0) || (lc1 <0) ){
4523: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4524: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4525: 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);
4526: 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);
4527: }
1.126 brouard 4528: #endif
4529: }
4530: return res;
4531: }
4532:
1.203 brouard 4533: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4534: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4535: /* { */
4536: /* int i; */
4537: /* int l=1, lmax=20; */
4538: /* double k1,k2,k3,k4,res,fx; */
4539: /* double p2[MAXPARM+1]; */
4540: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4541: /* int k=0,kmax=10; */
4542: /* double l1; */
4543:
4544: /* fx=func(x); */
4545: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4546: /* l1=pow(10,l); */
4547: /* delts=delt; */
4548: /* for(k=1 ; k <kmax; k=k+1){ */
4549: /* delt = delti*(l1*k); */
4550: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4551: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4552: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4553: /* k1=func(p2)-fx; */
4554:
4555: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4556: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4557: /* k2=func(p2)-fx; */
4558:
4559: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4560: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4561: /* k3=func(p2)-fx; */
4562:
4563: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4564: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4565: /* k4=func(p2)-fx; */
4566: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4567: /* #ifdef DEBUGHESSIJ */
4568: /* 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); */
4569: /* 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); */
4570: /* #endif */
4571: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4572: /* k=kmax; */
4573: /* } */
4574: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4575: /* k=kmax; l=lmax*10; */
4576: /* } */
4577: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4578: /* delts=delt; */
4579: /* } */
4580: /* } /\* End loop k *\/ */
4581: /* } */
4582: /* delti[theta]=delts; */
4583: /* return res; */
4584: /* } */
4585:
4586:
1.126 brouard 4587: /************** Inverse of matrix **************/
4588: void ludcmp(double **a, int n, int *indx, double *d)
4589: {
4590: int i,imax,j,k;
4591: double big,dum,sum,temp;
4592: double *vv;
4593:
4594: vv=vector(1,n);
4595: *d=1.0;
4596: for (i=1;i<=n;i++) {
4597: big=0.0;
4598: for (j=1;j<=n;j++)
4599: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4600: if (big == 0.0){
4601: printf(" Singular Hessian matrix at row %d:\n",i);
4602: for (j=1;j<=n;j++) {
4603: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4604: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4605: }
4606: fflush(ficlog);
4607: fclose(ficlog);
4608: nrerror("Singular matrix in routine ludcmp");
4609: }
1.126 brouard 4610: vv[i]=1.0/big;
4611: }
4612: for (j=1;j<=n;j++) {
4613: for (i=1;i<j;i++) {
4614: sum=a[i][j];
4615: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4616: a[i][j]=sum;
4617: }
4618: big=0.0;
4619: for (i=j;i<=n;i++) {
4620: sum=a[i][j];
4621: for (k=1;k<j;k++)
4622: sum -= a[i][k]*a[k][j];
4623: a[i][j]=sum;
4624: if ( (dum=vv[i]*fabs(sum)) >= big) {
4625: big=dum;
4626: imax=i;
4627: }
4628: }
4629: if (j != imax) {
4630: for (k=1;k<=n;k++) {
4631: dum=a[imax][k];
4632: a[imax][k]=a[j][k];
4633: a[j][k]=dum;
4634: }
4635: *d = -(*d);
4636: vv[imax]=vv[j];
4637: }
4638: indx[j]=imax;
4639: if (a[j][j] == 0.0) a[j][j]=TINY;
4640: if (j != n) {
4641: dum=1.0/(a[j][j]);
4642: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4643: }
4644: }
4645: free_vector(vv,1,n); /* Doesn't work */
4646: ;
4647: }
4648:
4649: void lubksb(double **a, int n, int *indx, double b[])
4650: {
4651: int i,ii=0,ip,j;
4652: double sum;
4653:
4654: for (i=1;i<=n;i++) {
4655: ip=indx[i];
4656: sum=b[ip];
4657: b[ip]=b[i];
4658: if (ii)
4659: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4660: else if (sum) ii=i;
4661: b[i]=sum;
4662: }
4663: for (i=n;i>=1;i--) {
4664: sum=b[i];
4665: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4666: b[i]=sum/a[i][i];
4667: }
4668: }
4669:
4670: void pstamp(FILE *fichier)
4671: {
1.196 brouard 4672: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4673: }
4674:
1.297 brouard 4675: void date2dmy(double date,double *day, double *month, double *year){
4676: double yp=0., yp1=0., yp2=0.;
4677:
4678: yp1=modf(date,&yp);/* extracts integral of date in yp and
4679: fractional in yp1 */
4680: *year=yp;
4681: yp2=modf((yp1*12),&yp);
4682: *month=yp;
4683: yp1=modf((yp2*30.5),&yp);
4684: *day=yp;
4685: if(*day==0) *day=1;
4686: if(*month==0) *month=1;
4687: }
4688:
1.253 brouard 4689:
4690:
1.126 brouard 4691: /************ Frequencies ********************/
1.251 brouard 4692: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4693: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4694: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4695: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4696:
1.265 brouard 4697: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4698: int iind=0, iage=0;
4699: int mi; /* Effective wave */
4700: int first;
4701: double ***freq; /* Frequencies */
1.268 brouard 4702: 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 */
4703: 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 4704: double *meanq, *stdq, *idq;
1.226 brouard 4705: double **meanqt;
4706: double *pp, **prop, *posprop, *pospropt;
4707: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4708: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4709: double agebegin, ageend;
4710:
4711: pp=vector(1,nlstate);
1.251 brouard 4712: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4713: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4714: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4715: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4716: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4717: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4718: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4719: meanqt=matrix(1,lastpass,1,nqtveff);
4720: strcpy(fileresp,"P_");
4721: strcat(fileresp,fileresu);
4722: /*strcat(fileresphtm,fileresu);*/
4723: if((ficresp=fopen(fileresp,"w"))==NULL) {
4724: printf("Problem with prevalence resultfile: %s\n", fileresp);
4725: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4726: exit(0);
4727: }
1.240 brouard 4728:
1.226 brouard 4729: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4730: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4731: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4732: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4733: fflush(ficlog);
4734: exit(70);
4735: }
4736: else{
4737: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4738: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4739: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4740: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4741: }
1.319 brouard 4742: 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 4743:
1.226 brouard 4744: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4745: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4746: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4747: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4748: fflush(ficlog);
4749: exit(70);
1.240 brouard 4750: } else{
1.226 brouard 4751: 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 4752: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4753: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4754: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4755: }
1.319 brouard 4756: 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 4757:
1.253 brouard 4758: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4759: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4760: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4761: j1=0;
1.126 brouard 4762:
1.227 brouard 4763: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4764: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4765: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4766:
4767:
1.226 brouard 4768: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4769: reference=low_education V1=0,V2=0
4770: med_educ V1=1 V2=0,
4771: high_educ V1=0 V2=1
4772: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4773: */
1.249 brouard 4774: dateintsum=0;
4775: k2cpt=0;
4776:
1.253 brouard 4777: if(cptcoveff == 0 )
1.265 brouard 4778: nl=1; /* Constant and age model only */
1.253 brouard 4779: else
4780: nl=2;
1.265 brouard 4781:
4782: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4783: /* Loop on nj=1 or 2 if dummy covariates j!=0
4784: * Loop on j1(1 to 2**cptcoveff) covariate combination
4785: * freq[s1][s2][iage] =0.
4786: * Loop on iind
4787: * ++freq[s1][s2][iage] weighted
4788: * end iind
4789: * if covariate and j!0
4790: * headers Variable on one line
4791: * endif cov j!=0
4792: * header of frequency table by age
4793: * Loop on age
4794: * pp[s1]+=freq[s1][s2][iage] weighted
4795: * pos+=freq[s1][s2][iage] weighted
4796: * Loop on s1 initial state
4797: * fprintf(ficresp
4798: * end s1
4799: * end age
4800: * if j!=0 computes starting values
4801: * end compute starting values
4802: * end j1
4803: * end nl
4804: */
1.253 brouard 4805: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4806: if(nj==1)
4807: j=0; /* First pass for the constant */
1.265 brouard 4808: else{
1.253 brouard 4809: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4810: }
1.251 brouard 4811: first=1;
1.265 brouard 4812: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on all covariates combination of the model, excluding quantitatives, V4=0, V3=0 for example, fixed or varying covariates */
1.251 brouard 4813: posproptt=0.;
4814: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4815: scanf("%d", i);*/
4816: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4817: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4818: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4819: freq[i][s2][m]=0;
1.251 brouard 4820:
4821: for (i=1; i<=nlstate; i++) {
1.240 brouard 4822: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4823: prop[i][m]=0;
4824: posprop[i]=0;
4825: pospropt[i]=0;
4826: }
1.283 brouard 4827: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4828: idq[z1]=0.;
4829: meanq[z1]=0.;
4830: stdq[z1]=0.;
1.283 brouard 4831: }
4832: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4833: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4834: /* meanqt[m][z1]=0.; */
4835: /* } */
4836: /* } */
1.251 brouard 4837: /* dateintsum=0; */
4838: /* k2cpt=0; */
4839:
1.265 brouard 4840: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4841: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4842: bool=1;
4843: if(j !=0){
4844: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4845: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4846: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4847: /* if(Tvaraff[z1] ==-20){ */
4848: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4849: /* }else if(Tvaraff[z1] ==-10){ */
4850: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4851: /* }else */
4852: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4853: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4854: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4855: /* 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",
4856: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4857: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4858: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4859: } /* Onlyf fixed */
4860: } /* end z1 */
4861: } /* cptcovn > 0 */
4862: } /* end any */
4863: }/* end j==0 */
1.265 brouard 4864: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4865: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4866: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4867: m=mw[mi][iind];
4868: if(j!=0){
4869: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4870: for (z1=1; z1<=cptcoveff; z1++) {
4871: if( Fixed[Tmodelind[z1]]==1){
4872: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4873: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4874: value is -1, we don't select. It differs from the
4875: constant and age model which counts them. */
4876: bool=0; /* not selected */
4877: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4878: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4879: bool=0;
4880: }
4881: }
4882: }
4883: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4884: } /* end j==0 */
4885: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4886: if(bool==1){ /*Selected */
1.251 brouard 4887: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4888: and mw[mi+1][iind]. dh depends on stepm. */
4889: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4890: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4891: if(m >=firstpass && m <=lastpass){
4892: k2=anint[m][iind]+(mint[m][iind]/12.);
4893: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4894: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4895: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4896: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4897: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4898: if (m<lastpass) {
4899: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4900: /* 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]); */
4901: if(s[m][iind]==-1)
4902: 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.));
4903: 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 4904: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
4905: if(!isnan(covar[ncovcol+z1][iind])){
4906: idq[z1]=idq[z1]+weight[iind];
4907: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4908: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
4909: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4910: }
1.284 brouard 4911: }
1.251 brouard 4912: /* if((int)agev[m][iind] == 55) */
4913: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4914: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4915: 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 4916: }
1.251 brouard 4917: } /* end if between passes */
4918: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4919: dateintsum=dateintsum+k2; /* on all covariates ?*/
4920: k2cpt++;
4921: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4922: }
1.251 brouard 4923: }else{
4924: bool=1;
4925: }/* end bool 2 */
4926: } /* end m */
1.284 brouard 4927: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4928: /* idq[z1]=idq[z1]+weight[iind]; */
4929: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4930: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4931: /* } */
1.251 brouard 4932: } /* end bool */
4933: } /* end iind = 1 to imx */
1.319 brouard 4934: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 4935: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4936:
4937:
4938: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4939: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4940: pstamp(ficresp);
1.251 brouard 4941: if (cptcoveff>0 && j!=0){
1.265 brouard 4942: pstamp(ficresp);
1.251 brouard 4943: printf( "\n#********** Variable ");
4944: fprintf(ficresp, "\n#********** Variable ");
4945: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4946: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4947: fprintf(ficlog, "\n#********** Variable ");
4948: for (z1=1; z1<=cptcoveff; z1++){
4949: if(!FixedV[Tvaraff[z1]]){
4950: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4951: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4952: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4953: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4954: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4955: }else{
1.251 brouard 4956: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4957: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4958: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4959: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4960: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4961: }
4962: }
4963: printf( "**********\n#");
4964: fprintf(ficresp, "**********\n#");
4965: fprintf(ficresphtm, "**********</h3>\n");
4966: fprintf(ficresphtmfr, "**********</h3>\n");
4967: fprintf(ficlog, "**********\n");
4968: }
1.284 brouard 4969: /*
4970: Printing means of quantitative variables if any
4971: */
4972: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 4973: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 4974: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 4975: if(weightopt==1){
4976: printf(" Weighted mean and standard deviation of");
4977: fprintf(ficlog," Weighted mean and standard deviation of");
4978: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4979: }
1.311 brouard 4980: /* mu = \frac{w x}{\sum w}
4981: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
4982: */
4983: 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]));
4984: 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]));
4985: 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 4986: }
4987: /* for (z1=1; z1<= nqtveff; z1++) { */
4988: /* for(m=1;m<=lastpass;m++){ */
4989: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4990: /* } */
4991: /* } */
1.283 brouard 4992:
1.251 brouard 4993: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4994: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4995: fprintf(ficresp, " Age");
4996: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.251 brouard 4997: for(i=1; i<=nlstate;i++) {
1.265 brouard 4998: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4999: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5000: }
1.265 brouard 5001: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5002: fprintf(ficresphtm, "\n");
5003:
5004: /* Header of frequency table by age */
5005: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5006: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5007: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5008: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5009: if(s2!=0 && m!=0)
5010: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5011: }
1.226 brouard 5012: }
1.251 brouard 5013: fprintf(ficresphtmfr, "\n");
5014:
5015: /* For each age */
5016: for(iage=iagemin; iage <= iagemax+3; iage++){
5017: fprintf(ficresphtm,"<tr>");
5018: if(iage==iagemax+1){
5019: fprintf(ficlog,"1");
5020: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5021: }else if(iage==iagemax+2){
5022: fprintf(ficlog,"0");
5023: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5024: }else if(iage==iagemax+3){
5025: fprintf(ficlog,"Total");
5026: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5027: }else{
1.240 brouard 5028: if(first==1){
1.251 brouard 5029: first=0;
5030: printf("See log file for details...\n");
5031: }
5032: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5033: fprintf(ficlog,"Age %d", iage);
5034: }
1.265 brouard 5035: for(s1=1; s1 <=nlstate ; s1++){
5036: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5037: pp[s1] += freq[s1][m][iage];
1.251 brouard 5038: }
1.265 brouard 5039: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5040: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5041: pos += freq[s1][m][iage];
5042: if(pp[s1]>=1.e-10){
1.251 brouard 5043: if(first==1){
1.265 brouard 5044: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5045: }
1.265 brouard 5046: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5047: }else{
5048: if(first==1)
1.265 brouard 5049: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5050: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5051: }
5052: }
5053:
1.265 brouard 5054: for(s1=1; s1 <=nlstate ; s1++){
5055: /* posprop[s1]=0; */
5056: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5057: pp[s1] += freq[s1][m][iage];
5058: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5059:
5060: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5061: pos += pp[s1]; /* pos is the total number of transitions until this age */
5062: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5063: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5064: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5065: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5066: }
5067:
5068: /* Writing ficresp */
5069: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
5070: if( iage <= iagemax){
5071: fprintf(ficresp," %d",iage);
5072: }
5073: }else if( nj==2){
5074: if( iage <= iagemax){
5075: fprintf(ficresp," %d",iage);
5076: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
5077: }
1.240 brouard 5078: }
1.265 brouard 5079: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5080: if(pos>=1.e-5){
1.251 brouard 5081: if(first==1)
1.265 brouard 5082: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5083: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5084: }else{
5085: if(first==1)
1.265 brouard 5086: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5087: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5088: }
5089: if( iage <= iagemax){
5090: if(pos>=1.e-5){
1.265 brouard 5091: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
5092: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5093: }else if( nj==2){
5094: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5095: }
5096: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5097: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5098: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5099: } else{
5100: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
5101: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5102: }
1.240 brouard 5103: }
1.265 brouard 5104: pospropt[s1] +=posprop[s1];
5105: } /* end loop s1 */
1.251 brouard 5106: /* pospropt=0.; */
1.265 brouard 5107: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5108: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5109: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5110: if(first==1){
1.265 brouard 5111: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5112: }
1.265 brouard 5113: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5114: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5115: }
1.265 brouard 5116: if(s1!=0 && m!=0)
5117: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5118: }
1.265 brouard 5119: } /* end loop s1 */
1.251 brouard 5120: posproptt=0.;
1.265 brouard 5121: for(s1=1; s1 <=nlstate; s1++){
5122: posproptt += pospropt[s1];
1.251 brouard 5123: }
5124: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5125: fprintf(ficresphtm,"</tr>\n");
5126: if((cptcoveff==0 && nj==1)|| nj==2 ) {
5127: if(iage <= iagemax)
5128: fprintf(ficresp,"\n");
1.240 brouard 5129: }
1.251 brouard 5130: if(first==1)
5131: printf("Others in log...\n");
5132: fprintf(ficlog,"\n");
5133: } /* end loop age iage */
1.265 brouard 5134:
1.251 brouard 5135: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5136: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5137: if(posproptt < 1.e-5){
1.265 brouard 5138: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5139: }else{
1.265 brouard 5140: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5141: }
1.226 brouard 5142: }
1.251 brouard 5143: fprintf(ficresphtm,"</tr>\n");
5144: fprintf(ficresphtm,"</table>\n");
5145: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5146: if(posproptt < 1.e-5){
1.251 brouard 5147: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5148: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5149: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5150: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5151: invalidvarcomb[j1]=1;
1.226 brouard 5152: }else{
1.251 brouard 5153: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
5154: invalidvarcomb[j1]=0;
1.226 brouard 5155: }
1.251 brouard 5156: fprintf(ficresphtmfr,"</table>\n");
5157: fprintf(ficlog,"\n");
5158: if(j!=0){
5159: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5160: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5161: for(k=1; k <=(nlstate+ndeath); k++){
5162: if (k != i) {
1.265 brouard 5163: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5164: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5165: if(j1==1){ /* All dummy covariates to zero */
5166: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5167: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5168: printf("%d%d ",i,k);
5169: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5170: 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]));
5171: 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]));
5172: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5173: }
1.253 brouard 5174: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5175: for(iage=iagemin; iage <= iagemax+3; iage++){
5176: x[iage]= (double)iage;
5177: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5178: /* 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 5179: }
1.268 brouard 5180: /* Some are not finite, but linreg will ignore these ages */
5181: no=0;
1.253 brouard 5182: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5183: pstart[s1]=b;
5184: pstart[s1-1]=a;
1.252 brouard 5185: }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 */
5186: 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]);
5187: 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 5188: 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 5189: printf("%d%d ",i,k);
5190: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5191: 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 5192: }else{ /* Other cases, like quantitative fixed or varying covariates */
5193: ;
5194: }
5195: /* printf("%12.7f )", param[i][jj][k]); */
5196: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5197: s1++;
1.251 brouard 5198: } /* end jj */
5199: } /* end k!= i */
5200: } /* end k */
1.265 brouard 5201: } /* end i, s1 */
1.251 brouard 5202: } /* end j !=0 */
5203: } /* end selected combination of covariate j1 */
5204: if(j==0){ /* We can estimate starting values from the occurences in each case */
5205: printf("#Freqsummary: Starting values for the constants:\n");
5206: fprintf(ficlog,"\n");
1.265 brouard 5207: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5208: for(k=1; k <=(nlstate+ndeath); k++){
5209: if (k != i) {
5210: printf("%d%d ",i,k);
5211: fprintf(ficlog,"%d%d ",i,k);
5212: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5213: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5214: if(jj==1){ /* Age has to be done */
1.265 brouard 5215: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5216: 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]));
5217: 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 5218: }
5219: /* printf("%12.7f )", param[i][jj][k]); */
5220: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5221: s1++;
1.250 brouard 5222: }
1.251 brouard 5223: printf("\n");
5224: fprintf(ficlog,"\n");
1.250 brouard 5225: }
5226: }
1.284 brouard 5227: } /* end of state i */
1.251 brouard 5228: printf("#Freqsummary\n");
5229: fprintf(ficlog,"\n");
1.265 brouard 5230: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5231: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5232: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5233: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5234: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5235: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5236: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5237: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5238: /* } */
5239: }
1.265 brouard 5240: } /* end loop s1 */
1.251 brouard 5241:
5242: printf("\n");
5243: fprintf(ficlog,"\n");
5244: } /* end j=0 */
1.249 brouard 5245: } /* end j */
1.252 brouard 5246:
1.253 brouard 5247: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5248: for(i=1, jk=1; i <=nlstate; i++){
5249: for(j=1; j <=nlstate+ndeath; j++){
5250: if(j!=i){
5251: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5252: printf("%1d%1d",i,j);
5253: fprintf(ficparo,"%1d%1d",i,j);
5254: for(k=1; k<=ncovmodel;k++){
5255: /* printf(" %lf",param[i][j][k]); */
5256: /* fprintf(ficparo," %lf",param[i][j][k]); */
5257: p[jk]=pstart[jk];
5258: printf(" %f ",pstart[jk]);
5259: fprintf(ficparo," %f ",pstart[jk]);
5260: jk++;
5261: }
5262: printf("\n");
5263: fprintf(ficparo,"\n");
5264: }
5265: }
5266: }
5267: } /* end mle=-2 */
1.226 brouard 5268: dateintmean=dateintsum/k2cpt;
1.296 brouard 5269: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5270:
1.226 brouard 5271: fclose(ficresp);
5272: fclose(ficresphtm);
5273: fclose(ficresphtmfr);
1.283 brouard 5274: free_vector(idq,1,nqfveff);
1.226 brouard 5275: free_vector(meanq,1,nqfveff);
1.284 brouard 5276: free_vector(stdq,1,nqfveff);
1.226 brouard 5277: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5278: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5279: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5280: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5281: free_vector(pospropt,1,nlstate);
5282: free_vector(posprop,1,nlstate);
1.251 brouard 5283: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5284: free_vector(pp,1,nlstate);
5285: /* End of freqsummary */
5286: }
1.126 brouard 5287:
1.268 brouard 5288: /* Simple linear regression */
5289: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5290:
5291: /* y=a+bx regression */
5292: double sumx = 0.0; /* sum of x */
5293: double sumx2 = 0.0; /* sum of x**2 */
5294: double sumxy = 0.0; /* sum of x * y */
5295: double sumy = 0.0; /* sum of y */
5296: double sumy2 = 0.0; /* sum of y**2 */
5297: double sume2 = 0.0; /* sum of square or residuals */
5298: double yhat;
5299:
5300: double denom=0;
5301: int i;
5302: int ne=*no;
5303:
5304: for ( i=ifi, ne=0;i<=ila;i++) {
5305: if(!isfinite(x[i]) || !isfinite(y[i])){
5306: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5307: continue;
5308: }
5309: ne=ne+1;
5310: sumx += x[i];
5311: sumx2 += x[i]*x[i];
5312: sumxy += x[i] * y[i];
5313: sumy += y[i];
5314: sumy2 += y[i]*y[i];
5315: denom = (ne * sumx2 - sumx*sumx);
5316: /* 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); */
5317: }
5318:
5319: denom = (ne * sumx2 - sumx*sumx);
5320: if (denom == 0) {
5321: // vertical, slope m is infinity
5322: *b = INFINITY;
5323: *a = 0;
5324: if (r) *r = 0;
5325: return 1;
5326: }
5327:
5328: *b = (ne * sumxy - sumx * sumy) / denom;
5329: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5330: if (r!=NULL) {
5331: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5332: sqrt((sumx2 - sumx*sumx/ne) *
5333: (sumy2 - sumy*sumy/ne));
5334: }
5335: *no=ne;
5336: for ( i=ifi, ne=0;i<=ila;i++) {
5337: if(!isfinite(x[i]) || !isfinite(y[i])){
5338: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5339: continue;
5340: }
5341: ne=ne+1;
5342: yhat = y[i] - *a -*b* x[i];
5343: sume2 += yhat * yhat ;
5344:
5345: denom = (ne * sumx2 - sumx*sumx);
5346: /* 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); */
5347: }
5348: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5349: *sa= *sb * sqrt(sumx2/ne);
5350:
5351: return 0;
5352: }
5353:
1.126 brouard 5354: /************ Prevalence ********************/
1.227 brouard 5355: 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)
5356: {
5357: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5358: in each health status at the date of interview (if between dateprev1 and dateprev2).
5359: We still use firstpass and lastpass as another selection.
5360: */
1.126 brouard 5361:
1.227 brouard 5362: int i, m, jk, j1, bool, z1,j, iv;
5363: int mi; /* Effective wave */
5364: int iage;
5365: double agebegin, ageend;
5366:
5367: double **prop;
5368: double posprop;
5369: double y2; /* in fractional years */
5370: int iagemin, iagemax;
5371: int first; /** to stop verbosity which is redirected to log file */
5372:
5373: iagemin= (int) agemin;
5374: iagemax= (int) agemax;
5375: /*pp=vector(1,nlstate);*/
1.251 brouard 5376: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5377: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5378: j1=0;
1.222 brouard 5379:
1.227 brouard 5380: /*j=cptcoveff;*/
5381: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5382:
1.288 brouard 5383: first=0;
1.227 brouard 5384: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5385: for (i=1; i<=nlstate; i++)
1.251 brouard 5386: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5387: prop[i][iage]=0.0;
5388: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5389: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5390: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5391:
5392: for (i=1; i<=imx; i++) { /* Each individual */
5393: bool=1;
5394: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5395: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5396: m=mw[mi][i];
5397: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5398: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5399: for (z1=1; z1<=cptcoveff; z1++){
5400: if( Fixed[Tmodelind[z1]]==1){
5401: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5402: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5403: bool=0;
5404: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5405: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5406: bool=0;
5407: }
5408: }
5409: if(bool==1){ /* Otherwise we skip that wave/person */
5410: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5411: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5412: if(m >=firstpass && m <=lastpass){
5413: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5414: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5415: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5416: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5417: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5418: 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);
5419: exit(1);
5420: }
5421: if (s[m][i]>0 && s[m][i]<=nlstate) {
5422: /*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]]);*/
5423: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5424: prop[s[m][i]][iagemax+3] += weight[i];
5425: } /* end valid statuses */
5426: } /* end selection of dates */
5427: } /* end selection of waves */
5428: } /* end bool */
5429: } /* end wave */
5430: } /* end individual */
5431: for(i=iagemin; i <= iagemax+3; i++){
5432: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5433: posprop += prop[jk][i];
5434: }
5435:
5436: for(jk=1; jk <=nlstate ; jk++){
5437: if( i <= iagemax){
5438: if(posprop>=1.e-5){
5439: probs[i][jk][j1]= prop[jk][i]/posprop;
5440: } else{
1.288 brouard 5441: if(!first){
5442: first=1;
1.266 brouard 5443: 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]);
5444: }else{
1.288 brouard 5445: 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 5446: }
5447: }
5448: }
5449: }/* end jk */
5450: }/* end i */
1.222 brouard 5451: /*} *//* end i1 */
1.227 brouard 5452: } /* end j1 */
1.222 brouard 5453:
1.227 brouard 5454: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5455: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5456: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5457: } /* End of prevalence */
1.126 brouard 5458:
5459: /************* Waves Concatenation ***************/
5460:
5461: 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)
5462: {
1.298 brouard 5463: /* 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 5464: Death is a valid wave (if date is known).
5465: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5466: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5467: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5468: */
1.126 brouard 5469:
1.224 brouard 5470: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5471: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5472: double sum=0., jmean=0.;*/
1.224 brouard 5473: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5474: int j, k=0,jk, ju, jl;
5475: double sum=0.;
5476: first=0;
1.214 brouard 5477: firstwo=0;
1.217 brouard 5478: firsthree=0;
1.218 brouard 5479: firstfour=0;
1.164 brouard 5480: jmin=100000;
1.126 brouard 5481: jmax=-1;
5482: jmean=0.;
1.224 brouard 5483:
5484: /* Treating live states */
1.214 brouard 5485: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5486: mi=0; /* First valid wave */
1.227 brouard 5487: mli=0; /* Last valid wave */
1.309 brouard 5488: m=firstpass; /* Loop on waves */
5489: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5490: 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 */
5491: mli=m-1;/* mw[++mi][i]=m-1; */
5492: }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 5493: 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 5494: mli=m;
1.224 brouard 5495: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5496: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5497: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5498: }
1.309 brouard 5499: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5500: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5501: break;
1.224 brouard 5502: #else
1.317 brouard 5503: 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 5504: if(firsthree == 0){
1.302 brouard 5505: 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 5506: firsthree=1;
1.317 brouard 5507: }else if(firsthree >=1 && firsthree < 10){
5508: 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);
5509: firsthree++;
5510: }else if(firsthree == 10){
5511: printf("Information, too many Information flags: no more reported to log either\n");
5512: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
5513: firsthree++;
5514: }else{
5515: firsthree++;
1.227 brouard 5516: }
1.309 brouard 5517: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5518: mli=m;
5519: }
5520: if(s[m][i]==-2){ /* Vital status is really unknown */
5521: nbwarn++;
1.309 brouard 5522: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5523: 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);
5524: 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);
5525: }
5526: break;
5527: }
5528: break;
1.224 brouard 5529: #endif
1.227 brouard 5530: }/* End m >= lastpass */
1.126 brouard 5531: }/* end while */
1.224 brouard 5532:
1.227 brouard 5533: /* 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 5534: /* After last pass */
1.224 brouard 5535: /* Treating death states */
1.214 brouard 5536: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5537: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5538: /* } */
1.126 brouard 5539: mi++; /* Death is another wave */
5540: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5541: /* Only death is a correct wave */
1.126 brouard 5542: mw[mi][i]=m;
1.257 brouard 5543: } /* else not in a death state */
1.224 brouard 5544: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5545: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5546: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5547: 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 5548: nbwarn++;
5549: if(firstfiv==0){
1.309 brouard 5550: 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 5551: firstfiv=1;
5552: }else{
1.309 brouard 5553: 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 5554: }
1.309 brouard 5555: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5556: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5557: nberr++;
5558: if(firstwo==0){
1.309 brouard 5559: 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 5560: firstwo=1;
5561: }
1.309 brouard 5562: 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 5563: }
1.257 brouard 5564: }else{ /* if date of interview is unknown */
1.227 brouard 5565: /* death is known but not confirmed by death status at any wave */
5566: if(firstfour==0){
1.309 brouard 5567: 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 5568: firstfour=1;
5569: }
1.309 brouard 5570: 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 5571: }
1.224 brouard 5572: } /* end if date of death is known */
5573: #endif
1.309 brouard 5574: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5575: /* wav[i]=mw[mi][i]; */
1.126 brouard 5576: if(mi==0){
5577: nbwarn++;
5578: if(first==0){
1.227 brouard 5579: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5580: first=1;
1.126 brouard 5581: }
5582: if(first==1){
1.227 brouard 5583: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5584: }
5585: } /* end mi==0 */
5586: } /* End individuals */
1.214 brouard 5587: /* wav and mw are no more changed */
1.223 brouard 5588:
1.317 brouard 5589: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
5590: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
5591:
5592:
1.126 brouard 5593: for(i=1; i<=imx; i++){
5594: for(mi=1; mi<wav[i];mi++){
5595: if (stepm <=0)
1.227 brouard 5596: dh[mi][i]=1;
1.126 brouard 5597: else{
1.260 brouard 5598: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5599: if (agedc[i] < 2*AGESUP) {
5600: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5601: if(j==0) j=1; /* Survives at least one month after exam */
5602: else if(j<0){
5603: nberr++;
5604: 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]);
5605: j=1; /* Temporary Dangerous patch */
5606: 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);
5607: 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]);
5608: 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);
5609: }
5610: k=k+1;
5611: if (j >= jmax){
5612: jmax=j;
5613: ijmax=i;
5614: }
5615: if (j <= jmin){
5616: jmin=j;
5617: ijmin=i;
5618: }
5619: sum=sum+j;
5620: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5621: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5622: }
5623: }
5624: else{
5625: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5626: /* 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 5627:
1.227 brouard 5628: k=k+1;
5629: if (j >= jmax) {
5630: jmax=j;
5631: ijmax=i;
5632: }
5633: else if (j <= jmin){
5634: jmin=j;
5635: ijmin=i;
5636: }
5637: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5638: /*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]);*/
5639: if(j<0){
5640: nberr++;
5641: 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]);
5642: 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]);
5643: }
5644: sum=sum+j;
5645: }
5646: jk= j/stepm;
5647: jl= j -jk*stepm;
5648: ju= j -(jk+1)*stepm;
5649: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5650: if(jl==0){
5651: dh[mi][i]=jk;
5652: bh[mi][i]=0;
5653: }else{ /* We want a negative bias in order to only have interpolation ie
5654: * to avoid the price of an extra matrix product in likelihood */
5655: dh[mi][i]=jk+1;
5656: bh[mi][i]=ju;
5657: }
5658: }else{
5659: if(jl <= -ju){
5660: dh[mi][i]=jk;
5661: bh[mi][i]=jl; /* bias is positive if real duration
5662: * is higher than the multiple of stepm and negative otherwise.
5663: */
5664: }
5665: else{
5666: dh[mi][i]=jk+1;
5667: bh[mi][i]=ju;
5668: }
5669: if(dh[mi][i]==0){
5670: dh[mi][i]=1; /* At least one step */
5671: bh[mi][i]=ju; /* At least one step */
5672: /* 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);*/
5673: }
5674: } /* end if mle */
1.126 brouard 5675: }
5676: } /* end wave */
5677: }
5678: jmean=sum/k;
5679: 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 5680: 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 5681: }
1.126 brouard 5682:
5683: /*********** Tricode ****************************/
1.220 brouard 5684: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5685: {
5686: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5687: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5688: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5689: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5690: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5691: */
1.130 brouard 5692:
1.242 brouard 5693: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5694: int modmaxcovj=0; /* Modality max of covariates j */
5695: int cptcode=0; /* Modality max of covariates j */
5696: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5697:
5698:
1.242 brouard 5699: /* cptcoveff=0; */
5700: /* *cptcov=0; */
1.126 brouard 5701:
1.242 brouard 5702: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5703: for (k=1; k <= maxncov; k++)
5704: for(j=1; j<=2; j++)
5705: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5706:
1.242 brouard 5707: /* Loop on covariates without age and products and no quantitative variable */
5708: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5709: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5710: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5711: switch(Fixed[k]) {
5712: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5713: modmaxcovj=0;
5714: modmincovj=0;
1.242 brouard 5715: 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*/
5716: ij=(int)(covar[Tvar[k]][i]);
5717: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5718: * If product of Vn*Vm, still boolean *:
5719: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5720: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5721: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5722: modality of the nth covariate of individual i. */
5723: if (ij > modmaxcovj)
5724: modmaxcovj=ij;
5725: else if (ij < modmincovj)
5726: modmincovj=ij;
1.287 brouard 5727: if (ij <0 || ij >1 ){
1.311 brouard 5728: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5729: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5730: fflush(ficlog);
5731: exit(1);
1.287 brouard 5732: }
5733: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5734: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5735: exit(1);
5736: }else
5737: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5738: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5739: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5740: /* getting the maximum value of the modality of the covariate
5741: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5742: female ies 1, then modmaxcovj=1.
5743: */
5744: } /* end for loop on individuals i */
5745: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5746: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5747: cptcode=modmaxcovj;
5748: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5749: /*for (i=0; i<=cptcode; i++) {*/
5750: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5751: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5752: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5753: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5754: if( j != -1){
5755: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5756: covariate for which somebody answered excluding
5757: undefined. Usually 2: 0 and 1. */
5758: }
5759: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5760: covariate for which somebody answered including
5761: undefined. Usually 3: -1, 0 and 1. */
5762: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5763: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5764: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5765:
1.242 brouard 5766: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5767: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5768: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5769: /* modmincovj=3; modmaxcovj = 7; */
5770: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5771: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5772: /* defining two dummy variables: variables V1_1 and V1_2.*/
5773: /* nbcode[Tvar[j]][ij]=k; */
5774: /* nbcode[Tvar[j]][1]=0; */
5775: /* nbcode[Tvar[j]][2]=1; */
5776: /* nbcode[Tvar[j]][3]=2; */
5777: /* To be continued (not working yet). */
5778: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5779:
5780: /* 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*/
5781: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5782: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5783: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5784: /*, could be restored in the future */
5785: 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 5786: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5787: break;
5788: }
5789: ij++;
1.287 brouard 5790: 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 5791: cptcode = ij; /* New max modality for covar j */
5792: } /* end of loop on modality i=-1 to 1 or more */
5793: break;
5794: case 1: /* Testing on varying covariate, could be simple and
5795: * should look at waves or product of fixed *
5796: * varying. No time to test -1, assuming 0 and 1 only */
5797: ij=0;
5798: for(i=0; i<=1;i++){
5799: nbcode[Tvar[k]][++ij]=i;
5800: }
5801: break;
5802: default:
5803: break;
5804: } /* end switch */
5805: } /* end dummy test */
1.311 brouard 5806: if(Dummy[k]==1 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5807: 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*/
5808: if(isnan(covar[Tvar[k]][i])){
5809: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5810: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5811: fflush(ficlog);
5812: exit(1);
5813: }
5814: }
5815: }
1.287 brouard 5816: } /* 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 5817:
5818: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5819: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5820: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5821: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5822: 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 */
5823: 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 */
5824: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5825: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5826:
5827: ij=0;
5828: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5829: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5830: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5831: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5832: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5833: /* If product not in single variable we don't print results */
5834: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5835: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5836: 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*/
5837: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5838: 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 */
5839: if(Fixed[k]!=0)
5840: anyvaryingduminmodel=1;
5841: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5842: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5843: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5844: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5845: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5846: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5847: }
5848: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5849: /* ij--; */
5850: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5851: *cptcov=ij; /*Number of total real effective covariates: effective
5852: * because they can be excluded from the model and real
5853: * if in the model but excluded because missing values, but how to get k from ij?*/
5854: for(j=ij+1; j<= cptcovt; j++){
5855: Tvaraff[j]=0;
5856: Tmodelind[j]=0;
5857: }
5858: for(j=ntveff+1; j<= cptcovt; j++){
5859: TmodelInvind[j]=0;
5860: }
5861: /* To be sorted */
5862: ;
5863: }
1.126 brouard 5864:
1.145 brouard 5865:
1.126 brouard 5866: /*********** Health Expectancies ****************/
5867:
1.235 brouard 5868: 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 5869:
5870: {
5871: /* Health expectancies, no variances */
1.164 brouard 5872: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5873: int nhstepma, nstepma; /* Decreasing with age */
5874: double age, agelim, hf;
5875: double ***p3mat;
5876: double eip;
5877:
1.238 brouard 5878: /* pstamp(ficreseij); */
1.126 brouard 5879: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5880: fprintf(ficreseij,"# Age");
5881: for(i=1; i<=nlstate;i++){
5882: for(j=1; j<=nlstate;j++){
5883: fprintf(ficreseij," e%1d%1d ",i,j);
5884: }
5885: fprintf(ficreseij," e%1d. ",i);
5886: }
5887: fprintf(ficreseij,"\n");
5888:
5889:
5890: if(estepm < stepm){
5891: printf ("Problem %d lower than %d\n",estepm, stepm);
5892: }
5893: else hstepm=estepm;
5894: /* We compute the life expectancy from trapezoids spaced every estepm months
5895: * This is mainly to measure the difference between two models: for example
5896: * if stepm=24 months pijx are given only every 2 years and by summing them
5897: * we are calculating an estimate of the Life Expectancy assuming a linear
5898: * progression in between and thus overestimating or underestimating according
5899: * to the curvature of the survival function. If, for the same date, we
5900: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5901: * to compare the new estimate of Life expectancy with the same linear
5902: * hypothesis. A more precise result, taking into account a more precise
5903: * curvature will be obtained if estepm is as small as stepm. */
5904:
5905: /* For example we decided to compute the life expectancy with the smallest unit */
5906: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5907: nhstepm is the number of hstepm from age to agelim
5908: nstepm is the number of stepm from age to agelin.
1.270 brouard 5909: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5910: and note for a fixed period like estepm months */
5911: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5912: survival function given by stepm (the optimization length). Unfortunately it
5913: means that if the survival funtion is printed only each two years of age and if
5914: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5915: results. So we changed our mind and took the option of the best precision.
5916: */
5917: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5918:
5919: agelim=AGESUP;
5920: /* If stepm=6 months */
5921: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5922: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5923:
5924: /* nhstepm age range expressed in number of stepm */
5925: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5926: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5927: /* if (stepm >= YEARM) hstepm=1;*/
5928: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5929: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5930:
5931: for (age=bage; age<=fage; age ++){
5932: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5933: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5934: /* if (stepm >= YEARM) hstepm=1;*/
5935: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5936:
5937: /* If stepm=6 months */
5938: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5939: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5940:
1.235 brouard 5941: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5942:
5943: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5944:
5945: printf("%d|",(int)age);fflush(stdout);
5946: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5947:
5948: /* Computing expectancies */
5949: for(i=1; i<=nlstate;i++)
5950: for(j=1; j<=nlstate;j++)
5951: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5952: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5953:
5954: /* 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]);*/
5955:
5956: }
5957:
5958: fprintf(ficreseij,"%3.0f",age );
5959: for(i=1; i<=nlstate;i++){
5960: eip=0;
5961: for(j=1; j<=nlstate;j++){
5962: eip +=eij[i][j][(int)age];
5963: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5964: }
5965: fprintf(ficreseij,"%9.4f", eip );
5966: }
5967: fprintf(ficreseij,"\n");
5968:
5969: }
5970: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5971: printf("\n");
5972: fprintf(ficlog,"\n");
5973:
5974: }
5975:
1.235 brouard 5976: 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 5977:
5978: {
5979: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5980: to initial status i, ei. .
1.126 brouard 5981: */
5982: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5983: int nhstepma, nstepma; /* Decreasing with age */
5984: double age, agelim, hf;
5985: double ***p3matp, ***p3matm, ***varhe;
5986: double **dnewm,**doldm;
5987: double *xp, *xm;
5988: double **gp, **gm;
5989: double ***gradg, ***trgradg;
5990: int theta;
5991:
5992: double eip, vip;
5993:
5994: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5995: xp=vector(1,npar);
5996: xm=vector(1,npar);
5997: dnewm=matrix(1,nlstate*nlstate,1,npar);
5998: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5999:
6000: pstamp(ficresstdeij);
6001: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6002: fprintf(ficresstdeij,"# Age");
6003: for(i=1; i<=nlstate;i++){
6004: for(j=1; j<=nlstate;j++)
6005: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6006: fprintf(ficresstdeij," e%1d. ",i);
6007: }
6008: fprintf(ficresstdeij,"\n");
6009:
6010: pstamp(ficrescveij);
6011: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6012: fprintf(ficrescveij,"# Age");
6013: for(i=1; i<=nlstate;i++)
6014: for(j=1; j<=nlstate;j++){
6015: cptj= (j-1)*nlstate+i;
6016: for(i2=1; i2<=nlstate;i2++)
6017: for(j2=1; j2<=nlstate;j2++){
6018: cptj2= (j2-1)*nlstate+i2;
6019: if(cptj2 <= cptj)
6020: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6021: }
6022: }
6023: fprintf(ficrescveij,"\n");
6024:
6025: if(estepm < stepm){
6026: printf ("Problem %d lower than %d\n",estepm, stepm);
6027: }
6028: else hstepm=estepm;
6029: /* We compute the life expectancy from trapezoids spaced every estepm months
6030: * This is mainly to measure the difference between two models: for example
6031: * if stepm=24 months pijx are given only every 2 years and by summing them
6032: * we are calculating an estimate of the Life Expectancy assuming a linear
6033: * progression in between and thus overestimating or underestimating according
6034: * to the curvature of the survival function. If, for the same date, we
6035: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6036: * to compare the new estimate of Life expectancy with the same linear
6037: * hypothesis. A more precise result, taking into account a more precise
6038: * curvature will be obtained if estepm is as small as stepm. */
6039:
6040: /* For example we decided to compute the life expectancy with the smallest unit */
6041: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6042: nhstepm is the number of hstepm from age to agelim
6043: nstepm is the number of stepm from age to agelin.
6044: Look at hpijx to understand the reason of that which relies in memory size
6045: and note for a fixed period like estepm months */
6046: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6047: survival function given by stepm (the optimization length). Unfortunately it
6048: means that if the survival funtion is printed only each two years of age and if
6049: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6050: results. So we changed our mind and took the option of the best precision.
6051: */
6052: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6053:
6054: /* If stepm=6 months */
6055: /* nhstepm age range expressed in number of stepm */
6056: agelim=AGESUP;
6057: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6058: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6059: /* if (stepm >= YEARM) hstepm=1;*/
6060: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6061:
6062: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6063: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6064: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6065: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6066: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6067: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6068:
6069: for (age=bage; age<=fage; age ++){
6070: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6071: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6072: /* if (stepm >= YEARM) hstepm=1;*/
6073: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6074:
1.126 brouard 6075: /* If stepm=6 months */
6076: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6077: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6078:
6079: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6080:
1.126 brouard 6081: /* Computing Variances of health expectancies */
6082: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6083: decrease memory allocation */
6084: for(theta=1; theta <=npar; theta++){
6085: for(i=1; i<=npar; i++){
1.222 brouard 6086: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6087: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6088: }
1.235 brouard 6089: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6090: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6091:
1.126 brouard 6092: for(j=1; j<= nlstate; j++){
1.222 brouard 6093: for(i=1; i<=nlstate; i++){
6094: for(h=0; h<=nhstepm-1; h++){
6095: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6096: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6097: }
6098: }
1.126 brouard 6099: }
1.218 brouard 6100:
1.126 brouard 6101: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6102: for(h=0; h<=nhstepm-1; h++){
6103: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6104: }
1.126 brouard 6105: }/* End theta */
6106:
6107:
6108: for(h=0; h<=nhstepm-1; h++)
6109: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6110: for(theta=1; theta <=npar; theta++)
6111: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6112:
1.218 brouard 6113:
1.222 brouard 6114: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6115: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6116: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6117:
1.222 brouard 6118: printf("%d|",(int)age);fflush(stdout);
6119: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6120: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6121: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6122: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6123: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6124: for(ij=1;ij<=nlstate*nlstate;ij++)
6125: for(ji=1;ji<=nlstate*nlstate;ji++)
6126: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6127: }
6128: }
1.320 brouard 6129: /* if((int)age ==50){ */
6130: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6131: /* } */
1.126 brouard 6132: /* Computing expectancies */
1.235 brouard 6133: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6134: for(i=1; i<=nlstate;i++)
6135: for(j=1; j<=nlstate;j++)
1.222 brouard 6136: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6137: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6138:
1.222 brouard 6139: /* 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 6140:
1.222 brouard 6141: }
1.269 brouard 6142:
6143: /* Standard deviation of expectancies ij */
1.126 brouard 6144: fprintf(ficresstdeij,"%3.0f",age );
6145: for(i=1; i<=nlstate;i++){
6146: eip=0.;
6147: vip=0.;
6148: for(j=1; j<=nlstate;j++){
1.222 brouard 6149: eip += eij[i][j][(int)age];
6150: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6151: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6152: 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 6153: }
6154: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6155: }
6156: fprintf(ficresstdeij,"\n");
1.218 brouard 6157:
1.269 brouard 6158: /* Variance of expectancies ij */
1.126 brouard 6159: fprintf(ficrescveij,"%3.0f",age );
6160: for(i=1; i<=nlstate;i++)
6161: for(j=1; j<=nlstate;j++){
1.222 brouard 6162: cptj= (j-1)*nlstate+i;
6163: for(i2=1; i2<=nlstate;i2++)
6164: for(j2=1; j2<=nlstate;j2++){
6165: cptj2= (j2-1)*nlstate+i2;
6166: if(cptj2 <= cptj)
6167: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6168: }
1.126 brouard 6169: }
6170: fprintf(ficrescveij,"\n");
1.218 brouard 6171:
1.126 brouard 6172: }
6173: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6174: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6175: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6176: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6177: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6178: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6179: printf("\n");
6180: fprintf(ficlog,"\n");
1.218 brouard 6181:
1.126 brouard 6182: free_vector(xm,1,npar);
6183: free_vector(xp,1,npar);
6184: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6185: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6186: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6187: }
1.218 brouard 6188:
1.126 brouard 6189: /************ Variance ******************/
1.235 brouard 6190: 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 6191: {
1.279 brouard 6192: /** Variance of health expectancies
6193: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6194: * double **newm;
6195: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6196: */
1.218 brouard 6197:
6198: /* int movingaverage(); */
6199: double **dnewm,**doldm;
6200: double **dnewmp,**doldmp;
6201: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6202: int first=0;
1.218 brouard 6203: int k;
6204: double *xp;
1.279 brouard 6205: double **gp, **gm; /**< for var eij */
6206: double ***gradg, ***trgradg; /**< for var eij */
6207: double **gradgp, **trgradgp; /**< for var p point j */
6208: double *gpp, *gmp; /**< for var p point j */
6209: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6210: double ***p3mat;
6211: double age,agelim, hf;
6212: /* double ***mobaverage; */
6213: int theta;
6214: char digit[4];
6215: char digitp[25];
6216:
6217: char fileresprobmorprev[FILENAMELENGTH];
6218:
6219: if(popbased==1){
6220: if(mobilav!=0)
6221: strcpy(digitp,"-POPULBASED-MOBILAV_");
6222: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6223: }
6224: else
6225: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6226:
1.218 brouard 6227: /* if (mobilav!=0) { */
6228: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6229: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6230: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6231: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6232: /* } */
6233: /* } */
6234:
6235: strcpy(fileresprobmorprev,"PRMORPREV-");
6236: sprintf(digit,"%-d",ij);
6237: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6238: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6239: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6240: strcat(fileresprobmorprev,fileresu);
6241: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6242: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6243: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6244: }
6245: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6246: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6247: pstamp(ficresprobmorprev);
6248: 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 6249: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
6250: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
6251: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
6252: }
6253: for(j=1;j<=cptcoveff;j++)
6254: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
6255: fprintf(ficresprobmorprev,"\n");
6256:
1.218 brouard 6257: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6258: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6259: fprintf(ficresprobmorprev," p.%-d SE",j);
6260: for(i=1; i<=nlstate;i++)
6261: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6262: }
6263: fprintf(ficresprobmorprev,"\n");
6264:
6265: fprintf(ficgp,"\n# Routine varevsij");
6266: fprintf(ficgp,"\nunset title \n");
6267: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6268: 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");
6269: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6270:
1.218 brouard 6271: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6272: pstamp(ficresvij);
6273: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6274: if(popbased==1)
6275: 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);
6276: else
6277: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6278: fprintf(ficresvij,"# Age");
6279: for(i=1; i<=nlstate;i++)
6280: for(j=1; j<=nlstate;j++)
6281: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6282: fprintf(ficresvij,"\n");
6283:
6284: xp=vector(1,npar);
6285: dnewm=matrix(1,nlstate,1,npar);
6286: doldm=matrix(1,nlstate,1,nlstate);
6287: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6288: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6289:
6290: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6291: gpp=vector(nlstate+1,nlstate+ndeath);
6292: gmp=vector(nlstate+1,nlstate+ndeath);
6293: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6294:
1.218 brouard 6295: if(estepm < stepm){
6296: printf ("Problem %d lower than %d\n",estepm, stepm);
6297: }
6298: else hstepm=estepm;
6299: /* For example we decided to compute the life expectancy with the smallest unit */
6300: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6301: nhstepm is the number of hstepm from age to agelim
6302: nstepm is the number of stepm from age to agelim.
6303: Look at function hpijx to understand why because of memory size limitations,
6304: we decided (b) to get a life expectancy respecting the most precise curvature of the
6305: survival function given by stepm (the optimization length). Unfortunately it
6306: means that if the survival funtion is printed every two years of age and if
6307: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6308: results. So we changed our mind and took the option of the best precision.
6309: */
6310: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6311: agelim = AGESUP;
6312: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6313: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6314: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6315: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6316: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6317: gp=matrix(0,nhstepm,1,nlstate);
6318: gm=matrix(0,nhstepm,1,nlstate);
6319:
6320:
6321: for(theta=1; theta <=npar; theta++){
6322: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6323: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6324: }
1.279 brouard 6325: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6326: * returns into prlim .
1.288 brouard 6327: */
1.242 brouard 6328: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6329:
6330: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6331: if (popbased==1) {
6332: if(mobilav ==0){
6333: for(i=1; i<=nlstate;i++)
6334: prlim[i][i]=probs[(int)age][i][ij];
6335: }else{ /* mobilav */
6336: for(i=1; i<=nlstate;i++)
6337: prlim[i][i]=mobaverage[(int)age][i][ij];
6338: }
6339: }
1.295 brouard 6340: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6341: */
6342: 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 6343: /**< 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 6344: * at horizon h in state j including mortality.
6345: */
1.218 brouard 6346: for(j=1; j<= nlstate; j++){
6347: for(h=0; h<=nhstepm; h++){
6348: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6349: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6350: }
6351: }
1.279 brouard 6352: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6353: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6354: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6355: */
6356: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6357: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6358: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6359: }
6360:
6361: /* Again with minus shift */
1.218 brouard 6362:
6363: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6364: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6365:
1.242 brouard 6366: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6367:
6368: if (popbased==1) {
6369: if(mobilav ==0){
6370: for(i=1; i<=nlstate;i++)
6371: prlim[i][i]=probs[(int)age][i][ij];
6372: }else{ /* mobilav */
6373: for(i=1; i<=nlstate;i++)
6374: prlim[i][i]=mobaverage[(int)age][i][ij];
6375: }
6376: }
6377:
1.235 brouard 6378: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6379:
6380: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6381: for(h=0; h<=nhstepm; h++){
6382: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6383: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6384: }
6385: }
6386: /* This for computing probability of death (h=1 means
6387: computed over hstepm matrices product = hstepm*stepm months)
6388: as a weighted average of prlim.
6389: */
6390: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6391: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6392: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6393: }
1.279 brouard 6394: /* end shifting computations */
6395:
6396: /**< Computing gradient matrix at horizon h
6397: */
1.218 brouard 6398: for(j=1; j<= nlstate; j++) /* vareij */
6399: for(h=0; h<=nhstepm; h++){
6400: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6401: }
1.279 brouard 6402: /**< Gradient of overall mortality p.3 (or p.j)
6403: */
6404: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6405: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6406: }
6407:
6408: } /* End theta */
1.279 brouard 6409:
6410: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6411: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6412:
6413: for(h=0; h<=nhstepm; h++) /* veij */
6414: for(j=1; j<=nlstate;j++)
6415: for(theta=1; theta <=npar; theta++)
6416: trgradg[h][j][theta]=gradg[h][theta][j];
6417:
6418: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6419: for(theta=1; theta <=npar; theta++)
6420: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6421: /**< as well as its transposed matrix
6422: */
1.218 brouard 6423:
6424: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6425: for(i=1;i<=nlstate;i++)
6426: for(j=1;j<=nlstate;j++)
6427: vareij[i][j][(int)age] =0.;
1.279 brouard 6428:
6429: /* Computing trgradg by matcov by gradg at age and summing over h
6430: * and k (nhstepm) formula 15 of article
6431: * Lievre-Brouard-Heathcote
6432: */
6433:
1.218 brouard 6434: for(h=0;h<=nhstepm;h++){
6435: for(k=0;k<=nhstepm;k++){
6436: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6437: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6438: for(i=1;i<=nlstate;i++)
6439: for(j=1;j<=nlstate;j++)
6440: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6441: }
6442: }
6443:
1.279 brouard 6444: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6445: * p.j overall mortality formula 49 but computed directly because
6446: * we compute the grad (wix pijx) instead of grad (pijx),even if
6447: * wix is independent of theta.
6448: */
1.218 brouard 6449: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6450: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6451: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6452: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6453: varppt[j][i]=doldmp[j][i];
6454: /* end ppptj */
6455: /* x centered again */
6456:
1.242 brouard 6457: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6458:
6459: if (popbased==1) {
6460: if(mobilav ==0){
6461: for(i=1; i<=nlstate;i++)
6462: prlim[i][i]=probs[(int)age][i][ij];
6463: }else{ /* mobilav */
6464: for(i=1; i<=nlstate;i++)
6465: prlim[i][i]=mobaverage[(int)age][i][ij];
6466: }
6467: }
6468:
6469: /* This for computing probability of death (h=1 means
6470: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6471: as a weighted average of prlim.
6472: */
1.235 brouard 6473: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6474: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6475: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6476: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6477: }
6478: /* end probability of death */
6479:
6480: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6481: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6482: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6483: for(i=1; i<=nlstate;i++){
6484: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6485: }
6486: }
6487: fprintf(ficresprobmorprev,"\n");
6488:
6489: fprintf(ficresvij,"%.0f ",age );
6490: for(i=1; i<=nlstate;i++)
6491: for(j=1; j<=nlstate;j++){
6492: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6493: }
6494: fprintf(ficresvij,"\n");
6495: free_matrix(gp,0,nhstepm,1,nlstate);
6496: free_matrix(gm,0,nhstepm,1,nlstate);
6497: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6498: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6499: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6500: } /* End age */
6501: free_vector(gpp,nlstate+1,nlstate+ndeath);
6502: free_vector(gmp,nlstate+1,nlstate+ndeath);
6503: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6504: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6505: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6506: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6507: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6508: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6509: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6510: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6511: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6512: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6513: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6514: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6515: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6516: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6517: 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);
6518: /* 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 6519: */
1.218 brouard 6520: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6521: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6522:
1.218 brouard 6523: free_vector(xp,1,npar);
6524: free_matrix(doldm,1,nlstate,1,nlstate);
6525: free_matrix(dnewm,1,nlstate,1,npar);
6526: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6527: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6528: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6529: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6530: fclose(ficresprobmorprev);
6531: fflush(ficgp);
6532: fflush(fichtm);
6533: } /* end varevsij */
1.126 brouard 6534:
6535: /************ Variance of prevlim ******************/
1.269 brouard 6536: 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 6537: {
1.205 brouard 6538: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6539: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6540:
1.268 brouard 6541: double **dnewmpar,**doldm;
1.126 brouard 6542: int i, j, nhstepm, hstepm;
6543: double *xp;
6544: double *gp, *gm;
6545: double **gradg, **trgradg;
1.208 brouard 6546: double **mgm, **mgp;
1.126 brouard 6547: double age,agelim;
6548: int theta;
6549:
6550: pstamp(ficresvpl);
1.288 brouard 6551: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6552: fprintf(ficresvpl,"# Age ");
6553: if(nresult >=1)
6554: fprintf(ficresvpl," Result# ");
1.126 brouard 6555: for(i=1; i<=nlstate;i++)
6556: fprintf(ficresvpl," %1d-%1d",i,i);
6557: fprintf(ficresvpl,"\n");
6558:
6559: xp=vector(1,npar);
1.268 brouard 6560: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6561: doldm=matrix(1,nlstate,1,nlstate);
6562:
6563: hstepm=1*YEARM; /* Every year of age */
6564: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6565: agelim = AGESUP;
6566: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6567: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6568: if (stepm >= YEARM) hstepm=1;
6569: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6570: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6571: mgp=matrix(1,npar,1,nlstate);
6572: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6573: gp=vector(1,nlstate);
6574: gm=vector(1,nlstate);
6575:
6576: for(theta=1; theta <=npar; theta++){
6577: for(i=1; i<=npar; i++){ /* Computes gradient */
6578: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6579: }
1.288 brouard 6580: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6581: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6582: /* else */
6583: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6584: for(i=1;i<=nlstate;i++){
1.126 brouard 6585: gp[i] = prlim[i][i];
1.208 brouard 6586: mgp[theta][i] = prlim[i][i];
6587: }
1.126 brouard 6588: for(i=1; i<=npar; i++) /* Computes gradient */
6589: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6590: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6591: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6592: /* else */
6593: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6594: for(i=1;i<=nlstate;i++){
1.126 brouard 6595: gm[i] = prlim[i][i];
1.208 brouard 6596: mgm[theta][i] = prlim[i][i];
6597: }
1.126 brouard 6598: for(i=1;i<=nlstate;i++)
6599: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6600: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6601: } /* End theta */
6602:
6603: trgradg =matrix(1,nlstate,1,npar);
6604:
6605: for(j=1; j<=nlstate;j++)
6606: for(theta=1; theta <=npar; theta++)
6607: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6608: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6609: /* printf("\nmgm mgp %d ",(int)age); */
6610: /* for(j=1; j<=nlstate;j++){ */
6611: /* printf(" %d ",j); */
6612: /* for(theta=1; theta <=npar; theta++) */
6613: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6614: /* printf("\n "); */
6615: /* } */
6616: /* } */
6617: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6618: /* printf("\n gradg %d ",(int)age); */
6619: /* for(j=1; j<=nlstate;j++){ */
6620: /* printf("%d ",j); */
6621: /* for(theta=1; theta <=npar; theta++) */
6622: /* printf("%d %lf ",theta,gradg[theta][j]); */
6623: /* printf("\n "); */
6624: /* } */
6625: /* } */
1.126 brouard 6626:
6627: for(i=1;i<=nlstate;i++)
6628: varpl[i][(int)age] =0.;
1.209 brouard 6629: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6630: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6631: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6632: }else{
1.268 brouard 6633: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6634: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6635: }
1.126 brouard 6636: for(i=1;i<=nlstate;i++)
6637: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6638:
6639: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6640: if(nresult >=1)
6641: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6642: for(i=1; i<=nlstate;i++){
1.126 brouard 6643: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6644: /* for(j=1;j<=nlstate;j++) */
6645: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6646: }
1.126 brouard 6647: fprintf(ficresvpl,"\n");
6648: free_vector(gp,1,nlstate);
6649: free_vector(gm,1,nlstate);
1.208 brouard 6650: free_matrix(mgm,1,npar,1,nlstate);
6651: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6652: free_matrix(gradg,1,npar,1,nlstate);
6653: free_matrix(trgradg,1,nlstate,1,npar);
6654: } /* End age */
6655:
6656: free_vector(xp,1,npar);
6657: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6658: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6659:
6660: }
6661:
6662:
6663: /************ Variance of backprevalence limit ******************/
1.269 brouard 6664: 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 6665: {
6666: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6667: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6668:
6669: double **dnewmpar,**doldm;
6670: int i, j, nhstepm, hstepm;
6671: double *xp;
6672: double *gp, *gm;
6673: double **gradg, **trgradg;
6674: double **mgm, **mgp;
6675: double age,agelim;
6676: int theta;
6677:
6678: pstamp(ficresvbl);
6679: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6680: fprintf(ficresvbl,"# Age ");
6681: if(nresult >=1)
6682: fprintf(ficresvbl," Result# ");
6683: for(i=1; i<=nlstate;i++)
6684: fprintf(ficresvbl," %1d-%1d",i,i);
6685: fprintf(ficresvbl,"\n");
6686:
6687: xp=vector(1,npar);
6688: dnewmpar=matrix(1,nlstate,1,npar);
6689: doldm=matrix(1,nlstate,1,nlstate);
6690:
6691: hstepm=1*YEARM; /* Every year of age */
6692: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6693: agelim = AGEINF;
6694: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6695: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6696: if (stepm >= YEARM) hstepm=1;
6697: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6698: gradg=matrix(1,npar,1,nlstate);
6699: mgp=matrix(1,npar,1,nlstate);
6700: mgm=matrix(1,npar,1,nlstate);
6701: gp=vector(1,nlstate);
6702: gm=vector(1,nlstate);
6703:
6704: for(theta=1; theta <=npar; theta++){
6705: for(i=1; i<=npar; i++){ /* Computes gradient */
6706: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6707: }
6708: if(mobilavproj > 0 )
6709: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6710: else
6711: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6712: for(i=1;i<=nlstate;i++){
6713: gp[i] = bprlim[i][i];
6714: mgp[theta][i] = bprlim[i][i];
6715: }
6716: for(i=1; i<=npar; i++) /* Computes gradient */
6717: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6718: if(mobilavproj > 0 )
6719: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6720: else
6721: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6722: for(i=1;i<=nlstate;i++){
6723: gm[i] = bprlim[i][i];
6724: mgm[theta][i] = bprlim[i][i];
6725: }
6726: for(i=1;i<=nlstate;i++)
6727: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6728: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6729: } /* End theta */
6730:
6731: trgradg =matrix(1,nlstate,1,npar);
6732:
6733: for(j=1; j<=nlstate;j++)
6734: for(theta=1; theta <=npar; theta++)
6735: trgradg[j][theta]=gradg[theta][j];
6736: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6737: /* printf("\nmgm mgp %d ",(int)age); */
6738: /* for(j=1; j<=nlstate;j++){ */
6739: /* printf(" %d ",j); */
6740: /* for(theta=1; theta <=npar; theta++) */
6741: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6742: /* printf("\n "); */
6743: /* } */
6744: /* } */
6745: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6746: /* printf("\n gradg %d ",(int)age); */
6747: /* for(j=1; j<=nlstate;j++){ */
6748: /* printf("%d ",j); */
6749: /* for(theta=1; theta <=npar; theta++) */
6750: /* printf("%d %lf ",theta,gradg[theta][j]); */
6751: /* printf("\n "); */
6752: /* } */
6753: /* } */
6754:
6755: for(i=1;i<=nlstate;i++)
6756: varbpl[i][(int)age] =0.;
6757: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6758: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6759: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6760: }else{
6761: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6762: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6763: }
6764: for(i=1;i<=nlstate;i++)
6765: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6766:
6767: fprintf(ficresvbl,"%.0f ",age );
6768: if(nresult >=1)
6769: fprintf(ficresvbl,"%d ",nres );
6770: for(i=1; i<=nlstate;i++)
6771: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6772: fprintf(ficresvbl,"\n");
6773: free_vector(gp,1,nlstate);
6774: free_vector(gm,1,nlstate);
6775: free_matrix(mgm,1,npar,1,nlstate);
6776: free_matrix(mgp,1,npar,1,nlstate);
6777: free_matrix(gradg,1,npar,1,nlstate);
6778: free_matrix(trgradg,1,nlstate,1,npar);
6779: } /* End age */
6780:
6781: free_vector(xp,1,npar);
6782: free_matrix(doldm,1,nlstate,1,npar);
6783: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6784:
6785: }
6786:
6787: /************ Variance of one-step probabilities ******************/
6788: 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 6789: {
6790: int i, j=0, k1, l1, tj;
6791: int k2, l2, j1, z1;
6792: int k=0, l;
6793: int first=1, first1, first2;
6794: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6795: double **dnewm,**doldm;
6796: double *xp;
6797: double *gp, *gm;
6798: double **gradg, **trgradg;
6799: double **mu;
6800: double age, cov[NCOVMAX+1];
6801: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6802: int theta;
6803: char fileresprob[FILENAMELENGTH];
6804: char fileresprobcov[FILENAMELENGTH];
6805: char fileresprobcor[FILENAMELENGTH];
6806: double ***varpij;
6807:
6808: strcpy(fileresprob,"PROB_");
6809: strcat(fileresprob,fileres);
6810: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6811: printf("Problem with resultfile: %s\n", fileresprob);
6812: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6813: }
6814: strcpy(fileresprobcov,"PROBCOV_");
6815: strcat(fileresprobcov,fileresu);
6816: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6817: printf("Problem with resultfile: %s\n", fileresprobcov);
6818: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6819: }
6820: strcpy(fileresprobcor,"PROBCOR_");
6821: strcat(fileresprobcor,fileresu);
6822: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6823: printf("Problem with resultfile: %s\n", fileresprobcor);
6824: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6825: }
6826: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6827: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6828: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6829: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6830: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6831: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6832: pstamp(ficresprob);
6833: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6834: fprintf(ficresprob,"# Age");
6835: pstamp(ficresprobcov);
6836: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6837: fprintf(ficresprobcov,"# Age");
6838: pstamp(ficresprobcor);
6839: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6840: fprintf(ficresprobcor,"# Age");
1.126 brouard 6841:
6842:
1.222 brouard 6843: for(i=1; i<=nlstate;i++)
6844: for(j=1; j<=(nlstate+ndeath);j++){
6845: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6846: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6847: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6848: }
6849: /* fprintf(ficresprob,"\n");
6850: fprintf(ficresprobcov,"\n");
6851: fprintf(ficresprobcor,"\n");
6852: */
6853: xp=vector(1,npar);
6854: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6855: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6856: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6857: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6858: first=1;
6859: fprintf(ficgp,"\n# Routine varprob");
6860: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6861: fprintf(fichtm,"\n");
6862:
1.288 brouard 6863: 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 6864: 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);
6865: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6866: and drawn. It helps understanding how is the covariance between two incidences.\
6867: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6868: 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 6869: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6870: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6871: standard deviations wide on each axis. <br>\
6872: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6873: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6874: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6875:
1.222 brouard 6876: cov[1]=1;
6877: /* tj=cptcoveff; */
1.225 brouard 6878: tj = (int) pow(2,cptcoveff);
1.222 brouard 6879: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6880: j1=0;
1.224 brouard 6881: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6882: if (cptcovn>0) {
6883: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6884: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6885: fprintf(ficresprob, "**********\n#\n");
6886: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6887: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6888: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6889:
1.222 brouard 6890: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6891: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6892: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6893:
6894:
1.222 brouard 6895: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.319 brouard 6896: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]); */
6897: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtmcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6898: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6899:
1.222 brouard 6900: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6901: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6902: fprintf(ficresprobcor, "**********\n#");
6903: if(invalidvarcomb[j1]){
6904: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6905: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6906: continue;
6907: }
6908: }
6909: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6910: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6911: gp=vector(1,(nlstate)*(nlstate+ndeath));
6912: gm=vector(1,(nlstate)*(nlstate+ndeath));
6913: for (age=bage; age<=fage; age ++){
6914: cov[2]=age;
6915: if(nagesqr==1)
6916: cov[3]= age*age;
6917: for (k=1; k<=cptcovn;k++) {
6918: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6919: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6920: * 1 1 1 1 1
6921: * 2 2 1 1 1
6922: * 3 1 2 1 1
6923: */
6924: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6925: }
1.319 brouard 6926: /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 */
6927: /* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] */
6928: /*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6929: for (k=1; k<=cptcovage;k++)
6930: cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.222 brouard 6931: for (k=1; k<=cptcovprod;k++)
6932: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6933:
6934:
1.222 brouard 6935: for(theta=1; theta <=npar; theta++){
6936: for(i=1; i<=npar; i++)
6937: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6938:
1.222 brouard 6939: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6940:
1.222 brouard 6941: k=0;
6942: for(i=1; i<= (nlstate); i++){
6943: for(j=1; j<=(nlstate+ndeath);j++){
6944: k=k+1;
6945: gp[k]=pmmij[i][j];
6946: }
6947: }
1.220 brouard 6948:
1.222 brouard 6949: for(i=1; i<=npar; i++)
6950: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6951:
1.222 brouard 6952: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6953: k=0;
6954: for(i=1; i<=(nlstate); i++){
6955: for(j=1; j<=(nlstate+ndeath);j++){
6956: k=k+1;
6957: gm[k]=pmmij[i][j];
6958: }
6959: }
1.220 brouard 6960:
1.222 brouard 6961: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6962: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6963: }
1.126 brouard 6964:
1.222 brouard 6965: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6966: for(theta=1; theta <=npar; theta++)
6967: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6968:
1.222 brouard 6969: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6970: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6971:
1.222 brouard 6972: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6973:
1.222 brouard 6974: k=0;
6975: for(i=1; i<=(nlstate); i++){
6976: for(j=1; j<=(nlstate+ndeath);j++){
6977: k=k+1;
6978: mu[k][(int) age]=pmmij[i][j];
6979: }
6980: }
6981: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6982: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6983: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6984:
1.222 brouard 6985: /*printf("\n%d ",(int)age);
6986: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6987: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6988: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6989: }*/
1.220 brouard 6990:
1.222 brouard 6991: fprintf(ficresprob,"\n%d ",(int)age);
6992: fprintf(ficresprobcov,"\n%d ",(int)age);
6993: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6994:
1.222 brouard 6995: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6996: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6997: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6998: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6999: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7000: }
7001: i=0;
7002: for (k=1; k<=(nlstate);k++){
7003: for (l=1; l<=(nlstate+ndeath);l++){
7004: i++;
7005: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7006: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7007: for (j=1; j<=i;j++){
7008: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7009: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7010: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7011: }
7012: }
7013: }/* end of loop for state */
7014: } /* end of loop for age */
7015: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7016: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7017: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7018: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7019:
7020: /* Confidence intervalle of pij */
7021: /*
7022: fprintf(ficgp,"\nunset parametric;unset label");
7023: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7024: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7025: 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);
7026: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7027: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7028: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7029: */
7030:
7031: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7032: first1=1;first2=2;
7033: for (k2=1; k2<=(nlstate);k2++){
7034: for (l2=1; l2<=(nlstate+ndeath);l2++){
7035: if(l2==k2) continue;
7036: j=(k2-1)*(nlstate+ndeath)+l2;
7037: for (k1=1; k1<=(nlstate);k1++){
7038: for (l1=1; l1<=(nlstate+ndeath);l1++){
7039: if(l1==k1) continue;
7040: i=(k1-1)*(nlstate+ndeath)+l1;
7041: if(i<=j) continue;
7042: for (age=bage; age<=fage; age ++){
7043: if ((int)age %5==0){
7044: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7045: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7046: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7047: mu1=mu[i][(int) age]/stepm*YEARM ;
7048: mu2=mu[j][(int) age]/stepm*YEARM;
7049: c12=cv12/sqrt(v1*v2);
7050: /* Computing eigen value of matrix of covariance */
7051: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7052: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7053: if ((lc2 <0) || (lc1 <0) ){
7054: if(first2==1){
7055: first1=0;
7056: 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);
7057: }
7058: 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);
7059: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7060: /* lc2=fabs(lc2); */
7061: }
1.220 brouard 7062:
1.222 brouard 7063: /* Eigen vectors */
1.280 brouard 7064: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7065: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7066: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7067: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7068: }else
7069: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7070: /*v21=sqrt(1.-v11*v11); *//* error */
7071: v21=(lc1-v1)/cv12*v11;
7072: v12=-v21;
7073: v22=v11;
7074: tnalp=v21/v11;
7075: if(first1==1){
7076: first1=0;
7077: 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);
7078: }
7079: 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);
7080: /*printf(fignu*/
7081: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7082: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7083: if(first==1){
7084: first=0;
7085: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7086: fprintf(ficgp,"\nset parametric;unset label");
7087: 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);
7088: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7089: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7090: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7091: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7092: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7093: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7094: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7095: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7096: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7097: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7098: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7099: 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 7100: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7101: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7102: }else{
7103: first=0;
7104: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7105: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7106: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7107: 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 7108: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7109: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7110: }/* if first */
7111: } /* age mod 5 */
7112: } /* end loop age */
7113: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7114: first=1;
7115: } /*l12 */
7116: } /* k12 */
7117: } /*l1 */
7118: }/* k1 */
7119: } /* loop on combination of covariates j1 */
7120: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7121: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7122: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7123: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7124: free_vector(xp,1,npar);
7125: fclose(ficresprob);
7126: fclose(ficresprobcov);
7127: fclose(ficresprobcor);
7128: fflush(ficgp);
7129: fflush(fichtmcov);
7130: }
1.126 brouard 7131:
7132:
7133: /******************* Printing html file ***********/
1.201 brouard 7134: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7135: int lastpass, int stepm, int weightopt, char model[],\
7136: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7137: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7138: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7139: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7140: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7141: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7142: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7143: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7144: </ul>");
1.319 brouard 7145: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7146: /* </ul>", model); */
1.214 brouard 7147: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7148: 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",
7149: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
7150: fprintf(fichtm,"<li> - Observed 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 7151: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7152: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7153: fprintf(fichtm,"\
7154: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7155: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7156: fprintf(fichtm,"\
1.217 brouard 7157: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7158: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7159: fprintf(fichtm,"\
1.288 brouard 7160: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7161: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7162: fprintf(fichtm,"\
1.288 brouard 7163: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7164: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7165: fprintf(fichtm,"\
1.211 brouard 7166: - (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 7167: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7168: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7169: if(prevfcast==1){
7170: fprintf(fichtm,"\
7171: - Prevalence projections by age and states: \
1.201 brouard 7172: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7173: }
1.126 brouard 7174:
7175:
1.225 brouard 7176: m=pow(2,cptcoveff);
1.222 brouard 7177: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7178:
1.317 brouard 7179: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7180:
7181: jj1=0;
7182:
7183: fprintf(fichtm," \n<ul>");
7184: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7185: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7186: if(m != 1 && TKresult[nres]!= k1)
7187: continue;
7188: jj1++;
7189: if (cptcovn > 0) {
7190: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
7191: for (cpt=1; cpt<=cptcoveff;cpt++){
7192: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7193: }
7194: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7195: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7196: }
7197: fprintf(fichtm,"\">");
7198:
7199: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7200: fprintf(fichtm,"************ Results for covariates");
7201: for (cpt=1; cpt<=cptcoveff;cpt++){
7202: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7203: }
7204: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7205: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7206: }
7207: if(invalidvarcomb[k1]){
7208: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7209: continue;
7210: }
7211: fprintf(fichtm,"</a></li>");
7212: } /* cptcovn >0 */
7213: }
1.317 brouard 7214: fprintf(fichtm," \n</ul>");
1.264 brouard 7215:
1.222 brouard 7216: jj1=0;
1.237 brouard 7217:
7218: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 7219: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 7220: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7221: continue;
1.220 brouard 7222:
1.222 brouard 7223: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7224: jj1++;
7225: if (cptcovn > 0) {
1.264 brouard 7226: fprintf(fichtm,"\n<p><a name=\"rescov");
7227: for (cpt=1; cpt<=cptcoveff;cpt++){
7228: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7229: }
7230: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7231: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7232: }
7233: fprintf(fichtm,"\"</a>");
7234:
1.222 brouard 7235: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7236: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 7237: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7238: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
7239: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7240: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7241: }
1.237 brouard 7242: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7243: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7244: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
7245: }
7246:
1.230 brouard 7247: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.321 brouard 7248: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7249: if(invalidvarcomb[k1]){
7250: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7251: printf("\nCombination (%d) ignored because no cases \n",k1);
7252: continue;
7253: }
7254: }
7255: /* aij, bij */
1.259 brouard 7256: 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 7257: <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 7258: /* Pij */
1.241 brouard 7259: 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> \
7260: <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 7261: /* Quasi-incidences */
7262: 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 7263: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7264: 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 7265: 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> \
7266: <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 7267: /* Survival functions (period) in state j */
7268: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7269: 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> \
1.241 brouard 7270: <img src=\"%s_%d-%d-%d.svg\">", cpt, cpt, nlstate, subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7271: }
7272: /* State specific survival functions (period) */
7273: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7274: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7275: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 7276: <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> <img src=\"%s_%d-%d-%d.svg\">", cpt, nlstate, cpt, subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7277: }
1.288 brouard 7278: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7279: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7280: 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> \
7281: <img src=\"%s_%d-%d-%d.svg\">", cpt, nlstate, cpt, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7282: }
1.296 brouard 7283: if(prevbcast==1){
1.288 brouard 7284: /* Backward prevalence in each health state */
1.222 brouard 7285: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7286: 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> \
1.241 brouard 7287: <img src=\"%s_%d-%d-%d.svg\">", cpt, cpt, nlstate, subdirf2(optionfilefiname,"PB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.222 brouard 7288: }
1.217 brouard 7289: }
1.222 brouard 7290: if(prevfcast==1){
1.288 brouard 7291: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7292: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7293: 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);
7294: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7295: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7296: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7297: }
7298: }
1.296 brouard 7299: if(prevbcast==1){
1.268 brouard 7300: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7301: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7302: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7303: 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 \
7304: 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 7305: 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);
7306: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7307: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7308: }
7309: }
1.220 brouard 7310:
1.222 brouard 7311: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7312: 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);
7313: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7314: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7315: }
7316: /* } /\* end i1 *\/ */
7317: }/* End k1 */
7318: fprintf(fichtm,"</ul>");
1.126 brouard 7319:
1.222 brouard 7320: fprintf(fichtm,"\
1.126 brouard 7321: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7322: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7323: - 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 7324: But because parameters are usually highly correlated (a higher incidence of disability \
7325: and a higher incidence of recovery can give very close observed transition) it might \
7326: be very useful to look not only at linear confidence intervals estimated from the \
7327: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7328: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7329: covariance matrix of the one-step probabilities. \
7330: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7331:
1.222 brouard 7332: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7333: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7334: fprintf(fichtm,"\
1.126 brouard 7335: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7336: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7337:
1.222 brouard 7338: fprintf(fichtm,"\
1.126 brouard 7339: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7340: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7341: fprintf(fichtm,"\
1.126 brouard 7342: - 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): \
7343: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7344: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7345: fprintf(fichtm,"\
1.126 brouard 7346: - (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): \
7347: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7348: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7349: fprintf(fichtm,"\
1.288 brouard 7350: - 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 7351: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7352: fprintf(fichtm,"\
1.128 brouard 7353: - 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 7354: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7355: fprintf(fichtm,"\
1.288 brouard 7356: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7357: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7358:
7359: /* if(popforecast==1) fprintf(fichtm,"\n */
7360: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7361: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7362: /* <br>",fileres,fileres,fileres,fileres); */
7363: /* else */
7364: /* fprintf(fichtm,"\n No population forecast: popforecast = %d (instead of 1) or stepm = %d (instead of 1) or model=%s (instead of .)<br><br></li>\n",popforecast, stepm, model); */
1.222 brouard 7365: fflush(fichtm);
1.126 brouard 7366:
1.225 brouard 7367: m=pow(2,cptcoveff);
1.222 brouard 7368: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7369:
1.317 brouard 7370: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
7371:
7372: jj1=0;
7373:
7374: fprintf(fichtm," \n<ul>");
7375: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7376: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7377: if(m != 1 && TKresult[nres]!= k1)
7378: continue;
7379: jj1++;
7380: if (cptcovn > 0) {
7381: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
7382: for (cpt=1; cpt<=cptcoveff;cpt++){
7383: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7384: }
7385: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7386: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7387: }
7388: fprintf(fichtm,"\">");
7389:
7390: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7391: fprintf(fichtm,"************ Results for covariates");
7392: for (cpt=1; cpt<=cptcoveff;cpt++){
7393: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7394: }
7395: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7396: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7397: }
7398: if(invalidvarcomb[k1]){
7399: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7400: continue;
7401: }
7402: fprintf(fichtm,"</a></li>");
7403: } /* cptcovn >0 */
7404: }
7405: fprintf(fichtm," \n</ul>");
7406:
1.222 brouard 7407: jj1=0;
1.237 brouard 7408:
1.241 brouard 7409: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7410: for(k1=1; k1<=m;k1++){
1.253 brouard 7411: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7412: continue;
1.222 brouard 7413: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7414: jj1++;
1.126 brouard 7415: if (cptcovn > 0) {
1.317 brouard 7416: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
7417: for (cpt=1; cpt<=cptcoveff;cpt++){
7418: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7419: }
7420: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7421: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7422: }
7423: fprintf(fichtm,"\"</a>");
7424:
1.126 brouard 7425: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.317 brouard 7426: for (cpt=1; cpt<=cptcoveff;cpt++){ /**< cptcoveff number of variables */
1.237 brouard 7427: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
1.317 brouard 7428: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
1.237 brouard 7429: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 7430: }
1.237 brouard 7431: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7432: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7433: }
7434:
1.321 brouard 7435: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 7436:
1.222 brouard 7437: if(invalidvarcomb[k1]){
7438: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7439: continue;
7440: }
1.126 brouard 7441: }
7442: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7443: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 7444: 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);
7445: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
7446: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 7447: }
7448: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 7449: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 7450: true period expectancies (those weighted with period prevalences are also\
7451: drawn in addition to the population based expectancies computed using\
1.314 brouard 7452: 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);
7453: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
7454: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7455: /* } /\* end i1 *\/ */
7456: }/* End k1 */
1.241 brouard 7457: }/* End nres */
1.222 brouard 7458: fprintf(fichtm,"</ul>");
7459: fflush(fichtm);
1.126 brouard 7460: }
7461:
7462: /******************* Gnuplot file **************/
1.296 brouard 7463: 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 7464:
7465: char dirfileres[132],optfileres[132];
1.264 brouard 7466: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7467: 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 7468: int lv=0, vlv=0, kl=0;
1.130 brouard 7469: int ng=0;
1.201 brouard 7470: int vpopbased;
1.223 brouard 7471: int ioffset; /* variable offset for columns */
1.270 brouard 7472: int iyearc=1; /* variable column for year of projection */
7473: int iagec=1; /* variable column for age of projection */
1.235 brouard 7474: int nres=0; /* Index of resultline */
1.266 brouard 7475: int istart=1; /* For starting graphs in projections */
1.219 brouard 7476:
1.126 brouard 7477: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7478: /* printf("Problem with file %s",optionfilegnuplot); */
7479: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7480: /* } */
7481:
7482: /*#ifdef windows */
7483: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7484: /*#endif */
1.225 brouard 7485: m=pow(2,cptcoveff);
1.126 brouard 7486:
1.274 brouard 7487: /* diagram of the model */
7488: fprintf(ficgp,"\n#Diagram of the model \n");
7489: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7490: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7491: 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);
7492:
7493: 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);
7494: fprintf(ficgp,"\n#show arrow\nunset label\n");
7495: 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);
7496: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7497: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7498: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7499: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7500:
1.202 brouard 7501: /* Contribution to likelihood */
7502: /* Plot the probability implied in the likelihood */
1.223 brouard 7503: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7504: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7505: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7506: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7507: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7508: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7509: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7510: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7511: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7512: 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));
7513: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7514: 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));
7515: for (i=1; i<= nlstate ; i ++) {
7516: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7517: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7518: 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);
7519: for (j=2; j<= nlstate+ndeath ; j ++) {
7520: 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);
7521: }
7522: fprintf(ficgp,";\nset out; unset ylabel;\n");
7523: }
7524: /* 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 */
7525: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7526: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7527: fprintf(ficgp,"\nset out;unset log\n");
7528: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7529:
1.126 brouard 7530: strcpy(dirfileres,optionfilefiname);
7531: strcpy(optfileres,"vpl");
1.223 brouard 7532: /* 1eme*/
1.238 brouard 7533: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7534: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7535: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7536: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7537: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7538: continue;
7539: /* We are interested in selected combination by the resultline */
1.246 brouard 7540: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7541: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7542: strcpy(gplotlabel,"(");
1.238 brouard 7543: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7544: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7545: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7546: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7547: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7548: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7549: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7550: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7551: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7552: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7553: }
7554: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7555: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7556: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7557: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7558: }
7559: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7560: /* printf("\n#\n"); */
1.238 brouard 7561: fprintf(ficgp,"\n#\n");
7562: if(invalidvarcomb[k1]){
1.260 brouard 7563: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7564: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7565: continue;
7566: }
1.235 brouard 7567:
1.241 brouard 7568: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7569: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7570: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
1.321 brouard 7571: fprintf(ficgp,"set title \"Alive state %d %s model=%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 7572: 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);
7573: /* 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); */
7574: /* k1-1 error should be nres-1*/
1.238 brouard 7575: for (i=1; i<= nlstate ; i ++) {
7576: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7577: else fprintf(ficgp," %%*lf (%%*lf)");
7578: }
1.288 brouard 7579: 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 7580: for (i=1; i<= nlstate ; i ++) {
7581: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7582: else fprintf(ficgp," %%*lf (%%*lf)");
7583: }
1.260 brouard 7584: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7585: for (i=1; i<= nlstate ; i ++) {
7586: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7587: else fprintf(ficgp," %%*lf (%%*lf)");
7588: }
1.265 brouard 7589: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7590:
7591: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7592: if(cptcoveff ==0){
1.271 brouard 7593: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7594: }else{
7595: kl=0;
7596: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7597: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7598: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7599: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7600: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7601: vlv= nbcode[Tvaraff[k]][lv];
7602: kl++;
7603: /* 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 *\/ */
7604: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7605: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7606: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7607: if(k==cptcoveff){
7608: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7609: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7610: }else{
7611: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7612: kl++;
7613: }
7614: } /* end covariate */
7615: } /* end if no covariate */
7616:
1.296 brouard 7617: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7618: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7619: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7620: if(cptcoveff ==0){
1.245 brouard 7621: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7622: }else{
7623: kl=0;
7624: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7625: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7626: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7627: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7628: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7629: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7630: kl++;
1.238 brouard 7631: /* 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 *\/ */
7632: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7633: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7634: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7635: if(k==cptcoveff){
1.245 brouard 7636: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7637: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7638: }else{
7639: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7640: kl++;
7641: }
7642: } /* end covariate */
7643: } /* end if no covariate */
1.296 brouard 7644: if(prevbcast == 1){
1.268 brouard 7645: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7646: /* k1-1 error should be nres-1*/
7647: for (i=1; i<= nlstate ; i ++) {
7648: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7649: else fprintf(ficgp," %%*lf (%%*lf)");
7650: }
1.271 brouard 7651: 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 7652: for (i=1; i<= nlstate ; i ++) {
7653: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7654: else fprintf(ficgp," %%*lf (%%*lf)");
7655: }
1.276 brouard 7656: 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 7657: for (i=1; i<= nlstate ; i ++) {
7658: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7659: else fprintf(ficgp," %%*lf (%%*lf)");
7660: }
1.274 brouard 7661: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7662: } /* end if backprojcast */
1.296 brouard 7663: } /* end if prevbcast */
1.276 brouard 7664: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7665: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7666: } /* nres */
1.201 brouard 7667: } /* k1 */
7668: } /* cpt */
1.235 brouard 7669:
7670:
1.126 brouard 7671: /*2 eme*/
1.238 brouard 7672: for (k1=1; k1<= m ; k1 ++){
7673: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7674: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7675: continue;
7676: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7677: strcpy(gplotlabel,"(");
1.238 brouard 7678: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7679: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7680: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7681: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7682: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7683: vlv= nbcode[Tvaraff[k]][lv];
7684: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7685: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7686: }
1.237 brouard 7687: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7688: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7689: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7690: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7691: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7692: }
1.264 brouard 7693: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7694: fprintf(ficgp,"\n#\n");
1.223 brouard 7695: if(invalidvarcomb[k1]){
7696: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7697: continue;
7698: }
1.219 brouard 7699:
1.241 brouard 7700: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7701: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7702: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7703: if(vpopbased==0){
1.238 brouard 7704: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7705: }else
1.238 brouard 7706: fprintf(ficgp,"\nreplot ");
7707: for (i=1; i<= nlstate+1 ; i ++) {
7708: k=2*i;
1.261 brouard 7709: 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 7710: for (j=1; j<= nlstate+1 ; j ++) {
7711: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7712: else fprintf(ficgp," %%*lf (%%*lf)");
7713: }
7714: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7715: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7716: 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 7717: for (j=1; j<= nlstate+1 ; j ++) {
7718: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7719: else fprintf(ficgp," %%*lf (%%*lf)");
7720: }
7721: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7722: 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 7723: for (j=1; j<= nlstate+1 ; j ++) {
7724: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7725: else fprintf(ficgp," %%*lf (%%*lf)");
7726: }
7727: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7728: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7729: } /* state */
7730: } /* vpopbased */
1.264 brouard 7731: 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 7732: } /* end nres */
7733: } /* k1 end 2 eme*/
7734:
7735:
7736: /*3eme*/
7737: for (k1=1; k1<= m ; k1 ++){
7738: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7739: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7740: continue;
7741:
7742: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7743: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7744: strcpy(gplotlabel,"(");
1.238 brouard 7745: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7746: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7747: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7748: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7749: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7750: vlv= nbcode[Tvaraff[k]][lv];
7751: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7752: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7753: }
7754: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7755: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7756: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7757: }
1.264 brouard 7758: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7759: fprintf(ficgp,"\n#\n");
7760: if(invalidvarcomb[k1]){
7761: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7762: continue;
7763: }
7764:
7765: /* k=2+nlstate*(2*cpt-2); */
7766: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7767: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7768: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7769: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7770: 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 7771: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7772: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7773: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7774: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7775: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7776: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7777:
1.238 brouard 7778: */
7779: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7780: 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 7781: /* 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 7782:
1.238 brouard 7783: }
1.261 brouard 7784: 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 7785: }
1.264 brouard 7786: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7787: } /* end nres */
7788: } /* end kl 3eme */
1.126 brouard 7789:
1.223 brouard 7790: /* 4eme */
1.201 brouard 7791: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7792: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7793: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7794: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7795: continue;
1.238 brouard 7796: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7797: strcpy(gplotlabel,"(");
1.238 brouard 7798: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7799: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7800: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7801: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7802: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7803: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7804: vlv= nbcode[Tvaraff[k]][lv];
7805: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7806: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7807: }
7808: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7809: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7810: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7811: }
1.264 brouard 7812: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7813: fprintf(ficgp,"\n#\n");
7814: if(invalidvarcomb[k1]){
7815: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7816: continue;
1.223 brouard 7817: }
1.238 brouard 7818:
1.241 brouard 7819: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7820: 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 7821: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7822: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7823: k=3;
7824: for (i=1; i<= nlstate ; i ++){
7825: if(i==1){
7826: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7827: }else{
7828: fprintf(ficgp,", '' ");
7829: }
7830: l=(nlstate+ndeath)*(i-1)+1;
7831: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7832: for (j=2; j<= nlstate+ndeath ; j ++)
7833: fprintf(ficgp,"+$%d",k+l+j-1);
7834: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7835: } /* nlstate */
1.264 brouard 7836: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7837: } /* end cpt state*/
7838: } /* end nres */
7839: } /* end covariate k1 */
7840:
1.220 brouard 7841: /* 5eme */
1.201 brouard 7842: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7843: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7844: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7845: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7846: continue;
1.238 brouard 7847: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7848: strcpy(gplotlabel,"(");
1.238 brouard 7849: 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);
7850: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7851: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7852: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7853: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7854: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7855: vlv= nbcode[Tvaraff[k]][lv];
7856: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7857: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7858: }
7859: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7860: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7861: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7862: }
1.264 brouard 7863: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7864: fprintf(ficgp,"\n#\n");
7865: if(invalidvarcomb[k1]){
7866: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7867: continue;
7868: }
1.227 brouard 7869:
1.241 brouard 7870: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7871: 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 7872: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7873: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7874: k=3;
7875: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7876: if(j==1)
7877: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7878: else
7879: fprintf(ficgp,", '' ");
7880: l=(nlstate+ndeath)*(cpt-1) +j;
7881: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7882: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7883: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7884: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7885: } /* nlstate */
7886: fprintf(ficgp,", '' ");
7887: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7888: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7889: l=(nlstate+ndeath)*(cpt-1) +j;
7890: if(j < nlstate)
7891: fprintf(ficgp,"$%d +",k+l);
7892: else
7893: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7894: }
1.264 brouard 7895: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7896: } /* end cpt state*/
7897: } /* end covariate */
7898: } /* end nres */
1.227 brouard 7899:
1.220 brouard 7900: /* 6eme */
1.202 brouard 7901: /* CV preval stable (period) for each covariate */
1.237 brouard 7902: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7903: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7904: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7905: continue;
1.255 brouard 7906: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7907: strcpy(gplotlabel,"(");
1.288 brouard 7908: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7909: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7910: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7911: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7912: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7913: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7914: vlv= nbcode[Tvaraff[k]][lv];
7915: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7916: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7917: }
1.237 brouard 7918: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7919: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7920: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7921: }
1.264 brouard 7922: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7923: fprintf(ficgp,"\n#\n");
1.223 brouard 7924: if(invalidvarcomb[k1]){
1.227 brouard 7925: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7926: continue;
1.223 brouard 7927: }
1.227 brouard 7928:
1.241 brouard 7929: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7930: 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 7931: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7932: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7933: k=3; /* Offset */
1.255 brouard 7934: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7935: if(i==1)
7936: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7937: else
7938: fprintf(ficgp,", '' ");
1.255 brouard 7939: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7940: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7941: for (j=2; j<= nlstate ; j ++)
7942: fprintf(ficgp,"+$%d",k+l+j-1);
7943: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7944: } /* nlstate */
1.264 brouard 7945: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7946: } /* end cpt state*/
7947: } /* end covariate */
1.227 brouard 7948:
7949:
1.220 brouard 7950: /* 7eme */
1.296 brouard 7951: if(prevbcast == 1){
1.288 brouard 7952: /* CV backward prevalence for each covariate */
1.237 brouard 7953: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7954: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7955: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7956: continue;
1.268 brouard 7957: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7958: strcpy(gplotlabel,"(");
1.288 brouard 7959: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7960: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7961: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7962: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7963: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7964: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7965: vlv= nbcode[Tvaraff[k]][lv];
7966: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7967: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7968: }
1.237 brouard 7969: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7970: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7971: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7972: }
1.264 brouard 7973: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7974: fprintf(ficgp,"\n#\n");
7975: if(invalidvarcomb[k1]){
7976: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7977: continue;
7978: }
7979:
1.241 brouard 7980: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7981: 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 7982: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7983: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7984: k=3; /* Offset */
1.268 brouard 7985: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7986: if(i==1)
7987: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7988: else
7989: fprintf(ficgp,", '' ");
7990: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7991: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7992: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7993: /* 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 7994: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7995: /* for (j=2; j<= nlstate ; j ++) */
7996: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7997: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7998: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7999: } /* nlstate */
1.264 brouard 8000: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8001: } /* end cpt state*/
8002: } /* end covariate */
1.296 brouard 8003: } /* End if prevbcast */
1.218 brouard 8004:
1.223 brouard 8005: /* 8eme */
1.218 brouard 8006: if(prevfcast==1){
1.288 brouard 8007: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8008:
1.237 brouard 8009: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8010: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8011: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8012: continue;
1.211 brouard 8013: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8014: strcpy(gplotlabel,"(");
1.288 brouard 8015: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8016: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8017: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8018: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8019: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8020: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8021: vlv= nbcode[Tvaraff[k]][lv];
8022: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8023: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8024: }
1.237 brouard 8025: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8026: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8027: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8028: }
1.264 brouard 8029: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8030: fprintf(ficgp,"\n#\n");
8031: if(invalidvarcomb[k1]){
8032: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8033: continue;
8034: }
8035:
8036: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8037: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8038: 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 8039: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8040: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8041:
8042: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8043: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8044: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8045: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8046: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8047: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8048: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8049: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8050: if(i==istart){
1.227 brouard 8051: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8052: }else{
8053: fprintf(ficgp,",\\\n '' ");
8054: }
8055: if(cptcoveff ==0){ /* No covariate */
8056: ioffset=2; /* Age is in 2 */
8057: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8058: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8059: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8060: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8061: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8062: if(i==nlstate+1){
1.270 brouard 8063: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8064: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8065: fprintf(ficgp,",\\\n '' ");
8066: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8067: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8068: offyear, \
1.268 brouard 8069: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8070: }else
1.227 brouard 8071: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8072: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8073: }else{ /* more than 2 covariates */
1.270 brouard 8074: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8075: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8076: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8077: iyearc=ioffset-1;
8078: iagec=ioffset;
1.227 brouard 8079: fprintf(ficgp," u %d:(",ioffset);
8080: kl=0;
8081: strcpy(gplotcondition,"(");
8082: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
8083: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
8084: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8085: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8086: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8087: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
8088: kl++;
8089: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8090: kl++;
8091: if(k <cptcoveff && cptcoveff>1)
8092: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8093: }
8094: strcpy(gplotcondition+strlen(gplotcondition),")");
8095: /* 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 *\/ */
8096: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8097: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8098: /* '' 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*/
8099: if(i==nlstate+1){
1.270 brouard 8100: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8101: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8102: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8103: fprintf(ficgp," u %d:(",iagec);
8104: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8105: iyearc, iagec, offyear, \
8106: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8107: /* '' 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 8108: }else{
8109: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8110: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8111: }
8112: } /* end if covariate */
8113: } /* nlstate */
1.264 brouard 8114: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8115: } /* end cpt state*/
8116: } /* end covariate */
8117: } /* End if prevfcast */
1.227 brouard 8118:
1.296 brouard 8119: if(prevbcast==1){
1.268 brouard 8120: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8121:
8122: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8123: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8124: if(m != 1 && TKresult[nres]!= k1)
8125: continue;
8126: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8127: strcpy(gplotlabel,"(");
8128: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
8129: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8130: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8131: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8132: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8133: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8134: vlv= nbcode[Tvaraff[k]][lv];
8135: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8136: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8137: }
8138: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8139: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8140: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8141: }
8142: strcpy(gplotlabel+strlen(gplotlabel),")");
8143: fprintf(ficgp,"\n#\n");
8144: if(invalidvarcomb[k1]){
8145: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8146: continue;
8147: }
8148:
8149: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8150: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8151: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8152: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8153: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8154:
8155: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8156: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8157: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8158: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8159: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8160: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8161: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8162: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8163: if(i==istart){
8164: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8165: }else{
8166: fprintf(ficgp,",\\\n '' ");
8167: }
8168: if(cptcoveff ==0){ /* No covariate */
8169: ioffset=2; /* Age is in 2 */
8170: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8171: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8172: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8173: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8174: fprintf(ficgp," u %d:(", ioffset);
8175: if(i==nlstate+1){
1.270 brouard 8176: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8177: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8178: fprintf(ficgp,",\\\n '' ");
8179: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8180: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8181: offbyear, \
8182: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8183: }else
8184: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8185: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8186: }else{ /* more than 2 covariates */
1.270 brouard 8187: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8188: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8189: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8190: iyearc=ioffset-1;
8191: iagec=ioffset;
1.268 brouard 8192: fprintf(ficgp," u %d:(",ioffset);
8193: kl=0;
8194: strcpy(gplotcondition,"(");
8195: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
8196: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
8197: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8198: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8199: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8200: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
8201: kl++;
8202: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8203: kl++;
8204: if(k <cptcoveff && cptcoveff>1)
8205: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8206: }
8207: strcpy(gplotcondition+strlen(gplotcondition),")");
8208: /* 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 *\/ */
8209: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8210: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8211: /* '' 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*/
8212: if(i==nlstate+1){
1.270 brouard 8213: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
8214: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 8215: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8216: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 8217: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 8218: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
8219: iyearc,iagec,offbyear, \
8220: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 8221: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
8222: }else{
8223: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
8224: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
8225: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
8226: }
8227: } /* end if covariate */
8228: } /* nlstate */
8229: fprintf(ficgp,"\nset out; unset label;\n");
8230: } /* end cpt state*/
8231: } /* end covariate */
1.296 brouard 8232: } /* End if prevbcast */
1.268 brouard 8233:
1.227 brouard 8234:
1.238 brouard 8235: /* 9eme writing MLE parameters */
8236: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 8237: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 8238: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 8239: for(k=1; k <=(nlstate+ndeath); k++){
8240: if (k != i) {
1.227 brouard 8241: fprintf(ficgp,"# current state %d\n",k);
8242: for(j=1; j <=ncovmodel; j++){
8243: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
8244: jk++;
8245: }
8246: fprintf(ficgp,"\n");
1.126 brouard 8247: }
8248: }
1.223 brouard 8249: }
1.187 brouard 8250: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8251:
1.145 brouard 8252: /*goto avoid;*/
1.238 brouard 8253: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8254: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8255: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8256: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8257: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8258: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8259: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8260: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8261: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8262: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8263: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8264: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8265: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8266: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8267: fprintf(ficgp,"#\n");
1.223 brouard 8268: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8269: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8270: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8271: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8272: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
8273: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 8274: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 8275: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8276: continue;
1.264 brouard 8277: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
8278: strcpy(gplotlabel,"(");
1.276 brouard 8279: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 8280: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8281: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8282: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8283: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8284: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8285: vlv= nbcode[Tvaraff[k]][lv];
8286: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8287: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8288: }
1.237 brouard 8289: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8290: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8291: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8292: }
1.264 brouard 8293: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8294: fprintf(ficgp,"\n#\n");
1.264 brouard 8295: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8296: fprintf(ficgp,"\nset key outside ");
8297: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8298: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8299: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8300: if (ng==1){
8301: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8302: fprintf(ficgp,"\nunset log y");
8303: }else if (ng==2){
8304: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8305: fprintf(ficgp,"\nset log y");
8306: }else if (ng==3){
8307: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8308: fprintf(ficgp,"\nset log y");
8309: }else
8310: fprintf(ficgp,"\nunset title ");
8311: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8312: i=1;
8313: for(k2=1; k2<=nlstate; k2++) {
8314: k3=i;
8315: for(k=1; k<=(nlstate+ndeath); k++) {
8316: if (k != k2){
8317: switch( ng) {
8318: case 1:
8319: if(nagesqr==0)
8320: fprintf(ficgp," p%d+p%d*x",i,i+1);
8321: else /* nagesqr =1 */
8322: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8323: break;
8324: case 2: /* ng=2 */
8325: if(nagesqr==0)
8326: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8327: else /* nagesqr =1 */
8328: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8329: break;
8330: case 3:
8331: if(nagesqr==0)
8332: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8333: else /* nagesqr =1 */
8334: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8335: break;
8336: }
8337: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8338: ijp=1; /* product no age */
8339: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8340: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8341: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8342: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8343: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
8344: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8345: if(DummyV[j]==0){
8346: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8347: }else{ /* quantitative */
8348: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8349: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8350: }
8351: ij++;
1.237 brouard 8352: }
1.268 brouard 8353: }
8354: }else if(cptcovprod >0){
8355: if(j==Tprod[ijp]) { /* */
8356: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8357: if(ijp <=cptcovprod) { /* Product */
8358: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8359: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8360: /* 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)]); */
8361: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8362: }else{ /* Vn is dummy and Vm is quanti */
8363: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8364: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8365: }
8366: }else{ /* Vn*Vm Vn is quanti */
8367: if(DummyV[Tvard[ijp][2]]==0){
8368: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8369: }else{ /* Both quanti */
8370: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8371: }
1.237 brouard 8372: }
1.268 brouard 8373: ijp++;
1.237 brouard 8374: }
1.268 brouard 8375: } /* end Tprod */
1.237 brouard 8376: } else{ /* simple covariate */
1.264 brouard 8377: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8378: if(Dummy[j]==0){
8379: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8380: }else{ /* quantitative */
8381: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8382: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8383: }
1.237 brouard 8384: } /* end simple */
8385: } /* end j */
1.223 brouard 8386: }else{
8387: i=i-ncovmodel;
8388: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8389: fprintf(ficgp," (1.");
8390: }
1.227 brouard 8391:
1.223 brouard 8392: if(ng != 1){
8393: fprintf(ficgp,")/(1");
1.227 brouard 8394:
1.264 brouard 8395: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8396: if(nagesqr==0)
1.264 brouard 8397: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8398: else /* nagesqr =1 */
1.264 brouard 8399: 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 8400:
1.223 brouard 8401: ij=1;
8402: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8403: if(cptcovage >0){
8404: if((j-2)==Tage[ij]) { /* Bug valgrind */
8405: if(ij <=cptcovage) { /* Bug valgrind */
8406: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8407: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8408: ij++;
8409: }
8410: }
8411: }else
8412: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/* Valgrind bug nbcode */
1.223 brouard 8413: }
8414: fprintf(ficgp,")");
8415: }
8416: fprintf(ficgp,")");
8417: if(ng ==2)
1.276 brouard 8418: 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 8419: else /* ng= 3 */
1.276 brouard 8420: 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);
1.223 brouard 8421: }else{ /* end ng <> 1 */
8422: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8423: 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 8424: }
8425: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8426: fprintf(ficgp,",");
8427: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8428: fprintf(ficgp,",");
8429: i=i+ncovmodel;
8430: } /* end k */
8431: } /* end k2 */
1.276 brouard 8432: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8433: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8434: } /* end k1 */
1.223 brouard 8435: } /* end ng */
8436: /* avoid: */
8437: fflush(ficgp);
1.126 brouard 8438: } /* end gnuplot */
8439:
8440:
8441: /*************** Moving average **************/
1.219 brouard 8442: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8443: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8444:
1.222 brouard 8445: int i, cpt, cptcod;
8446: int modcovmax =1;
8447: int mobilavrange, mob;
8448: int iage=0;
1.288 brouard 8449: int firstA1=0, firstA2=0;
1.222 brouard 8450:
1.266 brouard 8451: double sum=0., sumr=0.;
1.222 brouard 8452: double age;
1.266 brouard 8453: double *sumnewp, *sumnewm, *sumnewmr;
8454: double *agemingood, *agemaxgood;
8455: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8456:
8457:
1.278 brouard 8458: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8459: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8460:
8461: sumnewp = vector(1,ncovcombmax);
8462: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8463: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8464: agemingood = vector(1,ncovcombmax);
1.266 brouard 8465: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8466: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8467: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8468:
8469: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8470: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8471: sumnewp[cptcod]=0.;
1.266 brouard 8472: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8473: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8474: }
8475: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8476:
1.266 brouard 8477: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8478: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8479: else mobilavrange=mobilav;
8480: for (age=bage; age<=fage; age++)
8481: for (i=1; i<=nlstate;i++)
8482: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8483: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8484: /* We keep the original values on the extreme ages bage, fage and for
8485: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8486: we use a 5 terms etc. until the borders are no more concerned.
8487: */
8488: for (mob=3;mob <=mobilavrange;mob=mob+2){
8489: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8490: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8491: sumnewm[cptcod]=0.;
8492: for (i=1; i<=nlstate;i++){
1.222 brouard 8493: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8494: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8495: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8496: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8497: }
8498: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8499: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8500: } /* end i */
8501: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8502: } /* end cptcod */
1.222 brouard 8503: }/* end age */
8504: }/* end mob */
1.266 brouard 8505: }else{
8506: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8507: return -1;
1.266 brouard 8508: }
8509:
8510: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8511: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8512: if(invalidvarcomb[cptcod]){
8513: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8514: continue;
8515: }
1.219 brouard 8516:
1.266 brouard 8517: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8518: sumnewm[cptcod]=0.;
8519: sumnewmr[cptcod]=0.;
8520: for (i=1; i<=nlstate;i++){
8521: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8522: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8523: }
8524: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8525: agemingoodr[cptcod]=age;
8526: }
8527: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8528: agemingood[cptcod]=age;
8529: }
8530: } /* age */
8531: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8532: sumnewm[cptcod]=0.;
1.266 brouard 8533: sumnewmr[cptcod]=0.;
1.222 brouard 8534: for (i=1; i<=nlstate;i++){
8535: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8536: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8537: }
8538: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8539: agemaxgoodr[cptcod]=age;
1.222 brouard 8540: }
8541: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8542: agemaxgood[cptcod]=age;
8543: }
8544: } /* age */
8545: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8546: /* but they will change */
1.288 brouard 8547: firstA1=0;firstA2=0;
1.266 brouard 8548: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8549: sumnewm[cptcod]=0.;
8550: sumnewmr[cptcod]=0.;
8551: for (i=1; i<=nlstate;i++){
8552: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8553: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8554: }
8555: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8556: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8557: agemaxgoodr[cptcod]=age; /* age min */
8558: for (i=1; i<=nlstate;i++)
8559: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8560: }else{ /* bad we change the value with the values of good ages */
8561: for (i=1; i<=nlstate;i++){
8562: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8563: } /* i */
8564: } /* end bad */
8565: }else{
8566: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8567: agemaxgood[cptcod]=age;
8568: }else{ /* bad we change the value with the values of good ages */
8569: for (i=1; i<=nlstate;i++){
8570: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8571: } /* i */
8572: } /* end bad */
8573: }/* end else */
8574: sum=0.;sumr=0.;
8575: for (i=1; i<=nlstate;i++){
8576: sum+=mobaverage[(int)age][i][cptcod];
8577: sumr+=probs[(int)age][i][cptcod];
8578: }
8579: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8580: if(!firstA1){
8581: firstA1=1;
8582: 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);
8583: }
8584: 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 8585: } /* end bad */
8586: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8587: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8588: if(!firstA2){
8589: firstA2=1;
8590: 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);
8591: }
8592: 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 8593: } /* end bad */
8594: }/* age */
1.266 brouard 8595:
8596: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8597: sumnewm[cptcod]=0.;
1.266 brouard 8598: sumnewmr[cptcod]=0.;
1.222 brouard 8599: for (i=1; i<=nlstate;i++){
8600: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8601: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8602: }
8603: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8604: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8605: agemingoodr[cptcod]=age;
8606: for (i=1; i<=nlstate;i++)
8607: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8608: }else{ /* bad we change the value with the values of good ages */
8609: for (i=1; i<=nlstate;i++){
8610: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8611: } /* i */
8612: } /* end bad */
8613: }else{
8614: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8615: agemingood[cptcod]=age;
8616: }else{ /* bad */
8617: for (i=1; i<=nlstate;i++){
8618: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8619: } /* i */
8620: } /* end bad */
8621: }/* end else */
8622: sum=0.;sumr=0.;
8623: for (i=1; i<=nlstate;i++){
8624: sum+=mobaverage[(int)age][i][cptcod];
8625: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8626: }
1.266 brouard 8627: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8628: 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 8629: } /* end bad */
8630: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8631: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8632: 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 8633: } /* end bad */
8634: }/* age */
1.266 brouard 8635:
1.222 brouard 8636:
8637: for (age=bage; age<=fage; age++){
1.235 brouard 8638: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8639: sumnewp[cptcod]=0.;
8640: sumnewm[cptcod]=0.;
8641: for (i=1; i<=nlstate;i++){
8642: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8643: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8644: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8645: }
8646: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8647: }
8648: /* printf("\n"); */
8649: /* } */
1.266 brouard 8650:
1.222 brouard 8651: /* brutal averaging */
1.266 brouard 8652: /* for (i=1; i<=nlstate;i++){ */
8653: /* for (age=1; age<=bage; age++){ */
8654: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8655: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8656: /* } */
8657: /* for (age=fage; age<=AGESUP; age++){ */
8658: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8659: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8660: /* } */
8661: /* } /\* end i status *\/ */
8662: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8663: /* for (age=1; age<=AGESUP; age++){ */
8664: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8665: /* mobaverage[(int)age][i][cptcod]=0.; */
8666: /* } */
8667: /* } */
1.222 brouard 8668: }/* end cptcod */
1.266 brouard 8669: free_vector(agemaxgoodr,1, ncovcombmax);
8670: free_vector(agemaxgood,1, ncovcombmax);
8671: free_vector(agemingood,1, ncovcombmax);
8672: free_vector(agemingoodr,1, ncovcombmax);
8673: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8674: free_vector(sumnewm,1, ncovcombmax);
8675: free_vector(sumnewp,1, ncovcombmax);
8676: return 0;
8677: }/* End movingaverage */
1.218 brouard 8678:
1.126 brouard 8679:
1.296 brouard 8680:
1.126 brouard 8681: /************** Forecasting ******************/
1.296 brouard 8682: /* 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)*/
8683: 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){
8684: /* dateintemean, mean date of interviews
8685: dateprojd, year, month, day of starting projection
8686: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8687: agemin, agemax range of age
8688: dateprev1 dateprev2 range of dates during which prevalence is computed
8689: */
1.296 brouard 8690: /* double anprojd, mprojd, jprojd; */
8691: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8692: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8693: double agec; /* generic age */
1.296 brouard 8694: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8695: double *popeffectif,*popcount;
8696: double ***p3mat;
1.218 brouard 8697: /* double ***mobaverage; */
1.126 brouard 8698: char fileresf[FILENAMELENGTH];
8699:
8700: agelim=AGESUP;
1.211 brouard 8701: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8702: in each health status at the date of interview (if between dateprev1 and dateprev2).
8703: We still use firstpass and lastpass as another selection.
8704: */
1.214 brouard 8705: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8706: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8707:
1.201 brouard 8708: strcpy(fileresf,"F_");
8709: strcat(fileresf,fileresu);
1.126 brouard 8710: if((ficresf=fopen(fileresf,"w"))==NULL) {
8711: printf("Problem with forecast resultfile: %s\n", fileresf);
8712: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8713: }
1.235 brouard 8714: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8715: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8716:
1.225 brouard 8717: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8718:
8719:
8720: stepsize=(int) (stepm+YEARM-1)/YEARM;
8721: if (stepm<=12) stepsize=1;
8722: if(estepm < stepm){
8723: printf ("Problem %d lower than %d\n",estepm, stepm);
8724: }
1.270 brouard 8725: else{
8726: hstepm=estepm;
8727: }
8728: if(estepm > stepm){ /* Yes every two year */
8729: stepsize=2;
8730: }
1.296 brouard 8731: hstepm=hstepm/stepm;
1.126 brouard 8732:
1.296 brouard 8733:
8734: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8735: /* fractional in yp1 *\/ */
8736: /* aintmean=yp; */
8737: /* yp2=modf((yp1*12),&yp); */
8738: /* mintmean=yp; */
8739: /* yp1=modf((yp2*30.5),&yp); */
8740: /* jintmean=yp; */
8741: /* if(jintmean==0) jintmean=1; */
8742: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8743:
1.296 brouard 8744:
8745: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8746: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8747: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8748: i1=pow(2,cptcoveff);
1.126 brouard 8749: if (cptcovn < 1){i1=1;}
8750:
1.296 brouard 8751: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8752:
8753: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8754:
1.126 brouard 8755: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8756: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8757: for(k=1; k<=i1;k++){
1.253 brouard 8758: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8759: continue;
1.227 brouard 8760: if(invalidvarcomb[k]){
8761: printf("\nCombination (%d) projection ignored because no cases \n",k);
8762: continue;
8763: }
8764: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8765: for(j=1;j<=cptcoveff;j++) {
8766: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8767: }
1.235 brouard 8768: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8769: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8770: }
1.227 brouard 8771: fprintf(ficresf," yearproj age");
8772: for(j=1; j<=nlstate+ndeath;j++){
8773: for(i=1; i<=nlstate;i++)
8774: fprintf(ficresf," p%d%d",i,j);
8775: fprintf(ficresf," wp.%d",j);
8776: }
1.296 brouard 8777: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8778: fprintf(ficresf,"\n");
1.296 brouard 8779: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8780: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8781: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8782: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8783: nhstepm = nhstepm/hstepm;
8784: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8785: oldm=oldms;savm=savms;
1.268 brouard 8786: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8787: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8788: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8789: for (h=0; h<=nhstepm; h++){
8790: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8791: break;
8792: }
8793: }
8794: fprintf(ficresf,"\n");
8795: for(j=1;j<=cptcoveff;j++)
8796: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8797: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8798:
8799: for(j=1; j<=nlstate+ndeath;j++) {
8800: ppij=0.;
8801: for(i=1; i<=nlstate;i++) {
1.278 brouard 8802: if (mobilav>=1)
8803: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8804: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8805: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8806: }
1.268 brouard 8807: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8808: } /* end i */
8809: fprintf(ficresf," %.3f", ppij);
8810: }/* end j */
1.227 brouard 8811: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8812: } /* end agec */
1.266 brouard 8813: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8814: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8815: } /* end yearp */
8816: } /* end k */
1.219 brouard 8817:
1.126 brouard 8818: fclose(ficresf);
1.215 brouard 8819: printf("End of Computing forecasting \n");
8820: fprintf(ficlog,"End of Computing forecasting\n");
8821:
1.126 brouard 8822: }
8823:
1.269 brouard 8824: /************** Back Forecasting ******************/
1.296 brouard 8825: /* 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){ */
8826: 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){
8827: /* back1, year, month, day of starting backprojection
1.267 brouard 8828: agemin, agemax range of age
8829: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8830: anback2 year of end of backprojection (same day and month as back1).
8831: prevacurrent and prev are prevalences.
1.267 brouard 8832: */
8833: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8834: double agec; /* generic age */
1.302 brouard 8835: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 8836: double *popeffectif,*popcount;
8837: double ***p3mat;
8838: /* double ***mobaverage; */
8839: char fileresfb[FILENAMELENGTH];
8840:
1.268 brouard 8841: agelim=AGEINF;
1.267 brouard 8842: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8843: in each health status at the date of interview (if between dateprev1 and dateprev2).
8844: We still use firstpass and lastpass as another selection.
8845: */
8846: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8847: /* firstpass, lastpass, stepm, weightopt, model); */
8848:
8849: /*Do we need to compute prevalence again?*/
8850:
8851: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8852:
8853: strcpy(fileresfb,"FB_");
8854: strcat(fileresfb,fileresu);
8855: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8856: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8857: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8858: }
8859: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8860: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8861:
8862: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8863:
8864:
8865: stepsize=(int) (stepm+YEARM-1)/YEARM;
8866: if (stepm<=12) stepsize=1;
8867: if(estepm < stepm){
8868: printf ("Problem %d lower than %d\n",estepm, stepm);
8869: }
1.270 brouard 8870: else{
8871: hstepm=estepm;
8872: }
8873: if(estepm >= stepm){ /* Yes every two year */
8874: stepsize=2;
8875: }
1.267 brouard 8876:
8877: hstepm=hstepm/stepm;
1.296 brouard 8878: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8879: /* fractional in yp1 *\/ */
8880: /* aintmean=yp; */
8881: /* yp2=modf((yp1*12),&yp); */
8882: /* mintmean=yp; */
8883: /* yp1=modf((yp2*30.5),&yp); */
8884: /* jintmean=yp; */
8885: /* if(jintmean==0) jintmean=1; */
8886: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8887:
8888: i1=pow(2,cptcoveff);
8889: if (cptcovn < 1){i1=1;}
8890:
1.296 brouard 8891: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8892: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8893:
8894: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8895:
8896: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8897: for(k=1; k<=i1;k++){
8898: if(i1 != 1 && TKresult[nres]!= k)
8899: continue;
8900: if(invalidvarcomb[k]){
8901: printf("\nCombination (%d) projection ignored because no cases \n",k);
8902: continue;
8903: }
1.268 brouard 8904: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8905: for(j=1;j<=cptcoveff;j++) {
8906: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8907: }
8908: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8909: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8910: }
8911: fprintf(ficresfb," yearbproj age");
8912: for(j=1; j<=nlstate+ndeath;j++){
8913: for(i=1; i<=nlstate;i++)
1.268 brouard 8914: fprintf(ficresfb," b%d%d",i,j);
8915: fprintf(ficresfb," b.%d",j);
1.267 brouard 8916: }
1.296 brouard 8917: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8918: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8919: fprintf(ficresfb,"\n");
1.296 brouard 8920: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8921: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8922: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8923: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8924: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8925: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8926: nhstepm = nhstepm/hstepm;
8927: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8928: oldm=oldms;savm=savms;
1.268 brouard 8929: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8930: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8931: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8932: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8933: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8934: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8935: for (h=0; h<=nhstepm; h++){
1.268 brouard 8936: if (h*hstepm/YEARM*stepm ==-yearp) {
8937: break;
8938: }
8939: }
8940: fprintf(ficresfb,"\n");
8941: for(j=1;j<=cptcoveff;j++)
8942: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8943: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 8944: for(i=1; i<=nlstate+ndeath;i++) {
8945: ppij=0.;ppi=0.;
8946: for(j=1; j<=nlstate;j++) {
8947: /* if (mobilav==1) */
1.269 brouard 8948: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8949: ppi=ppi+prevacurrent[(int)agec][j][k];
8950: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8951: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8952: /* else { */
8953: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8954: /* } */
1.268 brouard 8955: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8956: } /* end j */
8957: if(ppi <0.99){
8958: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8959: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8960: }
8961: fprintf(ficresfb," %.3f", ppij);
8962: }/* end j */
1.267 brouard 8963: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8964: } /* end agec */
8965: } /* end yearp */
8966: } /* end k */
1.217 brouard 8967:
1.267 brouard 8968: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8969:
1.267 brouard 8970: fclose(ficresfb);
8971: printf("End of Computing Back forecasting \n");
8972: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8973:
1.267 brouard 8974: }
1.217 brouard 8975:
1.269 brouard 8976: /* Variance of prevalence limit: varprlim */
8977: 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 8978: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 8979:
8980: char fileresvpl[FILENAMELENGTH];
8981: FILE *ficresvpl;
8982: double **oldm, **savm;
8983: double **varpl; /* Variances of prevalence limits by age */
8984: int i1, k, nres, j ;
8985:
8986: strcpy(fileresvpl,"VPL_");
8987: strcat(fileresvpl,fileresu);
8988: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 8989: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 8990: exit(0);
8991: }
1.288 brouard 8992: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8993: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 8994:
8995: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8996: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8997:
8998: i1=pow(2,cptcoveff);
8999: if (cptcovn < 1){i1=1;}
9000:
9001: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9002: for(k=1; k<=i1;k++){
9003: if(i1 != 1 && TKresult[nres]!= k)
9004: continue;
9005: fprintf(ficresvpl,"\n#****** ");
9006: printf("\n#****** ");
9007: fprintf(ficlog,"\n#****** ");
9008: for(j=1;j<=cptcoveff;j++) {
9009: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9010: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9011: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9012: }
9013: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9014: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9015: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9016: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9017: }
9018: fprintf(ficresvpl,"******\n");
9019: printf("******\n");
9020: fprintf(ficlog,"******\n");
9021:
9022: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9023: oldm=oldms;savm=savms;
9024: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9025: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9026: /*}*/
9027: }
9028:
9029: fclose(ficresvpl);
1.288 brouard 9030: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9031: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9032:
9033: }
9034: /* Variance of back prevalence: varbprlim */
9035: 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){
9036: /*------- Variance of back (stable) prevalence------*/
9037:
9038: char fileresvbl[FILENAMELENGTH];
9039: FILE *ficresvbl;
9040:
9041: double **oldm, **savm;
9042: double **varbpl; /* Variances of back prevalence limits by age */
9043: int i1, k, nres, j ;
9044:
9045: strcpy(fileresvbl,"VBL_");
9046: strcat(fileresvbl,fileresu);
9047: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9048: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9049: exit(0);
9050: }
9051: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9052: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9053:
9054:
9055: i1=pow(2,cptcoveff);
9056: if (cptcovn < 1){i1=1;}
9057:
9058: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9059: for(k=1; k<=i1;k++){
9060: if(i1 != 1 && TKresult[nres]!= k)
9061: continue;
9062: fprintf(ficresvbl,"\n#****** ");
9063: printf("\n#****** ");
9064: fprintf(ficlog,"\n#****** ");
9065: for(j=1;j<=cptcoveff;j++) {
9066: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9067: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9068: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9069: }
9070: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9071: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9072: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9073: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9074: }
9075: fprintf(ficresvbl,"******\n");
9076: printf("******\n");
9077: fprintf(ficlog,"******\n");
9078:
9079: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9080: oldm=oldms;savm=savms;
9081:
9082: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9083: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9084: /*}*/
9085: }
9086:
9087: fclose(ficresvbl);
9088: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9089: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9090:
9091: } /* End of varbprlim */
9092:
1.126 brouard 9093: /************** Forecasting *****not tested NB*************/
1.227 brouard 9094: /* 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 9095:
1.227 brouard 9096: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
9097: /* int *popage; */
9098: /* double calagedatem, agelim, kk1, kk2; */
9099: /* double *popeffectif,*popcount; */
9100: /* double ***p3mat,***tabpop,***tabpopprev; */
9101: /* /\* double ***mobaverage; *\/ */
9102: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 9103:
1.227 brouard 9104: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9105: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9106: /* agelim=AGESUP; */
9107: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 9108:
1.227 brouard 9109: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 9110:
9111:
1.227 brouard 9112: /* strcpy(filerespop,"POP_"); */
9113: /* strcat(filerespop,fileresu); */
9114: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
9115: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
9116: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
9117: /* } */
9118: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
9119: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 9120:
1.227 brouard 9121: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 9122:
1.227 brouard 9123: /* /\* if (mobilav!=0) { *\/ */
9124: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
9125: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
9126: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9127: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9128: /* /\* } *\/ */
9129: /* /\* } *\/ */
1.126 brouard 9130:
1.227 brouard 9131: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
9132: /* if (stepm<=12) stepsize=1; */
1.126 brouard 9133:
1.227 brouard 9134: /* agelim=AGESUP; */
1.126 brouard 9135:
1.227 brouard 9136: /* hstepm=1; */
9137: /* hstepm=hstepm/stepm; */
1.218 brouard 9138:
1.227 brouard 9139: /* if (popforecast==1) { */
9140: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
9141: /* printf("Problem with population file : %s\n",popfile);exit(0); */
9142: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
9143: /* } */
9144: /* popage=ivector(0,AGESUP); */
9145: /* popeffectif=vector(0,AGESUP); */
9146: /* popcount=vector(0,AGESUP); */
1.126 brouard 9147:
1.227 brouard 9148: /* i=1; */
9149: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 9150:
1.227 brouard 9151: /* imx=i; */
9152: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
9153: /* } */
1.218 brouard 9154:
1.227 brouard 9155: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
9156: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
9157: /* k=k+1; */
9158: /* fprintf(ficrespop,"\n#******"); */
9159: /* for(j=1;j<=cptcoveff;j++) { */
9160: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
9161: /* } */
9162: /* fprintf(ficrespop,"******\n"); */
9163: /* fprintf(ficrespop,"# Age"); */
9164: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
9165: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 9166:
1.227 brouard 9167: /* for (cpt=0; cpt<=0;cpt++) { */
9168: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 9169:
1.227 brouard 9170: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9171: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9172: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9173:
1.227 brouard 9174: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9175: /* oldm=oldms;savm=savms; */
9176: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 9177:
1.227 brouard 9178: /* for (h=0; h<=nhstepm; h++){ */
9179: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9180: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9181: /* } */
9182: /* for(j=1; j<=nlstate+ndeath;j++) { */
9183: /* kk1=0.;kk2=0; */
9184: /* for(i=1; i<=nlstate;i++) { */
9185: /* if (mobilav==1) */
9186: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
9187: /* else { */
9188: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
9189: /* } */
9190: /* } */
9191: /* if (h==(int)(calagedatem+12*cpt)){ */
9192: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
9193: /* /\*fprintf(ficrespop," %.3f", kk1); */
9194: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
9195: /* } */
9196: /* } */
9197: /* for(i=1; i<=nlstate;i++){ */
9198: /* kk1=0.; */
9199: /* for(j=1; j<=nlstate;j++){ */
9200: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
9201: /* } */
9202: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
9203: /* } */
1.218 brouard 9204:
1.227 brouard 9205: /* if (h==(int)(calagedatem+12*cpt)) */
9206: /* for(j=1; j<=nlstate;j++) */
9207: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
9208: /* } */
9209: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9210: /* } */
9211: /* } */
1.218 brouard 9212:
1.227 brouard 9213: /* /\******\/ */
1.218 brouard 9214:
1.227 brouard 9215: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
9216: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
9217: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9218: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9219: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9220:
1.227 brouard 9221: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9222: /* oldm=oldms;savm=savms; */
9223: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9224: /* for (h=0; h<=nhstepm; h++){ */
9225: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9226: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9227: /* } */
9228: /* for(j=1; j<=nlstate+ndeath;j++) { */
9229: /* kk1=0.;kk2=0; */
9230: /* for(i=1; i<=nlstate;i++) { */
9231: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
9232: /* } */
9233: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
9234: /* } */
9235: /* } */
9236: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9237: /* } */
9238: /* } */
9239: /* } */
9240: /* } */
1.218 brouard 9241:
1.227 brouard 9242: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 9243:
1.227 brouard 9244: /* if (popforecast==1) { */
9245: /* free_ivector(popage,0,AGESUP); */
9246: /* free_vector(popeffectif,0,AGESUP); */
9247: /* free_vector(popcount,0,AGESUP); */
9248: /* } */
9249: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9250: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9251: /* fclose(ficrespop); */
9252: /* } /\* End of popforecast *\/ */
1.218 brouard 9253:
1.126 brouard 9254: int fileappend(FILE *fichier, char *optionfich)
9255: {
9256: if((fichier=fopen(optionfich,"a"))==NULL) {
9257: printf("Problem with file: %s\n", optionfich);
9258: fprintf(ficlog,"Problem with file: %s\n", optionfich);
9259: return (0);
9260: }
9261: fflush(fichier);
9262: return (1);
9263: }
9264:
9265:
9266: /**************** function prwizard **********************/
9267: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9268: {
9269:
9270: /* Wizard to print covariance matrix template */
9271:
1.164 brouard 9272: char ca[32], cb[32];
9273: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9274: int numlinepar;
9275:
9276: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9277: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9278: for(i=1; i <=nlstate; i++){
9279: jj=0;
9280: for(j=1; j <=nlstate+ndeath; j++){
9281: if(j==i) continue;
9282: jj++;
9283: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9284: printf("%1d%1d",i,j);
9285: fprintf(ficparo,"%1d%1d",i,j);
9286: for(k=1; k<=ncovmodel;k++){
9287: /* printf(" %lf",param[i][j][k]); */
9288: /* fprintf(ficparo," %lf",param[i][j][k]); */
9289: printf(" 0.");
9290: fprintf(ficparo," 0.");
9291: }
9292: printf("\n");
9293: fprintf(ficparo,"\n");
9294: }
9295: }
9296: printf("# Scales (for hessian or gradient estimation)\n");
9297: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9298: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9299: for(i=1; i <=nlstate; i++){
9300: jj=0;
9301: for(j=1; j <=nlstate+ndeath; j++){
9302: if(j==i) continue;
9303: jj++;
9304: fprintf(ficparo,"%1d%1d",i,j);
9305: printf("%1d%1d",i,j);
9306: fflush(stdout);
9307: for(k=1; k<=ncovmodel;k++){
9308: /* printf(" %le",delti3[i][j][k]); */
9309: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9310: printf(" 0.");
9311: fprintf(ficparo," 0.");
9312: }
9313: numlinepar++;
9314: printf("\n");
9315: fprintf(ficparo,"\n");
9316: }
9317: }
9318: printf("# Covariance matrix\n");
9319: /* # 121 Var(a12)\n\ */
9320: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9321: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9322: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9323: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9324: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9325: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9326: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9327: fflush(stdout);
9328: fprintf(ficparo,"# Covariance matrix\n");
9329: /* # 121 Var(a12)\n\ */
9330: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9331: /* # ...\n\ */
9332: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9333:
9334: for(itimes=1;itimes<=2;itimes++){
9335: jj=0;
9336: for(i=1; i <=nlstate; i++){
9337: for(j=1; j <=nlstate+ndeath; j++){
9338: if(j==i) continue;
9339: for(k=1; k<=ncovmodel;k++){
9340: jj++;
9341: ca[0]= k+'a'-1;ca[1]='\0';
9342: if(itimes==1){
9343: printf("#%1d%1d%d",i,j,k);
9344: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9345: }else{
9346: printf("%1d%1d%d",i,j,k);
9347: fprintf(ficparo,"%1d%1d%d",i,j,k);
9348: /* printf(" %.5le",matcov[i][j]); */
9349: }
9350: ll=0;
9351: for(li=1;li <=nlstate; li++){
9352: for(lj=1;lj <=nlstate+ndeath; lj++){
9353: if(lj==li) continue;
9354: for(lk=1;lk<=ncovmodel;lk++){
9355: ll++;
9356: if(ll<=jj){
9357: cb[0]= lk +'a'-1;cb[1]='\0';
9358: if(ll<jj){
9359: if(itimes==1){
9360: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9361: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9362: }else{
9363: printf(" 0.");
9364: fprintf(ficparo," 0.");
9365: }
9366: }else{
9367: if(itimes==1){
9368: printf(" Var(%s%1d%1d)",ca,i,j);
9369: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9370: }else{
9371: printf(" 0.");
9372: fprintf(ficparo," 0.");
9373: }
9374: }
9375: }
9376: } /* end lk */
9377: } /* end lj */
9378: } /* end li */
9379: printf("\n");
9380: fprintf(ficparo,"\n");
9381: numlinepar++;
9382: } /* end k*/
9383: } /*end j */
9384: } /* end i */
9385: } /* end itimes */
9386:
9387: } /* end of prwizard */
9388: /******************* Gompertz Likelihood ******************************/
9389: double gompertz(double x[])
9390: {
1.302 brouard 9391: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9392: int i,n=0; /* n is the size of the sample */
9393:
1.220 brouard 9394: for (i=1;i<=imx ; i++) {
1.126 brouard 9395: sump=sump+weight[i];
9396: /* sump=sump+1;*/
9397: num=num+1;
9398: }
1.302 brouard 9399: L=0.0;
9400: /* agegomp=AGEGOMP; */
1.126 brouard 9401: /* for (i=0; i<=imx; i++)
9402: 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]);*/
9403:
1.302 brouard 9404: for (i=1;i<=imx ; i++) {
9405: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9406: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9407: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9408: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9409: * +
9410: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9411: */
9412: if (wav[i] > 1 || agedc[i] < AGESUP) {
9413: if (cens[i] == 1){
9414: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9415: } else if (cens[i] == 0){
1.126 brouard 9416: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9417: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9418: } else
9419: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9420: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9421: L=L+A*weight[i];
1.126 brouard 9422: /* 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 9423: }
9424: }
1.126 brouard 9425:
1.302 brouard 9426: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9427:
9428: return -2*L*num/sump;
9429: }
9430:
1.136 brouard 9431: #ifdef GSL
9432: /******************* Gompertz_f Likelihood ******************************/
9433: double gompertz_f(const gsl_vector *v, void *params)
9434: {
1.302 brouard 9435: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9436: double *x= (double *) v->data;
9437: int i,n=0; /* n is the size of the sample */
9438:
9439: for (i=0;i<=imx-1 ; i++) {
9440: sump=sump+weight[i];
9441: /* sump=sump+1;*/
9442: num=num+1;
9443: }
9444:
9445:
9446: /* for (i=0; i<=imx; i++)
9447: 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]);*/
9448: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9449: for (i=1;i<=imx ; i++)
9450: {
9451: if (cens[i] == 1 && wav[i]>1)
9452: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9453:
9454: if (cens[i] == 0 && wav[i]>1)
9455: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9456: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9457:
9458: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9459: if (wav[i] > 1 ) { /* ??? */
9460: LL=LL+A*weight[i];
9461: /* 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]);*/
9462: }
9463: }
9464:
9465: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9466: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9467:
9468: return -2*LL*num/sump;
9469: }
9470: #endif
9471:
1.126 brouard 9472: /******************* Printing html file ***********/
1.201 brouard 9473: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9474: int lastpass, int stepm, int weightopt, char model[],\
9475: int imx, double p[],double **matcov,double agemortsup){
9476: int i,k;
9477:
9478: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9479: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9480: for (i=1;i<=2;i++)
9481: 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 9482: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9483: fprintf(fichtm,"</ul>");
9484:
9485: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9486:
9487: 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>");
9488:
9489: for (k=agegomp;k<(agemortsup-2);k++)
9490: 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]);
9491:
9492:
9493: fflush(fichtm);
9494: }
9495:
9496: /******************* Gnuplot file **************/
1.201 brouard 9497: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9498:
9499: char dirfileres[132],optfileres[132];
1.164 brouard 9500:
1.126 brouard 9501: int ng;
9502:
9503:
9504: /*#ifdef windows */
9505: fprintf(ficgp,"cd \"%s\" \n",pathc);
9506: /*#endif */
9507:
9508:
9509: strcpy(dirfileres,optionfilefiname);
9510: strcpy(optfileres,"vpl");
1.199 brouard 9511: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9512: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9513: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9514: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9515: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9516:
9517: }
9518:
1.136 brouard 9519: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9520: {
1.126 brouard 9521:
1.136 brouard 9522: /*-------- data file ----------*/
9523: FILE *fic;
9524: char dummy[]=" ";
1.240 brouard 9525: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9526: int lstra;
1.136 brouard 9527: int linei, month, year,iout;
1.302 brouard 9528: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9529: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9530: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9531: char *stratrunc;
1.223 brouard 9532:
1.240 brouard 9533: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9534: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9535:
1.240 brouard 9536: for(v=1; v <=ncovcol;v++){
9537: DummyV[v]=0;
9538: FixedV[v]=0;
9539: }
9540: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9541: DummyV[v]=1;
9542: FixedV[v]=0;
9543: }
9544: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9545: DummyV[v]=0;
9546: FixedV[v]=1;
9547: }
9548: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9549: DummyV[v]=1;
9550: FixedV[v]=1;
9551: }
9552: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9553: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9554: 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]);
9555: }
1.126 brouard 9556:
1.136 brouard 9557: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9558: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9559: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9560: }
1.126 brouard 9561:
1.302 brouard 9562: /* Is it a BOM UTF-8 Windows file? */
9563: /* First data line */
9564: linei=0;
9565: while(fgets(line, MAXLINE, fic)) {
9566: noffset=0;
9567: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
9568: {
9569: noffset=noffset+3;
9570: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
9571: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
9572: fflush(ficlog); return 1;
9573: }
9574: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
9575: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
9576: {
9577: noffset=noffset+2;
1.304 brouard 9578: 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);
9579: 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 9580: fflush(ficlog); return 1;
9581: }
9582: else if( line[0] == 0 && line[1] == 0)
9583: {
9584: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
9585: noffset=noffset+4;
1.304 brouard 9586: 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);
9587: 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 9588: fflush(ficlog); return 1;
9589: }
9590: } else{
9591: ;/*printf(" Not a BOM file\n");*/
9592: }
9593: /* If line starts with a # it is a comment */
9594: if (line[noffset] == '#') {
9595: linei=linei+1;
9596: break;
9597: }else{
9598: break;
9599: }
9600: }
9601: fclose(fic);
9602: if((fic=fopen(datafile,"r"))==NULL) {
9603: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9604: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
9605: }
9606: /* Not a Bom file */
9607:
1.136 brouard 9608: i=1;
9609: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9610: linei=linei+1;
9611: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9612: if(line[j] == '\t')
9613: line[j] = ' ';
9614: }
9615: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9616: ;
9617: };
9618: line[j+1]=0; /* Trims blanks at end of line */
9619: if(line[0]=='#'){
9620: fprintf(ficlog,"Comment line\n%s\n",line);
9621: printf("Comment line\n%s\n",line);
9622: continue;
9623: }
9624: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9625: strcpy(line, linetmp);
1.223 brouard 9626:
9627: /* Loops on waves */
9628: for (j=maxwav;j>=1;j--){
9629: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9630: cutv(stra, strb, line, ' ');
9631: if(strb[0]=='.') { /* Missing value */
9632: lval=-1;
9633: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9634: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9635: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9636: 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);
9637: 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);
9638: return 1;
9639: }
9640: }else{
9641: errno=0;
9642: /* what_kind_of_number(strb); */
9643: dval=strtod(strb,&endptr);
9644: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9645: /* if(strb != endptr && *endptr == '\0') */
9646: /* dval=dlval; */
9647: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9648: if( strb[0]=='\0' || (*endptr != '\0')){
9649: 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);
9650: 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);
9651: return 1;
9652: }
9653: cotqvar[j][iv][i]=dval;
9654: cotvar[j][ntv+iv][i]=dval;
9655: }
9656: strcpy(line,stra);
1.223 brouard 9657: }/* end loop ntqv */
1.225 brouard 9658:
1.223 brouard 9659: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9660: cutv(stra, strb, line, ' ');
9661: if(strb[0]=='.') { /* Missing value */
9662: lval=-1;
9663: }else{
9664: errno=0;
9665: lval=strtol(strb,&endptr,10);
9666: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9667: if( strb[0]=='\0' || (*endptr != '\0')){
9668: 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);
9669: 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);
9670: return 1;
9671: }
9672: }
9673: if(lval <-1 || lval >1){
9674: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 9675: 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 9676: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9677: For example, for multinomial values like 1, 2 and 3,\n \
9678: build V1=0 V2=0 for the reference value (1),\n \
9679: V1=1 V2=0 for (2) \n \
1.223 brouard 9680: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9681: output of IMaCh is often meaningless.\n \
1.319 brouard 9682: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 9683: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 9684: 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 9685: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9686: For example, for multinomial values like 1, 2 and 3,\n \
9687: build V1=0 V2=0 for the reference value (1),\n \
9688: V1=1 V2=0 for (2) \n \
1.223 brouard 9689: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9690: output of IMaCh is often meaningless.\n \
1.319 brouard 9691: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 9692: return 1;
9693: }
9694: cotvar[j][iv][i]=(double)(lval);
9695: strcpy(line,stra);
1.223 brouard 9696: }/* end loop ntv */
1.225 brouard 9697:
1.223 brouard 9698: /* Statuses at wave */
1.137 brouard 9699: cutv(stra, strb, line, ' ');
1.223 brouard 9700: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9701: lval=-1;
1.136 brouard 9702: }else{
1.238 brouard 9703: errno=0;
9704: lval=strtol(strb,&endptr,10);
9705: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9706: if( strb[0]=='\0' || (*endptr != '\0')){
9707: 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);
9708: 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);
9709: return 1;
9710: }
1.136 brouard 9711: }
1.225 brouard 9712:
1.136 brouard 9713: s[j][i]=lval;
1.225 brouard 9714:
1.223 brouard 9715: /* Date of Interview */
1.136 brouard 9716: strcpy(line,stra);
9717: cutv(stra, strb,line,' ');
1.169 brouard 9718: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9719: }
1.169 brouard 9720: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9721: month=99;
9722: year=9999;
1.136 brouard 9723: }else{
1.225 brouard 9724: 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);
9725: 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);
9726: return 1;
1.136 brouard 9727: }
9728: anint[j][i]= (double) year;
1.302 brouard 9729: mint[j][i]= (double)month;
9730: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
9731: /* 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]); */
9732: /* 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]); */
9733: /* } */
1.136 brouard 9734: strcpy(line,stra);
1.223 brouard 9735: } /* End loop on waves */
1.225 brouard 9736:
1.223 brouard 9737: /* Date of death */
1.136 brouard 9738: cutv(stra, strb,line,' ');
1.169 brouard 9739: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9740: }
1.169 brouard 9741: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9742: month=99;
9743: year=9999;
9744: }else{
1.141 brouard 9745: 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 9746: 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);
9747: return 1;
1.136 brouard 9748: }
9749: andc[i]=(double) year;
9750: moisdc[i]=(double) month;
9751: strcpy(line,stra);
9752:
1.223 brouard 9753: /* Date of birth */
1.136 brouard 9754: cutv(stra, strb,line,' ');
1.169 brouard 9755: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9756: }
1.169 brouard 9757: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9758: month=99;
9759: year=9999;
9760: }else{
1.141 brouard 9761: 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);
9762: 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 9763: return 1;
1.136 brouard 9764: }
9765: if (year==9999) {
1.141 brouard 9766: 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);
9767: 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 9768: return 1;
9769:
1.136 brouard 9770: }
9771: annais[i]=(double)(year);
1.302 brouard 9772: moisnais[i]=(double)(month);
9773: for (j=1;j<=maxwav;j++){
9774: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
9775: 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]);
9776: 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]);
9777: }
9778: }
9779:
1.136 brouard 9780: strcpy(line,stra);
1.225 brouard 9781:
1.223 brouard 9782: /* Sample weight */
1.136 brouard 9783: cutv(stra, strb,line,' ');
9784: errno=0;
9785: dval=strtod(strb,&endptr);
9786: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9787: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9788: 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 9789: fflush(ficlog);
9790: return 1;
9791: }
9792: weight[i]=dval;
9793: strcpy(line,stra);
1.225 brouard 9794:
1.223 brouard 9795: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9796: cutv(stra, strb, line, ' ');
9797: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9798: lval=-1;
1.311 brouard 9799: coqvar[iv][i]=NAN;
9800: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 9801: }else{
1.225 brouard 9802: errno=0;
9803: /* what_kind_of_number(strb); */
9804: dval=strtod(strb,&endptr);
9805: /* if(strb != endptr && *endptr == '\0') */
9806: /* dval=dlval; */
9807: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9808: if( strb[0]=='\0' || (*endptr != '\0')){
9809: 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);
9810: 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);
9811: return 1;
9812: }
9813: coqvar[iv][i]=dval;
1.226 brouard 9814: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9815: }
9816: strcpy(line,stra);
9817: }/* end loop nqv */
1.136 brouard 9818:
1.223 brouard 9819: /* Covariate values */
1.136 brouard 9820: for (j=ncovcol;j>=1;j--){
9821: cutv(stra, strb,line,' ');
1.223 brouard 9822: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9823: lval=-1;
1.136 brouard 9824: }else{
1.225 brouard 9825: errno=0;
9826: lval=strtol(strb,&endptr,10);
9827: if( strb[0]=='\0' || (*endptr != '\0')){
9828: 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);
9829: 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);
9830: return 1;
9831: }
1.136 brouard 9832: }
9833: if(lval <-1 || lval >1){
1.225 brouard 9834: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9835: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9836: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9837: For example, for multinomial values like 1, 2 and 3,\n \
9838: build V1=0 V2=0 for the reference value (1),\n \
9839: V1=1 V2=0 for (2) \n \
1.136 brouard 9840: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9841: output of IMaCh is often meaningless.\n \
1.136 brouard 9842: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9843: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9844: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9845: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9846: For example, for multinomial values like 1, 2 and 3,\n \
9847: build V1=0 V2=0 for the reference value (1),\n \
9848: V1=1 V2=0 for (2) \n \
1.136 brouard 9849: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9850: output of IMaCh is often meaningless.\n \
1.136 brouard 9851: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9852: return 1;
1.136 brouard 9853: }
9854: covar[j][i]=(double)(lval);
9855: strcpy(line,stra);
9856: }
9857: lstra=strlen(stra);
1.225 brouard 9858:
1.136 brouard 9859: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9860: stratrunc = &(stra[lstra-9]);
9861: num[i]=atol(stratrunc);
9862: }
9863: else
9864: num[i]=atol(stra);
9865: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9866: 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;}*/
9867:
9868: i=i+1;
9869: } /* End loop reading data */
1.225 brouard 9870:
1.136 brouard 9871: *imax=i-1; /* Number of individuals */
9872: fclose(fic);
1.225 brouard 9873:
1.136 brouard 9874: return (0);
1.164 brouard 9875: /* endread: */
1.225 brouard 9876: printf("Exiting readdata: ");
9877: fclose(fic);
9878: return (1);
1.223 brouard 9879: }
1.126 brouard 9880:
1.234 brouard 9881: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9882: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9883: while (*p2 == ' ')
1.234 brouard 9884: p2++;
9885: /* while ((*p1++ = *p2++) !=0) */
9886: /* ; */
9887: /* do */
9888: /* while (*p2 == ' ') */
9889: /* p2++; */
9890: /* while (*p1++ == *p2++); */
9891: *stri=p2;
1.145 brouard 9892: }
9893:
1.235 brouard 9894: int decoderesult ( char resultline[], int nres)
1.230 brouard 9895: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9896: {
1.235 brouard 9897: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9898: char resultsav[MAXLINE];
1.234 brouard 9899: int resultmodel[MAXLINE];
9900: int modelresult[MAXLINE];
1.230 brouard 9901: char stra[80], strb[80], strc[80], strd[80],stre[80];
9902:
1.234 brouard 9903: removefirstspace(&resultline);
1.230 brouard 9904:
9905: if (strstr(resultline,"v") !=0){
9906: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9907: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9908: return 1;
9909: }
9910: trimbb(resultsav, resultline);
9911: if (strlen(resultsav) >1){
9912: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9913: }
1.253 brouard 9914: if(j == 0){ /* Resultline but no = */
9915: TKresult[nres]=0; /* Combination for the nresult and the model */
9916: return (0);
9917: }
1.234 brouard 9918: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.318 brouard 9919: printf("ERROR: the number of variables in this result line, %d, differs from the number of variables used in the model line, %d.\n",j, cptcovs);
1.310 brouard 9920: fprintf(ficlog,"ERROR: the number of variables in the resultline, %d, differs from the number of variables used in the model line, %d.\n",j, cptcovs);
1.234 brouard 9921: }
9922: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9923: if(nbocc(resultsav,'=') >1){
1.318 brouard 9924: 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" */
9925: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.234 brouard 9926: }else
9927: cutl(strc,strd,resultsav,'=');
1.318 brouard 9928: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 9929:
1.230 brouard 9930: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 9931: 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 9932: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9933: /* cptcovsel++; */
9934: if (nbocc(stra,'=') >0)
9935: strcpy(resultsav,stra); /* and analyzes it */
9936: }
1.235 brouard 9937: /* Checking for missing or useless values in comparison of current model needs */
1.318 brouard 9938: for(k1=1; k1<= cptcovt ;k1++){ /* Loop on model. model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9939: if(Typevar[k1]==0){ /* Single covariate in model *//*0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 9940: match=0;
1.318 brouard 9941: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9942: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9943: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 9944: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 9945: break;
9946: }
9947: }
9948: if(match == 0){
1.310 brouard 9949: printf("Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
9950: fprintf(ficlog,"Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
9951: return 1;
1.234 brouard 9952: }
9953: }
9954: }
1.235 brouard 9955: /* Checking for missing or useless values in comparison of current model needs */
1.318 brouard 9956: for(k2=1; k2 <=j;k2++){ /* Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9957: match=0;
1.318 brouard 9958: for(k1=1; k1<= cptcovt ;k1++){ /* loop on model: model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.235 brouard 9959: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9960: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.318 brouard 9961: resultmodel[k1]=k2; /* k2th variable of the model corresponds to k1 variable of the model. resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9962: ++match;
9963: }
9964: }
9965: }
9966: if(match == 0){
9967: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
1.310 brouard 9968: fprintf(ficlog,"Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9969: return 1;
1.234 brouard 9970: }else if(match > 1){
9971: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 9972: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9973: return 1;
1.234 brouard 9974: }
9975: }
1.235 brouard 9976:
1.234 brouard 9977: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9978: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9979: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9980: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9981: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9982: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9983: /* 1 0 0 0 */
9984: /* 2 1 0 0 */
9985: /* 3 0 1 0 */
9986: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9987: /* 5 0 0 1 */
9988: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9989: /* 7 0 1 1 */
9990: /* 8 1 1 1 */
1.237 brouard 9991: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9992: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9993: /* V5*age V5 known which value for nres? */
9994: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.318 brouard 9995: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* loop on model line */
1.235 brouard 9996: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9997: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9998: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9999: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 10000: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
10001: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
10002: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 10003: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
10004: k4++;;
10005: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
1.318 brouard 10006: k3q= resultmodel[k1]; /* resultmodel[1(V5)] = 25.1=k3q */
10007: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.237 brouard 10008: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
10009: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
10010: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 10011: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
10012: k4q++;;
10013: }
10014: }
1.234 brouard 10015:
1.235 brouard 10016: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 10017: return (0);
10018: }
1.235 brouard 10019:
1.230 brouard 10020: int decodemodel( char model[], int lastobs)
10021: /**< This routine decodes the model and returns:
1.224 brouard 10022: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
10023: * - nagesqr = 1 if age*age in the model, otherwise 0.
10024: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
10025: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
10026: * - cptcovage number of covariates with age*products =2
10027: * - cptcovs number of simple covariates
10028: * - 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
10029: * which is a new column after the 9 (ncovcol) variables.
1.319 brouard 10030: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 10031: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
10032: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
10033: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
10034: */
1.319 brouard 10035: /* 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 10036: {
1.238 brouard 10037: int i, j, k, ks, v;
1.227 brouard 10038: int j1, k1, k2, k3, k4;
1.136 brouard 10039: char modelsav[80];
1.145 brouard 10040: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 10041: char *strpt;
1.136 brouard 10042:
1.145 brouard 10043: /*removespace(model);*/
1.136 brouard 10044: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 10045: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 10046: if (strstr(model,"AGE") !=0){
1.192 brouard 10047: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
10048: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 10049: return 1;
10050: }
1.141 brouard 10051: if (strstr(model,"v") !=0){
10052: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
10053: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
10054: return 1;
10055: }
1.187 brouard 10056: strcpy(modelsav,model);
10057: if ((strpt=strstr(model,"age*age")) !=0){
10058: printf(" strpt=%s, model=%s\n",strpt, model);
10059: if(strpt != model){
1.234 brouard 10060: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10061: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10062: corresponding column of parameters.\n",model);
1.234 brouard 10063: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10064: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10065: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 10066: return 1;
1.225 brouard 10067: }
1.187 brouard 10068: nagesqr=1;
10069: if (strstr(model,"+age*age") !=0)
1.234 brouard 10070: substrchaine(modelsav, model, "+age*age");
1.187 brouard 10071: else if (strstr(model,"age*age+") !=0)
1.234 brouard 10072: substrchaine(modelsav, model, "age*age+");
1.187 brouard 10073: else
1.234 brouard 10074: substrchaine(modelsav, model, "age*age");
1.187 brouard 10075: }else
10076: nagesqr=0;
10077: if (strlen(modelsav) >1){
10078: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
10079: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 10080: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 10081: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 10082: * cst, age and age*age
10083: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
10084: /* including age products which are counted in cptcovage.
10085: * but the covariates which are products must be treated
10086: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 10087: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
10088: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 10089:
10090:
1.187 brouard 10091: /* Design
10092: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
10093: * < ncovcol=8 >
10094: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
10095: * k= 1 2 3 4 5 6 7 8
10096: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
10097: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 10098: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
10099: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 10100: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
10101: * Tage[++cptcovage]=k
10102: * if products, new covar are created after ncovcol with k1
10103: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
10104: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
10105: * 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
10106: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
10107: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
10108: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
10109: * < ncovcol=8 >
10110: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
10111: * k= 1 2 3 4 5 6 7 8 9 10 11 12
10112: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 10113: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 10114: * p Tprod[1]@2={ 6, 5}
10115: *p Tvard[1][1]@4= {7, 8, 5, 6}
10116: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
10117: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 10118: *How to reorganize? Tvars(orted)
1.187 brouard 10119: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
10120: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
10121: * {2, 1, 4, 8, 5, 6, 3, 7}
10122: * Struct []
10123: */
1.225 brouard 10124:
1.187 brouard 10125: /* This loop fills the array Tvar from the string 'model'.*/
10126: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
10127: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
10128: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
10129: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
10130: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
10131: /* k=1 Tvar[1]=2 (from V2) */
10132: /* k=5 Tvar[5] */
10133: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 10134: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 10135: /* } */
1.198 brouard 10136: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 10137: /*
10138: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 10139: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
10140: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
10141: }
1.187 brouard 10142: cptcovage=0;
1.319 brouard 10143: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
10144: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
10145: 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" */
10146: if (nbocc(modelsav,'+')==0)
10147: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 10148: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
10149: /*scanf("%d",i);*/
1.319 brouard 10150: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
10151: 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 10152: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
10153: /* covar is not filled and then is empty */
10154: cptcovprod--;
10155: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 10156: 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 10157: Typevar[k]=1; /* 1 for age product */
1.319 brouard 10158: cptcovage++; /* Counts the number of covariates which include age as a product */
10159: 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 10160: /*printf("stre=%s ", stre);*/
10161: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
10162: cptcovprod--;
10163: cutl(stre,strb,strc,'V');
10164: Tvar[k]=atoi(stre);
10165: Typevar[k]=1; /* 1 for age product */
10166: cptcovage++;
10167: Tage[cptcovage]=k;
10168: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
10169: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
10170: cptcovn++;
10171: cptcovprodnoage++;k1++;
10172: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
10173: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
10174: because this model-covariate is a construction we invent a new column
10175: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.319 brouard 10176: If already ncovcol=4 and model=V2 + V1 +V1*V4 +age*V3 +V3*V2
10177: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
10178: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=4 etc */
1.234 brouard 10179: Typevar[k]=2; /* 2 for double fixed dummy covariates */
10180: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
10181: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 10182: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 10183: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
10184: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
10185: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
10186: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
10187: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 10188: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 10189: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
10190: for (i=1; i<=lastobs;i++){
10191: /* Computes the new covariate which is a product of
10192: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
10193: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
10194: }
10195: } /* End age is not in the model */
10196: } /* End if model includes a product */
1.319 brouard 10197: else { /* not a product */
1.234 brouard 10198: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
10199: /* scanf("%d",i);*/
10200: cutl(strd,strc,strb,'V');
10201: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
10202: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
10203: Tvar[k]=atoi(strd);
10204: Typevar[k]=0; /* 0 for simple covariates */
10205: }
10206: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 10207: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 10208: scanf("%d",i);*/
1.187 brouard 10209: } /* end of loop + on total covariates */
10210: } /* end if strlen(modelsave == 0) age*age might exist */
10211: } /* end if strlen(model == 0) */
1.136 brouard 10212:
10213: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
10214: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 10215:
1.136 brouard 10216: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 10217: printf("cptcovprod=%d ", cptcovprod);
10218: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
10219: scanf("%d ",i);*/
10220:
10221:
1.230 brouard 10222: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
10223: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 10224: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
10225: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
10226: k = 1 2 3 4 5 6 7 8 9
10227: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 10228: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 10229: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
10230: Dummy[k] 1 0 0 0 3 1 1 2 3
10231: Tmodelind[combination of covar]=k;
1.225 brouard 10232: */
10233: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 10234: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 10235: /* 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 10236: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 10237: printf("Model=1+age+%s\n\
1.227 brouard 10238: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10239: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10240: 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 10241: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 10242: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10243: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10244: 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);
1.285 brouard 10245: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 10246: for(k=1, ncovf=0, nsd=0, nsq=0, ncovv=0, ncova=0, ncoveff=0, nqfveff=0, ntveff=0, nqtveff=0;k<=cptcovt; k++){ /* or cptocvt */
10247: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 10248: Fixed[k]= 0;
10249: Dummy[k]= 0;
1.225 brouard 10250: ncoveff++;
1.232 brouard 10251: ncovf++;
1.234 brouard 10252: nsd++;
10253: modell[k].maintype= FTYPE;
10254: TvarsD[nsd]=Tvar[k];
10255: TvarsDind[nsd]=k;
10256: TvarF[ncovf]=Tvar[k];
10257: TvarFind[ncovf]=k;
10258: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10259: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10260: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
10261: Fixed[k]= 0;
10262: Dummy[k]= 0;
10263: ncoveff++;
10264: ncovf++;
10265: modell[k].maintype= FTYPE;
10266: TvarF[ncovf]=Tvar[k];
10267: TvarFind[ncovf]=k;
1.230 brouard 10268: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10269: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10270: }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 10271: Fixed[k]= 0;
10272: Dummy[k]= 1;
1.230 brouard 10273: nqfveff++;
1.234 brouard 10274: modell[k].maintype= FTYPE;
10275: modell[k].subtype= FQ;
10276: nsq++;
10277: TvarsQ[nsq]=Tvar[k];
10278: TvarsQind[nsq]=k;
1.232 brouard 10279: ncovf++;
1.234 brouard 10280: TvarF[ncovf]=Tvar[k];
10281: TvarFind[ncovf]=k;
1.231 brouard 10282: 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 10283: 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 10284: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10285: Fixed[k]= 1;
10286: Dummy[k]= 0;
1.225 brouard 10287: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10288: modell[k].maintype= VTYPE;
10289: modell[k].subtype= VD;
10290: nsd++;
10291: TvarsD[nsd]=Tvar[k];
10292: TvarsDind[nsd]=k;
10293: ncovv++; /* Only simple time varying variables */
10294: TvarV[ncovv]=Tvar[k];
1.242 brouard 10295: 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 10296: 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 */
10297: 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 10298: 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);
10299: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10300: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10301: Fixed[k]= 1;
10302: Dummy[k]= 1;
10303: nqtveff++;
10304: modell[k].maintype= VTYPE;
10305: modell[k].subtype= VQ;
10306: ncovv++; /* Only simple time varying variables */
10307: nsq++;
1.319 brouard 10308: TvarsQ[nsq]=Tvar[k]; /* k=1 Tvar=5 nsq=1 TvarsQ[1]=5 */
1.234 brouard 10309: TvarsQind[nsq]=k;
10310: TvarV[ncovv]=Tvar[k];
1.242 brouard 10311: 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 10312: 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 */
10313: 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 10314: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10315: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10316: 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);
1.228 brouard 10317: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 10318: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 10319: ncova++;
10320: TvarA[ncova]=Tvar[k];
10321: TvarAind[ncova]=k;
1.231 brouard 10322: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 10323: Fixed[k]= 2;
10324: Dummy[k]= 2;
10325: modell[k].maintype= ATYPE;
10326: modell[k].subtype= APFD;
10327: /* ncoveff++; */
1.227 brouard 10328: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 10329: Fixed[k]= 2;
10330: Dummy[k]= 3;
10331: modell[k].maintype= ATYPE;
10332: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10333: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10334: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10335: Fixed[k]= 3;
10336: Dummy[k]= 2;
10337: modell[k].maintype= ATYPE;
10338: modell[k].subtype= APVD; /* Product age * varying dummy */
10339: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10340: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10341: Fixed[k]= 3;
10342: Dummy[k]= 3;
10343: modell[k].maintype= ATYPE;
10344: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10345: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10346: }
10347: }else if (Typevar[k] == 2) { /* product without age */
10348: k1=Tposprod[k];
10349: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10350: if(Tvard[k1][2] <=ncovcol){
10351: Fixed[k]= 1;
10352: Dummy[k]= 0;
10353: modell[k].maintype= FTYPE;
10354: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10355: ncovf++; /* Fixed variables without age */
10356: TvarF[ncovf]=Tvar[k];
10357: TvarFind[ncovf]=k;
10358: }else if(Tvard[k1][2] <=ncovcol+nqv){
10359: Fixed[k]= 0; /* or 2 ?*/
10360: Dummy[k]= 1;
10361: modell[k].maintype= FTYPE;
10362: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10363: ncovf++; /* Varying variables without age */
10364: TvarF[ncovf]=Tvar[k];
10365: TvarFind[ncovf]=k;
10366: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10367: Fixed[k]= 1;
10368: Dummy[k]= 0;
10369: modell[k].maintype= VTYPE;
10370: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10371: ncovv++; /* Varying variables without age */
10372: TvarV[ncovv]=Tvar[k];
10373: TvarVind[ncovv]=k;
10374: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10375: Fixed[k]= 1;
10376: Dummy[k]= 1;
10377: modell[k].maintype= VTYPE;
10378: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10379: ncovv++; /* Varying variables without age */
10380: TvarV[ncovv]=Tvar[k];
10381: TvarVind[ncovv]=k;
10382: }
1.227 brouard 10383: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10384: if(Tvard[k1][2] <=ncovcol){
10385: Fixed[k]= 0; /* or 2 ?*/
10386: Dummy[k]= 1;
10387: modell[k].maintype= FTYPE;
10388: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10389: ncovf++; /* Fixed variables without age */
10390: TvarF[ncovf]=Tvar[k];
10391: TvarFind[ncovf]=k;
10392: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10393: Fixed[k]= 1;
10394: Dummy[k]= 1;
10395: modell[k].maintype= VTYPE;
10396: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10397: ncovv++; /* Varying variables without age */
10398: TvarV[ncovv]=Tvar[k];
10399: TvarVind[ncovv]=k;
10400: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10401: Fixed[k]= 1;
10402: Dummy[k]= 1;
10403: modell[k].maintype= VTYPE;
10404: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10405: ncovv++; /* Varying variables without age */
10406: TvarV[ncovv]=Tvar[k];
10407: TvarVind[ncovv]=k;
10408: ncovv++; /* Varying variables without age */
10409: TvarV[ncovv]=Tvar[k];
10410: TvarVind[ncovv]=k;
10411: }
1.227 brouard 10412: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10413: if(Tvard[k1][2] <=ncovcol){
10414: Fixed[k]= 1;
10415: Dummy[k]= 1;
10416: modell[k].maintype= VTYPE;
10417: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10418: ncovv++; /* Varying variables without age */
10419: TvarV[ncovv]=Tvar[k];
10420: TvarVind[ncovv]=k;
10421: }else if(Tvard[k1][2] <=ncovcol+nqv){
10422: Fixed[k]= 1;
10423: Dummy[k]= 1;
10424: modell[k].maintype= VTYPE;
10425: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10426: ncovv++; /* Varying variables without age */
10427: TvarV[ncovv]=Tvar[k];
10428: TvarVind[ncovv]=k;
10429: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10430: Fixed[k]= 1;
10431: Dummy[k]= 0;
10432: modell[k].maintype= VTYPE;
10433: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10434: ncovv++; /* Varying variables without age */
10435: TvarV[ncovv]=Tvar[k];
10436: TvarVind[ncovv]=k;
10437: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10438: Fixed[k]= 1;
10439: Dummy[k]= 1;
10440: modell[k].maintype= VTYPE;
10441: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10442: ncovv++; /* Varying variables without age */
10443: TvarV[ncovv]=Tvar[k];
10444: TvarVind[ncovv]=k;
10445: }
1.227 brouard 10446: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10447: if(Tvard[k1][2] <=ncovcol){
10448: Fixed[k]= 1;
10449: Dummy[k]= 1;
10450: modell[k].maintype= VTYPE;
10451: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10452: ncovv++; /* Varying variables without age */
10453: TvarV[ncovv]=Tvar[k];
10454: TvarVind[ncovv]=k;
10455: }else if(Tvard[k1][2] <=ncovcol+nqv){
10456: Fixed[k]= 1;
10457: Dummy[k]= 1;
10458: modell[k].maintype= VTYPE;
10459: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10460: ncovv++; /* Varying variables without age */
10461: TvarV[ncovv]=Tvar[k];
10462: TvarVind[ncovv]=k;
10463: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10464: Fixed[k]= 1;
10465: Dummy[k]= 1;
10466: modell[k].maintype= VTYPE;
10467: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10468: ncovv++; /* Varying variables without age */
10469: TvarV[ncovv]=Tvar[k];
10470: TvarVind[ncovv]=k;
10471: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10472: Fixed[k]= 1;
10473: Dummy[k]= 1;
10474: modell[k].maintype= VTYPE;
10475: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10476: ncovv++; /* Varying variables without age */
10477: TvarV[ncovv]=Tvar[k];
10478: TvarVind[ncovv]=k;
10479: }
1.227 brouard 10480: }else{
1.240 brouard 10481: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10482: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10483: } /*end k1*/
1.225 brouard 10484: }else{
1.226 brouard 10485: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10486: 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 10487: }
1.227 brouard 10488: 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]);
1.231 brouard 10489: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10490: 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]);
10491: }
10492: /* Searching for doublons in the model */
10493: for(k1=1; k1<= cptcovt;k1++){
10494: for(k2=1; k2 <k1;k2++){
1.285 brouard 10495: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10496: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10497: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10498: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10499: printf("Error duplication in the model=%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]);
10500: fprintf(ficlog,"Error duplication in the model=%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 10501: return(1);
10502: }
10503: }else if (Typevar[k1] ==2){
10504: k3=Tposprod[k1];
10505: k4=Tposprod[k2];
10506: 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])) ){
10507: printf("Error duplication in the model=%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]]);
10508: fprintf(ficlog,"Error duplication in the model=%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);
10509: return(1);
10510: }
10511: }
1.227 brouard 10512: }
10513: }
1.225 brouard 10514: }
10515: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10516: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10517: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10518: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10519: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10520: /*endread:*/
1.225 brouard 10521: printf("Exiting decodemodel: ");
10522: return (1);
1.136 brouard 10523: }
10524:
1.169 brouard 10525: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10526: {/* Check ages at death */
1.136 brouard 10527: int i, m;
1.218 brouard 10528: int firstone=0;
10529:
1.136 brouard 10530: for (i=1; i<=imx; i++) {
10531: for(m=2; (m<= maxwav); m++) {
10532: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10533: anint[m][i]=9999;
1.216 brouard 10534: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10535: s[m][i]=-1;
1.136 brouard 10536: }
10537: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10538: *nberr = *nberr + 1;
1.218 brouard 10539: if(firstone == 0){
10540: firstone=1;
1.260 brouard 10541: 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 10542: }
1.262 brouard 10543: 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 10544: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10545: }
10546: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10547: (*nberr)++;
1.259 brouard 10548: 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 10549: 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 10550: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10551: }
10552: }
10553: }
10554:
10555: for (i=1; i<=imx; i++) {
10556: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10557: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10558: 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 10559: if (s[m][i] >= nlstate+1) {
1.169 brouard 10560: if(agedc[i]>0){
10561: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10562: agev[m][i]=agedc[i];
1.214 brouard 10563: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10564: }else {
1.136 brouard 10565: if ((int)andc[i]!=9999){
10566: nbwarn++;
10567: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10568: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10569: agev[m][i]=-1;
10570: }
10571: }
1.169 brouard 10572: } /* agedc > 0 */
1.214 brouard 10573: } /* end if */
1.136 brouard 10574: else if(s[m][i] !=9){ /* Standard case, age in fractional
10575: years but with the precision of a month */
10576: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10577: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10578: agev[m][i]=1;
10579: else if(agev[m][i] < *agemin){
10580: *agemin=agev[m][i];
10581: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10582: }
10583: else if(agev[m][i] >*agemax){
10584: *agemax=agev[m][i];
1.156 brouard 10585: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10586: }
10587: /*agev[m][i]=anint[m][i]-annais[i];*/
10588: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10589: } /* en if 9*/
1.136 brouard 10590: else { /* =9 */
1.214 brouard 10591: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10592: agev[m][i]=1;
10593: s[m][i]=-1;
10594: }
10595: }
1.214 brouard 10596: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10597: agev[m][i]=1;
1.214 brouard 10598: else{
10599: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10600: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10601: agev[m][i]=0;
10602: }
10603: } /* End for lastpass */
10604: }
1.136 brouard 10605:
10606: for (i=1; i<=imx; i++) {
10607: for(m=firstpass; (m<=lastpass); m++){
10608: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10609: (*nberr)++;
1.136 brouard 10610: 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);
10611: 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);
10612: return 1;
10613: }
10614: }
10615: }
10616:
10617: /*for (i=1; i<=imx; i++){
10618: for (m=firstpass; (m<lastpass); m++){
10619: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10620: }
10621:
10622: }*/
10623:
10624:
1.139 brouard 10625: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10626: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10627:
10628: return (0);
1.164 brouard 10629: /* endread:*/
1.136 brouard 10630: printf("Exiting calandcheckages: ");
10631: return (1);
10632: }
10633:
1.172 brouard 10634: #if defined(_MSC_VER)
10635: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10636: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10637: //#include "stdafx.h"
10638: //#include <stdio.h>
10639: //#include <tchar.h>
10640: //#include <windows.h>
10641: //#include <iostream>
10642: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10643:
10644: LPFN_ISWOW64PROCESS fnIsWow64Process;
10645:
10646: BOOL IsWow64()
10647: {
10648: BOOL bIsWow64 = FALSE;
10649:
10650: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10651: // (HANDLE, PBOOL);
10652:
10653: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10654:
10655: HMODULE module = GetModuleHandle(_T("kernel32"));
10656: const char funcName[] = "IsWow64Process";
10657: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10658: GetProcAddress(module, funcName);
10659:
10660: if (NULL != fnIsWow64Process)
10661: {
10662: if (!fnIsWow64Process(GetCurrentProcess(),
10663: &bIsWow64))
10664: //throw std::exception("Unknown error");
10665: printf("Unknown error\n");
10666: }
10667: return bIsWow64 != FALSE;
10668: }
10669: #endif
1.177 brouard 10670:
1.191 brouard 10671: void syscompilerinfo(int logged)
1.292 brouard 10672: {
10673: #include <stdint.h>
10674:
10675: /* #include "syscompilerinfo.h"*/
1.185 brouard 10676: /* command line Intel compiler 32bit windows, XP compatible:*/
10677: /* /GS /W3 /Gy
10678: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10679: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10680: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10681: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10682: */
10683: /* 64 bits */
1.185 brouard 10684: /*
10685: /GS /W3 /Gy
10686: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10687: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10688: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10689: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10690: /* Optimization are useless and O3 is slower than O2 */
10691: /*
10692: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10693: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10694: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10695: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10696: */
1.186 brouard 10697: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10698: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10699: /PDB:"visual studio
10700: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10701: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10702: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10703: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10704: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10705: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10706: uiAccess='false'"
10707: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10708: /NOLOGO /TLBID:1
10709: */
1.292 brouard 10710:
10711:
1.177 brouard 10712: #if defined __INTEL_COMPILER
1.178 brouard 10713: #if defined(__GNUC__)
10714: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10715: #endif
1.177 brouard 10716: #elif defined(__GNUC__)
1.179 brouard 10717: #ifndef __APPLE__
1.174 brouard 10718: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10719: #endif
1.177 brouard 10720: struct utsname sysInfo;
1.178 brouard 10721: int cross = CROSS;
10722: if (cross){
10723: printf("Cross-");
1.191 brouard 10724: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10725: }
1.174 brouard 10726: #endif
10727:
1.191 brouard 10728: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10729: #if defined(__clang__)
1.191 brouard 10730: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10731: #endif
10732: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10733: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10734: #endif
10735: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10736: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10737: #endif
10738: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10739: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10740: #endif
10741: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10742: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10743: #endif
10744: #if defined(_MSC_VER)
1.191 brouard 10745: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10746: #endif
10747: #if defined(__PGI)
1.191 brouard 10748: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10749: #endif
10750: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10751: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10752: #endif
1.191 brouard 10753: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10754:
1.167 brouard 10755: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10756: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10757: // Windows (x64 and x86)
1.191 brouard 10758: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10759: #elif __unix__ // all unices, not all compilers
10760: // Unix
1.191 brouard 10761: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10762: #elif __linux__
10763: // linux
1.191 brouard 10764: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10765: #elif __APPLE__
1.174 brouard 10766: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10767: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10768: #endif
10769:
10770: /* __MINGW32__ */
10771: /* __CYGWIN__ */
10772: /* __MINGW64__ */
10773: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10774: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10775: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10776: /* _WIN64 // Defined for applications for Win64. */
10777: /* _M_X64 // Defined for compilations that target x64 processors. */
10778: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10779:
1.167 brouard 10780: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10781: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10782: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10783: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10784: #else
1.191 brouard 10785: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10786: #endif
10787:
1.169 brouard 10788: #if defined(__GNUC__)
10789: # if defined(__GNUC_PATCHLEVEL__)
10790: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10791: + __GNUC_MINOR__ * 100 \
10792: + __GNUC_PATCHLEVEL__)
10793: # else
10794: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10795: + __GNUC_MINOR__ * 100)
10796: # endif
1.174 brouard 10797: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10798: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10799:
10800: if (uname(&sysInfo) != -1) {
10801: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10802: 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 10803: }
10804: else
10805: perror("uname() error");
1.179 brouard 10806: //#ifndef __INTEL_COMPILER
10807: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10808: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10809: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10810: #endif
1.169 brouard 10811: #endif
1.172 brouard 10812:
1.286 brouard 10813: // void main ()
1.172 brouard 10814: // {
1.169 brouard 10815: #if defined(_MSC_VER)
1.174 brouard 10816: if (IsWow64()){
1.191 brouard 10817: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10818: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10819: }
10820: else{
1.191 brouard 10821: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10822: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10823: }
1.172 brouard 10824: // printf("\nPress Enter to continue...");
10825: // getchar();
10826: // }
10827:
1.169 brouard 10828: #endif
10829:
1.167 brouard 10830:
1.219 brouard 10831: }
1.136 brouard 10832:
1.219 brouard 10833: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10834: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10835: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10836: /* double ftolpl = 1.e-10; */
1.180 brouard 10837: double age, agebase, agelim;
1.203 brouard 10838: double tot;
1.180 brouard 10839:
1.202 brouard 10840: strcpy(filerespl,"PL_");
10841: strcat(filerespl,fileresu);
10842: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10843: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10844: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10845: }
1.288 brouard 10846: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10847: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10848: pstamp(ficrespl);
1.288 brouard 10849: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10850: fprintf(ficrespl,"#Age ");
10851: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10852: fprintf(ficrespl,"\n");
1.180 brouard 10853:
1.219 brouard 10854: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10855:
1.219 brouard 10856: agebase=ageminpar;
10857: agelim=agemaxpar;
1.180 brouard 10858:
1.227 brouard 10859: /* i1=pow(2,ncoveff); */
1.234 brouard 10860: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10861: if (cptcovn < 1){i1=1;}
1.180 brouard 10862:
1.238 brouard 10863: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10864: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10865: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10866: continue;
1.235 brouard 10867:
1.238 brouard 10868: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10869: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10870: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10871: /* k=k+1; */
10872: /* to clean */
10873: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10874: fprintf(ficrespl,"#******");
10875: printf("#******");
10876: fprintf(ficlog,"#******");
10877: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10878: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10879: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10880: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10881: }
10882: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10883: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10884: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10885: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10886: }
10887: fprintf(ficrespl,"******\n");
10888: printf("******\n");
10889: fprintf(ficlog,"******\n");
10890: if(invalidvarcomb[k]){
10891: printf("\nCombination (%d) ignored because no case \n",k);
10892: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10893: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10894: continue;
10895: }
1.219 brouard 10896:
1.238 brouard 10897: fprintf(ficrespl,"#Age ");
10898: for(j=1;j<=cptcoveff;j++) {
10899: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10900: }
10901: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10902: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10903:
1.238 brouard 10904: for (age=agebase; age<=agelim; age++){
10905: /* for (age=agebase; age<=agebase; age++){ */
10906: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10907: fprintf(ficrespl,"%.0f ",age );
10908: for(j=1;j<=cptcoveff;j++)
10909: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10910: tot=0.;
10911: for(i=1; i<=nlstate;i++){
10912: tot += prlim[i][i];
10913: fprintf(ficrespl," %.5f", prlim[i][i]);
10914: }
10915: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10916: } /* Age */
10917: /* was end of cptcod */
10918: } /* cptcov */
10919: } /* nres */
1.219 brouard 10920: return 0;
1.180 brouard 10921: }
10922:
1.218 brouard 10923: 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 10924: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10925:
10926: /* Computes the back prevalence limit for any combination of covariate values
10927: * at any age between ageminpar and agemaxpar
10928: */
1.235 brouard 10929: int i, j, k, i1, nres=0 ;
1.217 brouard 10930: /* double ftolpl = 1.e-10; */
10931: double age, agebase, agelim;
10932: double tot;
1.218 brouard 10933: /* double ***mobaverage; */
10934: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10935:
10936: strcpy(fileresplb,"PLB_");
10937: strcat(fileresplb,fileresu);
10938: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 10939: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
10940: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 10941: }
1.288 brouard 10942: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
10943: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 10944: pstamp(ficresplb);
1.288 brouard 10945: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 10946: fprintf(ficresplb,"#Age ");
10947: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10948: fprintf(ficresplb,"\n");
10949:
1.218 brouard 10950:
10951: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10952:
10953: agebase=ageminpar;
10954: agelim=agemaxpar;
10955:
10956:
1.227 brouard 10957: i1=pow(2,cptcoveff);
1.218 brouard 10958: if (cptcovn < 1){i1=1;}
1.227 brouard 10959:
1.238 brouard 10960: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10961: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10962: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10963: continue;
10964: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10965: fprintf(ficresplb,"#******");
10966: printf("#******");
10967: fprintf(ficlog,"#******");
10968: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10969: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10970: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10971: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10972: }
10973: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10974: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10975: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10976: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10977: }
10978: fprintf(ficresplb,"******\n");
10979: printf("******\n");
10980: fprintf(ficlog,"******\n");
10981: if(invalidvarcomb[k]){
10982: printf("\nCombination (%d) ignored because no cases \n",k);
10983: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10984: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10985: continue;
10986: }
1.218 brouard 10987:
1.238 brouard 10988: fprintf(ficresplb,"#Age ");
10989: for(j=1;j<=cptcoveff;j++) {
10990: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10991: }
10992: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10993: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10994:
10995:
1.238 brouard 10996: for (age=agebase; age<=agelim; age++){
10997: /* for (age=agebase; age<=agebase; age++){ */
10998: if(mobilavproj > 0){
10999: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
11000: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11001: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 11002: }else if (mobilavproj == 0){
11003: 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);
11004: 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);
11005: exit(1);
11006: }else{
11007: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11008: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 11009: /* printf("TOTOT\n"); */
11010: /* exit(1); */
1.238 brouard 11011: }
11012: fprintf(ficresplb,"%.0f ",age );
11013: for(j=1;j<=cptcoveff;j++)
11014: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11015: tot=0.;
11016: for(i=1; i<=nlstate;i++){
11017: tot += bprlim[i][i];
11018: fprintf(ficresplb," %.5f", bprlim[i][i]);
11019: }
11020: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
11021: } /* Age */
11022: /* was end of cptcod */
1.255 brouard 11023: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 11024: } /* end of any combination */
11025: } /* end of nres */
1.218 brouard 11026: /* hBijx(p, bage, fage); */
11027: /* fclose(ficrespijb); */
11028:
11029: return 0;
1.217 brouard 11030: }
1.218 brouard 11031:
1.180 brouard 11032: int hPijx(double *p, int bage, int fage){
11033: /*------------- h Pij x at various ages ------------*/
11034:
11035: int stepsize;
11036: int agelim;
11037: int hstepm;
11038: int nhstepm;
1.235 brouard 11039: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 11040:
11041: double agedeb;
11042: double ***p3mat;
11043:
1.201 brouard 11044: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 11045: if((ficrespij=fopen(filerespij,"w"))==NULL) {
11046: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
11047: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
11048: }
11049: printf("Computing pij: result on file '%s' \n", filerespij);
11050: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
11051:
11052: stepsize=(int) (stepm+YEARM-1)/YEARM;
11053: /*if (stepm<=24) stepsize=2;*/
11054:
11055: agelim=AGESUP;
11056: hstepm=stepsize*YEARM; /* Every year of age */
11057: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 11058:
1.180 brouard 11059: /* hstepm=1; aff par mois*/
11060: pstamp(ficrespij);
11061: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 11062: i1= pow(2,cptcoveff);
1.218 brouard 11063: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11064: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11065: /* k=k+1; */
1.235 brouard 11066: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11067: for(k=1; k<=i1;k++){
1.253 brouard 11068: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11069: continue;
1.183 brouard 11070: fprintf(ficrespij,"\n#****** ");
1.227 brouard 11071: for(j=1;j<=cptcoveff;j++)
1.198 brouard 11072: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11073: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11074: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11075: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11076: }
1.183 brouard 11077: fprintf(ficrespij,"******\n");
11078:
11079: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
11080: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
11081: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
11082:
11083: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 11084:
1.183 brouard 11085: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11086: oldm=oldms;savm=savms;
1.235 brouard 11087: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 11088: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
11089: for(i=1; i<=nlstate;i++)
11090: for(j=1; j<=nlstate+ndeath;j++)
11091: fprintf(ficrespij," %1d-%1d",i,j);
11092: fprintf(ficrespij,"\n");
11093: for (h=0; h<=nhstepm; h++){
11094: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11095: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 11096: for(i=1; i<=nlstate;i++)
11097: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 11098: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 11099: fprintf(ficrespij,"\n");
11100: }
1.183 brouard 11101: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11102: fprintf(ficrespij,"\n");
11103: }
1.180 brouard 11104: /*}*/
11105: }
1.218 brouard 11106: return 0;
1.180 brouard 11107: }
1.218 brouard 11108:
11109: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 11110: /*------------- h Bij x at various ages ------------*/
11111:
11112: int stepsize;
1.218 brouard 11113: /* int agelim; */
11114: int ageminl;
1.217 brouard 11115: int hstepm;
11116: int nhstepm;
1.238 brouard 11117: int h, i, i1, j, k, nres;
1.218 brouard 11118:
1.217 brouard 11119: double agedeb;
11120: double ***p3mat;
1.218 brouard 11121:
11122: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
11123: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
11124: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11125: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11126: }
11127: printf("Computing pij back: result on file '%s' \n", filerespijb);
11128: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
11129:
11130: stepsize=(int) (stepm+YEARM-1)/YEARM;
11131: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 11132:
1.218 brouard 11133: /* agelim=AGESUP; */
1.289 brouard 11134: ageminl=AGEINF; /* was 30 */
1.218 brouard 11135: hstepm=stepsize*YEARM; /* Every year of age */
11136: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
11137:
11138: /* hstepm=1; aff par mois*/
11139: pstamp(ficrespijb);
1.255 brouard 11140: 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 11141: i1= pow(2,cptcoveff);
1.218 brouard 11142: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11143: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11144: /* k=k+1; */
1.238 brouard 11145: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11146: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11147: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11148: continue;
11149: fprintf(ficrespijb,"\n#****** ");
11150: for(j=1;j<=cptcoveff;j++)
11151: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11152: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11153: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11154: }
11155: fprintf(ficrespijb,"******\n");
1.264 brouard 11156: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 11157: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
11158: continue;
11159: }
11160:
11161: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
11162: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
11163: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 11164: 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 */
11165: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 11166:
11167: /* nhstepm=nhstepm*YEARM; aff par mois*/
11168:
1.266 brouard 11169: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
11170: /* and memory limitations if stepm is small */
11171:
1.238 brouard 11172: /* oldm=oldms;savm=savms; */
11173: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 11174: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 11175: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 11176: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 11177: for(i=1; i<=nlstate;i++)
11178: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 11179: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 11180: fprintf(ficrespijb,"\n");
1.238 brouard 11181: for (h=0; h<=nhstepm; h++){
11182: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11183: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
11184: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
11185: for(i=1; i<=nlstate;i++)
11186: for(j=1; j<=nlstate+ndeath;j++)
11187: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
11188: fprintf(ficrespijb,"\n");
11189: }
11190: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11191: fprintf(ficrespijb,"\n");
11192: } /* end age deb */
11193: } /* end combination */
11194: } /* end nres */
1.218 brouard 11195: return 0;
11196: } /* hBijx */
1.217 brouard 11197:
1.180 brouard 11198:
1.136 brouard 11199: /***********************************************/
11200: /**************** Main Program *****************/
11201: /***********************************************/
11202:
11203: int main(int argc, char *argv[])
11204: {
11205: #ifdef GSL
11206: const gsl_multimin_fminimizer_type *T;
11207: size_t iteri = 0, it;
11208: int rval = GSL_CONTINUE;
11209: int status = GSL_SUCCESS;
11210: double ssval;
11211: #endif
11212: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 11213: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
11214: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 11215: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 11216: int jj, ll, li, lj, lk;
1.136 brouard 11217: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 11218: int num_filled;
1.136 brouard 11219: int itimes;
11220: int NDIM=2;
11221: int vpopbased=0;
1.235 brouard 11222: int nres=0;
1.258 brouard 11223: int endishere=0;
1.277 brouard 11224: int noffset=0;
1.274 brouard 11225: int ncurrv=0; /* Temporary variable */
11226:
1.164 brouard 11227: char ca[32], cb[32];
1.136 brouard 11228: /* FILE *fichtm; *//* Html File */
11229: /* FILE *ficgp;*/ /*Gnuplot File */
11230: struct stat info;
1.191 brouard 11231: double agedeb=0.;
1.194 brouard 11232:
11233: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 11234: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 11235:
1.165 brouard 11236: double fret;
1.191 brouard 11237: double dum=0.; /* Dummy variable */
1.136 brouard 11238: double ***p3mat;
1.218 brouard 11239: /* double ***mobaverage; */
1.319 brouard 11240: double wald;
1.164 brouard 11241:
11242: char line[MAXLINE];
1.197 brouard 11243: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
11244:
1.234 brouard 11245: char modeltemp[MAXLINE];
1.230 brouard 11246: char resultline[MAXLINE];
11247:
1.136 brouard 11248: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 11249: char *tok, *val; /* pathtot */
1.290 brouard 11250: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 11251: int c, h , cpt, c2;
1.191 brouard 11252: int jl=0;
11253: int i1, j1, jk, stepsize=0;
1.194 brouard 11254: int count=0;
11255:
1.164 brouard 11256: int *tab;
1.136 brouard 11257: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 11258: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
11259: /* double anprojf, mprojf, jprojf; */
11260: /* double jintmean,mintmean,aintmean; */
11261: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11262: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11263: double yrfproj= 10.0; /* Number of years of forward projections */
11264: double yrbproj= 10.0; /* Number of years of backward projections */
11265: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11266: int mobilav=0,popforecast=0;
1.191 brouard 11267: int hstepm=0, nhstepm=0;
1.136 brouard 11268: int agemortsup;
11269: float sumlpop=0.;
11270: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11271: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11272:
1.191 brouard 11273: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11274: double ftolpl=FTOL;
11275: double **prlim;
1.217 brouard 11276: double **bprlim;
1.317 brouard 11277: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
11278: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 11279: double ***paramstart; /* Matrix of starting parameter values */
11280: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11281: double **matcov; /* Matrix of covariance */
1.203 brouard 11282: double **hess; /* Hessian matrix */
1.136 brouard 11283: double ***delti3; /* Scale */
11284: double *delti; /* Scale */
11285: double ***eij, ***vareij;
11286: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11287:
1.136 brouard 11288: double *epj, vepp;
1.164 brouard 11289:
1.273 brouard 11290: double dateprev1, dateprev2;
1.296 brouard 11291: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11292: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11293:
1.217 brouard 11294:
1.136 brouard 11295: double **ximort;
1.145 brouard 11296: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11297: int *dcwave;
11298:
1.164 brouard 11299: char z[1]="c";
1.136 brouard 11300:
11301: /*char *strt;*/
11302: char strtend[80];
1.126 brouard 11303:
1.164 brouard 11304:
1.126 brouard 11305: /* setlocale (LC_ALL, ""); */
11306: /* bindtextdomain (PACKAGE, LOCALEDIR); */
11307: /* textdomain (PACKAGE); */
11308: /* setlocale (LC_CTYPE, ""); */
11309: /* setlocale (LC_MESSAGES, ""); */
11310:
11311: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 11312: rstart_time = time(NULL);
11313: /* (void) gettimeofday(&start_time,&tzp);*/
11314: start_time = *localtime(&rstart_time);
1.126 brouard 11315: curr_time=start_time;
1.157 brouard 11316: /*tml = *localtime(&start_time.tm_sec);*/
11317: /* strcpy(strstart,asctime(&tml)); */
11318: strcpy(strstart,asctime(&start_time));
1.126 brouard 11319:
11320: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 11321: /* tp.tm_sec = tp.tm_sec +86400; */
11322: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 11323: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
11324: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
11325: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 11326: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 11327: /* strt=asctime(&tmg); */
11328: /* printf("Time(after) =%s",strstart); */
11329: /* (void) time (&time_value);
11330: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
11331: * tm = *localtime(&time_value);
11332: * strstart=asctime(&tm);
11333: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11334: */
11335:
11336: nberr=0; /* Number of errors and warnings */
11337: nbwarn=0;
1.184 brouard 11338: #ifdef WIN32
11339: _getcwd(pathcd, size);
11340: #else
1.126 brouard 11341: getcwd(pathcd, size);
1.184 brouard 11342: #endif
1.191 brouard 11343: syscompilerinfo(0);
1.196 brouard 11344: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11345: if(argc <=1){
11346: printf("\nEnter the parameter file name: ");
1.205 brouard 11347: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11348: printf("ERROR Empty parameter file name\n");
11349: goto end;
11350: }
1.126 brouard 11351: i=strlen(pathr);
11352: if(pathr[i-1]=='\n')
11353: pathr[i-1]='\0';
1.156 brouard 11354: i=strlen(pathr);
1.205 brouard 11355: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11356: pathr[i-1]='\0';
1.205 brouard 11357: }
11358: i=strlen(pathr);
11359: if( i==0 ){
11360: printf("ERROR Empty parameter file name\n");
11361: goto end;
11362: }
11363: for (tok = pathr; tok != NULL; ){
1.126 brouard 11364: printf("Pathr |%s|\n",pathr);
11365: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11366: printf("val= |%s| pathr=%s\n",val,pathr);
11367: strcpy (pathtot, val);
11368: if(pathr[0] == '\0') break; /* Dirty */
11369: }
11370: }
1.281 brouard 11371: else if (argc<=2){
11372: strcpy(pathtot,argv[1]);
11373: }
1.126 brouard 11374: else{
11375: strcpy(pathtot,argv[1]);
1.281 brouard 11376: strcpy(z,argv[2]);
11377: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11378: }
11379: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11380: /*cygwin_split_path(pathtot,path,optionfile);
11381: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11382: /* cutv(path,optionfile,pathtot,'\\');*/
11383:
11384: /* Split argv[0], imach program to get pathimach */
11385: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11386: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11387: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11388: /* strcpy(pathimach,argv[0]); */
11389: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11390: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11391: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11392: #ifdef WIN32
11393: _chdir(path); /* Can be a relative path */
11394: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11395: #else
1.126 brouard 11396: chdir(path); /* Can be a relative path */
1.184 brouard 11397: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11398: #endif
11399: printf("Current directory %s!\n",pathcd);
1.126 brouard 11400: strcpy(command,"mkdir ");
11401: strcat(command,optionfilefiname);
11402: if((outcmd=system(command)) != 0){
1.169 brouard 11403: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11404: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11405: /* fclose(ficlog); */
11406: /* exit(1); */
11407: }
11408: /* if((imk=mkdir(optionfilefiname))<0){ */
11409: /* perror("mkdir"); */
11410: /* } */
11411:
11412: /*-------- arguments in the command line --------*/
11413:
1.186 brouard 11414: /* Main Log file */
1.126 brouard 11415: strcat(filelog, optionfilefiname);
11416: strcat(filelog,".log"); /* */
11417: if((ficlog=fopen(filelog,"w"))==NULL) {
11418: printf("Problem with logfile %s\n",filelog);
11419: goto end;
11420: }
11421: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11422: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11423: fprintf(ficlog,"\nEnter the parameter file name: \n");
11424: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11425: path=%s \n\
11426: optionfile=%s\n\
11427: optionfilext=%s\n\
1.156 brouard 11428: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11429:
1.197 brouard 11430: syscompilerinfo(1);
1.167 brouard 11431:
1.126 brouard 11432: printf("Local time (at start):%s",strstart);
11433: fprintf(ficlog,"Local time (at start): %s",strstart);
11434: fflush(ficlog);
11435: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11436: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11437:
11438: /* */
11439: strcpy(fileres,"r");
11440: strcat(fileres, optionfilefiname);
1.201 brouard 11441: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11442: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11443: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11444:
1.186 brouard 11445: /* Main ---------arguments file --------*/
1.126 brouard 11446:
11447: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11448: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11449: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11450: fflush(ficlog);
1.149 brouard 11451: /* goto end; */
11452: exit(70);
1.126 brouard 11453: }
11454:
11455: strcpy(filereso,"o");
1.201 brouard 11456: strcat(filereso,fileresu);
1.126 brouard 11457: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11458: printf("Problem with Output resultfile: %s\n", filereso);
11459: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11460: fflush(ficlog);
11461: goto end;
11462: }
1.278 brouard 11463: /*-------- Rewriting parameter file ----------*/
11464: strcpy(rfileres,"r"); /* "Rparameterfile */
11465: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11466: strcat(rfileres,"."); /* */
11467: strcat(rfileres,optionfilext); /* Other files have txt extension */
11468: if((ficres =fopen(rfileres,"w"))==NULL) {
11469: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11470: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11471: fflush(ficlog);
11472: goto end;
11473: }
11474: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11475:
1.278 brouard 11476:
1.126 brouard 11477: /* Reads comments: lines beginning with '#' */
11478: numlinepar=0;
1.277 brouard 11479: /* Is it a BOM UTF-8 Windows file? */
11480: /* First parameter line */
1.197 brouard 11481: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11482: noffset=0;
11483: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11484: {
11485: noffset=noffset+3;
11486: printf("# File is an UTF8 Bom.\n"); // 0xBF
11487: }
1.302 brouard 11488: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
11489: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 11490: {
11491: noffset=noffset+2;
11492: printf("# File is an UTF16BE BOM file\n");
11493: }
11494: else if( line[0] == 0 && line[1] == 0)
11495: {
11496: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11497: noffset=noffset+4;
11498: printf("# File is an UTF16BE BOM file\n");
11499: }
11500: } else{
11501: ;/*printf(" Not a BOM file\n");*/
11502: }
11503:
1.197 brouard 11504: /* If line starts with a # it is a comment */
1.277 brouard 11505: if (line[noffset] == '#') {
1.197 brouard 11506: numlinepar++;
11507: fputs(line,stdout);
11508: fputs(line,ficparo);
1.278 brouard 11509: fputs(line,ficres);
1.197 brouard 11510: fputs(line,ficlog);
11511: continue;
11512: }else
11513: break;
11514: }
11515: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11516: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11517: if (num_filled != 5) {
11518: printf("Should be 5 parameters\n");
1.283 brouard 11519: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11520: }
1.126 brouard 11521: numlinepar++;
1.197 brouard 11522: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11523: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11524: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11525: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11526: }
11527: /* Second parameter line */
11528: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11529: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11530: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11531: if (line[0] == '#') {
11532: numlinepar++;
1.283 brouard 11533: printf("%s",line);
11534: fprintf(ficres,"%s",line);
11535: fprintf(ficparo,"%s",line);
11536: fprintf(ficlog,"%s",line);
1.197 brouard 11537: continue;
11538: }else
11539: break;
11540: }
1.223 brouard 11541: 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", \
11542: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11543: if (num_filled != 11) {
11544: 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 11545: printf("but line=%s\n",line);
1.283 brouard 11546: 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");
11547: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11548: }
1.286 brouard 11549: if( lastpass > maxwav){
11550: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11551: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11552: fflush(ficlog);
11553: goto end;
11554: }
11555: 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 11556: 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 11557: 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 11558: 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 11559: }
1.203 brouard 11560: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11561: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11562: /* Third parameter line */
11563: while(fgets(line, MAXLINE, ficpar)) {
11564: /* If line starts with a # it is a comment */
11565: if (line[0] == '#') {
11566: numlinepar++;
1.283 brouard 11567: printf("%s",line);
11568: fprintf(ficres,"%s",line);
11569: fprintf(ficparo,"%s",line);
11570: fprintf(ficlog,"%s",line);
1.197 brouard 11571: continue;
11572: }else
11573: break;
11574: }
1.201 brouard 11575: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11576: if (num_filled != 1){
1.302 brouard 11577: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
11578: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 11579: model[0]='\0';
11580: goto end;
11581: }
11582: else{
11583: if (model[0]=='+'){
11584: for(i=1; i<=strlen(model);i++)
11585: modeltemp[i-1]=model[i];
1.201 brouard 11586: strcpy(model,modeltemp);
1.197 brouard 11587: }
11588: }
1.199 brouard 11589: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11590: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11591: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11592: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11593: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11594: }
11595: /* 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); */
11596: /* numlinepar=numlinepar+3; /\* In general *\/ */
11597: /* 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 11598: /* 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); */
11599: /* 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 11600: fflush(ficlog);
1.190 brouard 11601: /* if(model[0]=='#'|| model[0]== '\0'){ */
11602: if(model[0]=='#'){
1.279 brouard 11603: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11604: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11605: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11606: if(mle != -1){
1.279 brouard 11607: 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 11608: exit(1);
11609: }
11610: }
1.126 brouard 11611: while((c=getc(ficpar))=='#' && c!= EOF){
11612: ungetc(c,ficpar);
11613: fgets(line, MAXLINE, ficpar);
11614: numlinepar++;
1.195 brouard 11615: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11616: z[0]=line[1];
11617: }
11618: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11619: fputs(line, stdout);
11620: //puts(line);
1.126 brouard 11621: fputs(line,ficparo);
11622: fputs(line,ficlog);
11623: }
11624: ungetc(c,ficpar);
11625:
11626:
1.290 brouard 11627: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11628: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11629: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11630: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11631: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11632: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11633: v1+v2*age+v2*v3 makes cptcovn = 3
11634: */
11635: if (strlen(model)>1)
1.187 brouard 11636: 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 11637: else
1.187 brouard 11638: ncovmodel=2; /* Constant and age */
1.133 brouard 11639: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11640: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11641: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11642: 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);
11643: 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);
11644: fflush(stdout);
11645: fclose (ficlog);
11646: goto end;
11647: }
1.126 brouard 11648: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11649: delti=delti3[1][1];
11650: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11651: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11652: /* We could also provide initial parameters values giving by simple logistic regression
11653: * only one way, that is without matrix product. We will have nlstate maximizations */
11654: /* for(i=1;i<nlstate;i++){ */
11655: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11656: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11657: /* } */
1.126 brouard 11658: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11659: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11660: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11661: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11662: fclose (ficparo);
11663: fclose (ficlog);
11664: goto end;
11665: exit(0);
1.220 brouard 11666: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11667: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11668: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11669: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11670: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11671: matcov=matrix(1,npar,1,npar);
1.203 brouard 11672: hess=matrix(1,npar,1,npar);
1.220 brouard 11673: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11674: /* Read guessed parameters */
1.126 brouard 11675: /* Reads comments: lines beginning with '#' */
11676: while((c=getc(ficpar))=='#' && c!= EOF){
11677: ungetc(c,ficpar);
11678: fgets(line, MAXLINE, ficpar);
11679: numlinepar++;
1.141 brouard 11680: fputs(line,stdout);
1.126 brouard 11681: fputs(line,ficparo);
11682: fputs(line,ficlog);
11683: }
11684: ungetc(c,ficpar);
11685:
11686: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11687: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11688: for(i=1; i <=nlstate; i++){
1.234 brouard 11689: j=0;
1.126 brouard 11690: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11691: if(jj==i) continue;
11692: j++;
1.292 brouard 11693: while((c=getc(ficpar))=='#' && c!= EOF){
11694: ungetc(c,ficpar);
11695: fgets(line, MAXLINE, ficpar);
11696: numlinepar++;
11697: fputs(line,stdout);
11698: fputs(line,ficparo);
11699: fputs(line,ficlog);
11700: }
11701: ungetc(c,ficpar);
1.234 brouard 11702: fscanf(ficpar,"%1d%1d",&i1,&j1);
11703: if ((i1 != i) || (j1 != jj)){
11704: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11705: It might be a problem of design; if ncovcol and the model are correct\n \
11706: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11707: exit(1);
11708: }
11709: fprintf(ficparo,"%1d%1d",i1,j1);
11710: if(mle==1)
11711: printf("%1d%1d",i,jj);
11712: fprintf(ficlog,"%1d%1d",i,jj);
11713: for(k=1; k<=ncovmodel;k++){
11714: fscanf(ficpar," %lf",¶m[i][j][k]);
11715: if(mle==1){
11716: printf(" %lf",param[i][j][k]);
11717: fprintf(ficlog," %lf",param[i][j][k]);
11718: }
11719: else
11720: fprintf(ficlog," %lf",param[i][j][k]);
11721: fprintf(ficparo," %lf",param[i][j][k]);
11722: }
11723: fscanf(ficpar,"\n");
11724: numlinepar++;
11725: if(mle==1)
11726: printf("\n");
11727: fprintf(ficlog,"\n");
11728: fprintf(ficparo,"\n");
1.126 brouard 11729: }
11730: }
11731: fflush(ficlog);
1.234 brouard 11732:
1.251 brouard 11733: /* Reads parameters values */
1.126 brouard 11734: p=param[1][1];
1.251 brouard 11735: pstart=paramstart[1][1];
1.126 brouard 11736:
11737: /* Reads comments: lines beginning with '#' */
11738: while((c=getc(ficpar))=='#' && c!= EOF){
11739: ungetc(c,ficpar);
11740: fgets(line, MAXLINE, ficpar);
11741: numlinepar++;
1.141 brouard 11742: fputs(line,stdout);
1.126 brouard 11743: fputs(line,ficparo);
11744: fputs(line,ficlog);
11745: }
11746: ungetc(c,ficpar);
11747:
11748: for(i=1; i <=nlstate; i++){
11749: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11750: fscanf(ficpar,"%1d%1d",&i1,&j1);
11751: if ( (i1-i) * (j1-j) != 0){
11752: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11753: exit(1);
11754: }
11755: printf("%1d%1d",i,j);
11756: fprintf(ficparo,"%1d%1d",i1,j1);
11757: fprintf(ficlog,"%1d%1d",i1,j1);
11758: for(k=1; k<=ncovmodel;k++){
11759: fscanf(ficpar,"%le",&delti3[i][j][k]);
11760: printf(" %le",delti3[i][j][k]);
11761: fprintf(ficparo," %le",delti3[i][j][k]);
11762: fprintf(ficlog," %le",delti3[i][j][k]);
11763: }
11764: fscanf(ficpar,"\n");
11765: numlinepar++;
11766: printf("\n");
11767: fprintf(ficparo,"\n");
11768: fprintf(ficlog,"\n");
1.126 brouard 11769: }
11770: }
11771: fflush(ficlog);
1.234 brouard 11772:
1.145 brouard 11773: /* Reads covariance matrix */
1.126 brouard 11774: delti=delti3[1][1];
1.220 brouard 11775:
11776:
1.126 brouard 11777: /* 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 11778:
1.126 brouard 11779: /* Reads comments: lines beginning with '#' */
11780: while((c=getc(ficpar))=='#' && c!= EOF){
11781: ungetc(c,ficpar);
11782: fgets(line, MAXLINE, ficpar);
11783: numlinepar++;
1.141 brouard 11784: fputs(line,stdout);
1.126 brouard 11785: fputs(line,ficparo);
11786: fputs(line,ficlog);
11787: }
11788: ungetc(c,ficpar);
1.220 brouard 11789:
1.126 brouard 11790: matcov=matrix(1,npar,1,npar);
1.203 brouard 11791: hess=matrix(1,npar,1,npar);
1.131 brouard 11792: for(i=1; i <=npar; i++)
11793: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11794:
1.194 brouard 11795: /* Scans npar lines */
1.126 brouard 11796: for(i=1; i <=npar; i++){
1.226 brouard 11797: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11798: if(count != 3){
1.226 brouard 11799: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11800: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11801: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11802: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11803: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11804: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11805: exit(1);
1.220 brouard 11806: }else{
1.226 brouard 11807: if(mle==1)
11808: printf("%1d%1d%d",i1,j1,jk);
11809: }
11810: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11811: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11812: for(j=1; j <=i; j++){
1.226 brouard 11813: fscanf(ficpar," %le",&matcov[i][j]);
11814: if(mle==1){
11815: printf(" %.5le",matcov[i][j]);
11816: }
11817: fprintf(ficlog," %.5le",matcov[i][j]);
11818: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11819: }
11820: fscanf(ficpar,"\n");
11821: numlinepar++;
11822: if(mle==1)
1.220 brouard 11823: printf("\n");
1.126 brouard 11824: fprintf(ficlog,"\n");
11825: fprintf(ficparo,"\n");
11826: }
1.194 brouard 11827: /* End of read covariance matrix npar lines */
1.126 brouard 11828: for(i=1; i <=npar; i++)
11829: for(j=i+1;j<=npar;j++)
1.226 brouard 11830: matcov[i][j]=matcov[j][i];
1.126 brouard 11831:
11832: if(mle==1)
11833: printf("\n");
11834: fprintf(ficlog,"\n");
11835:
11836: fflush(ficlog);
11837:
11838: } /* End of mle != -3 */
1.218 brouard 11839:
1.186 brouard 11840: /* Main data
11841: */
1.290 brouard 11842: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11843: /* num=lvector(1,n); */
11844: /* moisnais=vector(1,n); */
11845: /* annais=vector(1,n); */
11846: /* moisdc=vector(1,n); */
11847: /* andc=vector(1,n); */
11848: /* weight=vector(1,n); */
11849: /* agedc=vector(1,n); */
11850: /* cod=ivector(1,n); */
11851: /* for(i=1;i<=n;i++){ */
11852: num=lvector(firstobs,lastobs);
11853: moisnais=vector(firstobs,lastobs);
11854: annais=vector(firstobs,lastobs);
11855: moisdc=vector(firstobs,lastobs);
11856: andc=vector(firstobs,lastobs);
11857: weight=vector(firstobs,lastobs);
11858: agedc=vector(firstobs,lastobs);
11859: cod=ivector(firstobs,lastobs);
11860: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11861: num[i]=0;
11862: moisnais[i]=0;
11863: annais[i]=0;
11864: moisdc[i]=0;
11865: andc[i]=0;
11866: agedc[i]=0;
11867: cod[i]=0;
11868: weight[i]=1.0; /* Equal weights, 1 by default */
11869: }
1.290 brouard 11870: mint=matrix(1,maxwav,firstobs,lastobs);
11871: anint=matrix(1,maxwav,firstobs,lastobs);
11872: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11873: tab=ivector(1,NCOVMAX);
1.144 brouard 11874: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11875: 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 11876:
1.136 brouard 11877: /* Reads data from file datafile */
11878: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11879: goto end;
11880:
11881: /* Calculation of the number of parameters from char model */
1.234 brouard 11882: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11883: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11884: k=3 V4 Tvar[k=3]= 4 (from V4)
11885: k=2 V1 Tvar[k=2]= 1 (from V1)
11886: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11887: */
11888:
11889: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11890: TvarsDind=ivector(1,NCOVMAX); /* */
11891: TvarsD=ivector(1,NCOVMAX); /* */
11892: TvarsQind=ivector(1,NCOVMAX); /* */
11893: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11894: TvarF=ivector(1,NCOVMAX); /* */
11895: TvarFind=ivector(1,NCOVMAX); /* */
11896: TvarV=ivector(1,NCOVMAX); /* */
11897: TvarVind=ivector(1,NCOVMAX); /* */
11898: TvarA=ivector(1,NCOVMAX); /* */
11899: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11900: TvarFD=ivector(1,NCOVMAX); /* */
11901: TvarFDind=ivector(1,NCOVMAX); /* */
11902: TvarFQ=ivector(1,NCOVMAX); /* */
11903: TvarFQind=ivector(1,NCOVMAX); /* */
11904: TvarVD=ivector(1,NCOVMAX); /* */
11905: TvarVDind=ivector(1,NCOVMAX); /* */
11906: TvarVQ=ivector(1,NCOVMAX); /* */
11907: TvarVQind=ivector(1,NCOVMAX); /* */
11908:
1.230 brouard 11909: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11910: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11911: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11912: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11913: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11914: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11915: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11916: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11917: */
11918: /* For model-covariate k tells which data-covariate to use but
11919: because this model-covariate is a construction we invent a new column
11920: ncovcol + k1
11921: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11922: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11923: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11924: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11925: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11926: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11927: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11928: */
1.145 brouard 11929: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11930: 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 11931: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11932: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11933: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11934: 4 covariates (3 plus signs)
11935: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11936: */
1.230 brouard 11937: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11938: * individual dummy, fixed or varying:
11939: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11940: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11941: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11942: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11943: * Tmodelind[1]@9={9,0,3,2,}*/
11944: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11945: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11946: * individual quantitative, fixed or varying:
11947: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11948: * 3, 1, 0, 0, 0, 0, 0, 0},
11949: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11950: /* Main decodemodel */
11951:
1.187 brouard 11952:
1.223 brouard 11953: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11954: goto end;
11955:
1.137 brouard 11956: if((double)(lastobs-imx)/(double)imx > 1.10){
11957: nbwarn++;
11958: 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);
11959: 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);
11960: }
1.136 brouard 11961: /* if(mle==1){*/
1.137 brouard 11962: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11963: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11964: }
11965:
11966: /*-calculation of age at interview from date of interview and age at death -*/
11967: agev=matrix(1,maxwav,1,imx);
11968:
11969: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11970: goto end;
11971:
1.126 brouard 11972:
1.136 brouard 11973: agegomp=(int)agemin;
1.290 brouard 11974: free_vector(moisnais,firstobs,lastobs);
11975: free_vector(annais,firstobs,lastobs);
1.126 brouard 11976: /* free_matrix(mint,1,maxwav,1,n);
11977: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11978: /* free_vector(moisdc,1,n); */
11979: /* free_vector(andc,1,n); */
1.145 brouard 11980: /* */
11981:
1.126 brouard 11982: wav=ivector(1,imx);
1.214 brouard 11983: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11984: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11985: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11986: 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.*/
11987: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11988: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11989:
11990: /* Concatenates waves */
1.214 brouard 11991: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11992: Death is a valid wave (if date is known).
11993: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11994: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11995: and mw[mi+1][i]. dh depends on stepm.
11996: */
11997:
1.126 brouard 11998: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11999: /* Concatenates waves */
1.145 brouard 12000:
1.290 brouard 12001: free_vector(moisdc,firstobs,lastobs);
12002: free_vector(andc,firstobs,lastobs);
1.215 brouard 12003:
1.126 brouard 12004: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
12005: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
12006: ncodemax[1]=1;
1.145 brouard 12007: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 12008: cptcoveff=0;
1.220 brouard 12009: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
12010: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 12011: }
12012:
12013: ncovcombmax=pow(2,cptcoveff);
12014: invalidvarcomb=ivector(1, ncovcombmax);
12015: for(i=1;i<ncovcombmax;i++)
12016: invalidvarcomb[i]=0;
12017:
1.211 brouard 12018: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 12019: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 12020: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 12021:
1.200 brouard 12022: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 12023: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 12024: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 12025: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
12026: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
12027: * (currently 0 or 1) in the data.
12028: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
12029: * corresponding modality (h,j).
12030: */
12031:
1.145 brouard 12032: h=0;
12033: /*if (cptcovn > 0) */
1.126 brouard 12034: m=pow(2,cptcoveff);
12035:
1.144 brouard 12036: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 12037: * For k=4 covariates, h goes from 1 to m=2**k
12038: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
12039: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 12040: * h\k 1 2 3 4
1.143 brouard 12041: *______________________________
12042: * 1 i=1 1 i=1 1 i=1 1 i=1 1
12043: * 2 2 1 1 1
12044: * 3 i=2 1 2 1 1
12045: * 4 2 2 1 1
12046: * 5 i=3 1 i=2 1 2 1
12047: * 6 2 1 2 1
12048: * 7 i=4 1 2 2 1
12049: * 8 2 2 2 1
1.197 brouard 12050: * 9 i=5 1 i=3 1 i=2 1 2
12051: * 10 2 1 1 2
12052: * 11 i=6 1 2 1 2
12053: * 12 2 2 1 2
12054: * 13 i=7 1 i=4 1 2 2
12055: * 14 2 1 2 2
12056: * 15 i=8 1 2 2 2
12057: * 16 2 2 2 2
1.143 brouard 12058: */
1.212 brouard 12059: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 12060: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
12061: * and the value of each covariate?
12062: * V1=1, V2=1, V3=2, V4=1 ?
12063: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
12064: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
12065: * In order to get the real value in the data, we use nbcode
12066: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
12067: * We are keeping this crazy system in order to be able (in the future?)
12068: * to have more than 2 values (0 or 1) for a covariate.
12069: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
12070: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
12071: * bbbbbbbb
12072: * 76543210
12073: * h-1 00000101 (6-1=5)
1.219 brouard 12074: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 12075: * &
12076: * 1 00000001 (1)
1.219 brouard 12077: * 00000000 = 1 & ((h-1) >> (k-1))
12078: * +1= 00000001 =1
1.211 brouard 12079: *
12080: * h=14, k=3 => h'=h-1=13, k'=k-1=2
12081: * h' 1101 =2^3+2^2+0x2^1+2^0
12082: * >>k' 11
12083: * & 00000001
12084: * = 00000001
12085: * +1 = 00000010=2 = codtabm(14,3)
12086: * Reverse h=6 and m=16?
12087: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
12088: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
12089: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
12090: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
12091: * V3=decodtabm(14,3,2**4)=2
12092: * h'=13 1101 =2^3+2^2+0x2^1+2^0
12093: *(h-1) >> (j-1) 0011 =13 >> 2
12094: * &1 000000001
12095: * = 000000001
12096: * +1= 000000010 =2
12097: * 2211
12098: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
12099: * V3=2
1.220 brouard 12100: * codtabm and decodtabm are identical
1.211 brouard 12101: */
12102:
1.145 brouard 12103:
12104: free_ivector(Ndum,-1,NCOVMAX);
12105:
12106:
1.126 brouard 12107:
1.186 brouard 12108: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 12109: strcpy(optionfilegnuplot,optionfilefiname);
12110: if(mle==-3)
1.201 brouard 12111: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 12112: strcat(optionfilegnuplot,".gp");
12113:
12114: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
12115: printf("Problem with file %s",optionfilegnuplot);
12116: }
12117: else{
1.204 brouard 12118: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 12119: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 12120: //fprintf(ficgp,"set missing 'NaNq'\n");
12121: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 12122: }
12123: /* fclose(ficgp);*/
1.186 brouard 12124:
12125:
12126: /* Initialisation of --------- index.htm --------*/
1.126 brouard 12127:
12128: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
12129: if(mle==-3)
1.201 brouard 12130: strcat(optionfilehtm,"-MORT_");
1.126 brouard 12131: strcat(optionfilehtm,".htm");
12132: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 12133: printf("Problem with %s \n",optionfilehtm);
12134: exit(0);
1.126 brouard 12135: }
12136:
12137: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
12138: strcat(optionfilehtmcov,"-cov.htm");
12139: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
12140: printf("Problem with %s \n",optionfilehtmcov), exit(0);
12141: }
12142: else{
12143: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
12144: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12145: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 12146: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
12147: }
12148:
1.213 brouard 12149: fprintf(fichtm,"<html><head>\n<head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<title>IMaCh %s</title></head>\n <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n<font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>-EUROREVES-Institut de longévité-2013-2016-Japan Society for the Promotion of Sciences 日本å¦è¡“振興会 (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \
1.204 brouard 12150: <hr size=\"2\" color=\"#EC5E5E\"> \n\
12151: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 12152: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12153: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 12154: \n\
12155: <hr size=\"2\" color=\"#EC5E5E\">\
12156: <ul><li><h4>Parameter files</h4>\n\
12157: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
12158: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
12159: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
12160: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
12161: - Date and time at start: %s</ul>\n",\
12162: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
12163: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
12164: fileres,fileres,\
12165: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
12166: fflush(fichtm);
12167:
12168: strcpy(pathr,path);
12169: strcat(pathr,optionfilefiname);
1.184 brouard 12170: #ifdef WIN32
12171: _chdir(optionfilefiname); /* Move to directory named optionfile */
12172: #else
1.126 brouard 12173: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 12174: #endif
12175:
1.126 brouard 12176:
1.220 brouard 12177: /* Calculates basic frequencies. Computes observed prevalence at single age
12178: and for any valid combination of covariates
1.126 brouard 12179: and prints on file fileres'p'. */
1.251 brouard 12180: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 12181: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 12182:
12183: fprintf(fichtm,"\n");
1.286 brouard 12184: 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 12185: ftol, stepm);
12186: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
12187: ncurrv=1;
12188: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
12189: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
12190: ncurrv=i;
12191: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12192: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 12193: ncurrv=i;
12194: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12195: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 12196: ncurrv=i;
12197: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
12198: 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", \
12199: nlstate, ndeath, maxwav, mle, weightopt);
12200:
12201: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
12202: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
12203:
12204:
1.317 brouard 12205: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 12206: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
12207: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 12208: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 12209: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 12210: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12211: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12212: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12213: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 12214:
1.126 brouard 12215: /* For Powell, parameters are in a vector p[] starting at p[1]
12216: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
12217: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
12218:
12219: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 12220: /* For mortality only */
1.126 brouard 12221: if (mle==-3){
1.136 brouard 12222: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 12223: for(i=1;i<=NDIM;i++)
12224: for(j=1;j<=NDIM;j++)
12225: ximort[i][j]=0.;
1.186 brouard 12226: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 12227: cens=ivector(firstobs,lastobs);
12228: ageexmed=vector(firstobs,lastobs);
12229: agecens=vector(firstobs,lastobs);
12230: dcwave=ivector(firstobs,lastobs);
1.223 brouard 12231:
1.126 brouard 12232: for (i=1; i<=imx; i++){
12233: dcwave[i]=-1;
12234: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 12235: if (s[m][i]>nlstate) {
12236: dcwave[i]=m;
12237: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
12238: break;
12239: }
1.126 brouard 12240: }
1.226 brouard 12241:
1.126 brouard 12242: for (i=1; i<=imx; i++) {
12243: if (wav[i]>0){
1.226 brouard 12244: ageexmed[i]=agev[mw[1][i]][i];
12245: j=wav[i];
12246: agecens[i]=1.;
12247:
12248: if (ageexmed[i]> 1 && wav[i] > 0){
12249: agecens[i]=agev[mw[j][i]][i];
12250: cens[i]= 1;
12251: }else if (ageexmed[i]< 1)
12252: cens[i]= -1;
12253: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
12254: cens[i]=0 ;
1.126 brouard 12255: }
12256: else cens[i]=-1;
12257: }
12258:
12259: for (i=1;i<=NDIM;i++) {
12260: for (j=1;j<=NDIM;j++)
1.226 brouard 12261: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 12262: }
12263:
1.302 brouard 12264: p[1]=0.0268; p[NDIM]=0.083;
12265: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12266:
12267:
1.136 brouard 12268: #ifdef GSL
12269: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12270: #else
1.126 brouard 12271: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12272: #endif
1.201 brouard 12273: strcpy(filerespow,"POW-MORT_");
12274: strcat(filerespow,fileresu);
1.126 brouard 12275: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12276: printf("Problem with resultfile: %s\n", filerespow);
12277: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12278: }
1.136 brouard 12279: #ifdef GSL
12280: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12281: #else
1.126 brouard 12282: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12283: #endif
1.126 brouard 12284: /* for (i=1;i<=nlstate;i++)
12285: for(j=1;j<=nlstate+ndeath;j++)
12286: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12287: */
12288: fprintf(ficrespow,"\n");
1.136 brouard 12289: #ifdef GSL
12290: /* gsl starts here */
12291: T = gsl_multimin_fminimizer_nmsimplex;
12292: gsl_multimin_fminimizer *sfm = NULL;
12293: gsl_vector *ss, *x;
12294: gsl_multimin_function minex_func;
12295:
12296: /* Initial vertex size vector */
12297: ss = gsl_vector_alloc (NDIM);
12298:
12299: if (ss == NULL){
12300: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
12301: }
12302: /* Set all step sizes to 1 */
12303: gsl_vector_set_all (ss, 0.001);
12304:
12305: /* Starting point */
1.126 brouard 12306:
1.136 brouard 12307: x = gsl_vector_alloc (NDIM);
12308:
12309: if (x == NULL){
12310: gsl_vector_free(ss);
12311: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
12312: }
12313:
12314: /* Initialize method and iterate */
12315: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 12316: /* gsl_vector_set(x, 0, 0.0268); */
12317: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 12318: gsl_vector_set(x, 0, p[1]);
12319: gsl_vector_set(x, 1, p[2]);
12320:
12321: minex_func.f = &gompertz_f;
12322: minex_func.n = NDIM;
12323: minex_func.params = (void *)&p; /* ??? */
12324:
12325: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
12326: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
12327:
12328: printf("Iterations beginning .....\n\n");
12329: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
12330:
12331: iteri=0;
12332: while (rval == GSL_CONTINUE){
12333: iteri++;
12334: status = gsl_multimin_fminimizer_iterate(sfm);
12335:
12336: if (status) printf("error: %s\n", gsl_strerror (status));
12337: fflush(0);
12338:
12339: if (status)
12340: break;
12341:
12342: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12343: ssval = gsl_multimin_fminimizer_size (sfm);
12344:
12345: if (rval == GSL_SUCCESS)
12346: printf ("converged to a local maximum at\n");
12347:
12348: printf("%5d ", iteri);
12349: for (it = 0; it < NDIM; it++){
12350: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12351: }
12352: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12353: }
12354:
12355: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12356:
12357: gsl_vector_free(x); /* initial values */
12358: gsl_vector_free(ss); /* inital step size */
12359: for (it=0; it<NDIM; it++){
12360: p[it+1]=gsl_vector_get(sfm->x,it);
12361: fprintf(ficrespow," %.12lf", p[it]);
12362: }
12363: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12364: #endif
12365: #ifdef POWELL
12366: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12367: #endif
1.126 brouard 12368: fclose(ficrespow);
12369:
1.203 brouard 12370: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12371:
12372: for(i=1; i <=NDIM; i++)
12373: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12374: matcov[i][j]=matcov[j][i];
1.126 brouard 12375:
12376: printf("\nCovariance matrix\n ");
1.203 brouard 12377: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12378: for(i=1; i <=NDIM; i++) {
12379: for(j=1;j<=NDIM;j++){
1.220 brouard 12380: printf("%f ",matcov[i][j]);
12381: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12382: }
1.203 brouard 12383: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12384: }
12385:
12386: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12387: for (i=1;i<=NDIM;i++) {
1.126 brouard 12388: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12389: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12390: }
1.302 brouard 12391: lsurv=vector(agegomp,AGESUP);
12392: lpop=vector(agegomp,AGESUP);
12393: tpop=vector(agegomp,AGESUP);
1.126 brouard 12394: lsurv[agegomp]=100000;
12395:
12396: for (k=agegomp;k<=AGESUP;k++) {
12397: agemortsup=k;
12398: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12399: }
12400:
12401: for (k=agegomp;k<agemortsup;k++)
12402: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12403:
12404: for (k=agegomp;k<agemortsup;k++){
12405: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12406: sumlpop=sumlpop+lpop[k];
12407: }
12408:
12409: tpop[agegomp]=sumlpop;
12410: for (k=agegomp;k<(agemortsup-3);k++){
12411: /* tpop[k+1]=2;*/
12412: tpop[k+1]=tpop[k]-lpop[k];
12413: }
12414:
12415:
12416: printf("\nAge lx qx dx Lx Tx e(x)\n");
12417: for (k=agegomp;k<(agemortsup-2);k++)
12418: 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]);
12419:
12420:
12421: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12422: ageminpar=50;
12423: agemaxpar=100;
1.194 brouard 12424: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12425: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12426: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12427: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12428: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12429: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12430: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12431: }else{
12432: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12433: 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 12434: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12435: }
1.201 brouard 12436: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12437: stepm, weightopt,\
12438: model,imx,p,matcov,agemortsup);
12439:
1.302 brouard 12440: free_vector(lsurv,agegomp,AGESUP);
12441: free_vector(lpop,agegomp,AGESUP);
12442: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 12443: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12444: free_ivector(dcwave,firstobs,lastobs);
12445: free_vector(agecens,firstobs,lastobs);
12446: free_vector(ageexmed,firstobs,lastobs);
12447: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12448: #ifdef GSL
1.136 brouard 12449: #endif
1.186 brouard 12450: } /* Endof if mle==-3 mortality only */
1.205 brouard 12451: /* Standard */
12452: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12453: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12454: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12455: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12456: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12457: for (k=1; k<=npar;k++)
12458: printf(" %d %8.5f",k,p[k]);
12459: printf("\n");
1.205 brouard 12460: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12461: /* mlikeli uses func not funcone */
1.247 brouard 12462: /* for(i=1;i<nlstate;i++){ */
12463: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12464: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12465: /* } */
1.205 brouard 12466: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12467: }
12468: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12469: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12470: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12471: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12472: }
12473: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12474: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12475: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12476: for (k=1; k<=npar;k++)
12477: printf(" %d %8.5f",k,p[k]);
12478: printf("\n");
12479:
12480: /*--------- results files --------------*/
1.283 brouard 12481: /* 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 12482:
12483:
12484: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12485: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 12486: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12487:
12488: printf("#model= 1 + age ");
12489: fprintf(ficres,"#model= 1 + age ");
12490: fprintf(ficlog,"#model= 1 + age ");
12491: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
12492: </ul>", model);
12493:
12494: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
12495: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
12496: if(nagesqr==1){
12497: printf(" + age*age ");
12498: fprintf(ficres," + age*age ");
12499: fprintf(ficlog," + age*age ");
12500: fprintf(fichtm, "<th>+ age*age</th>");
12501: }
12502: for(j=1;j <=ncovmodel-2;j++){
12503: if(Typevar[j]==0) {
12504: printf(" + V%d ",Tvar[j]);
12505: fprintf(ficres," + V%d ",Tvar[j]);
12506: fprintf(ficlog," + V%d ",Tvar[j]);
12507: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
12508: }else if(Typevar[j]==1) {
12509: printf(" + V%d*age ",Tvar[j]);
12510: fprintf(ficres," + V%d*age ",Tvar[j]);
12511: fprintf(ficlog," + V%d*age ",Tvar[j]);
12512: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
12513: }else if(Typevar[j]==2) {
12514: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12515: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12516: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12517: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12518: }
12519: }
12520: printf("\n");
12521: fprintf(ficres,"\n");
12522: fprintf(ficlog,"\n");
12523: fprintf(fichtm, "</tr>");
12524: fprintf(fichtm, "\n");
12525:
12526:
1.126 brouard 12527: for(i=1,jk=1; i <=nlstate; i++){
12528: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12529: if (k != i) {
1.319 brouard 12530: fprintf(fichtm, "<tr>");
1.225 brouard 12531: printf("%d%d ",i,k);
12532: fprintf(ficlog,"%d%d ",i,k);
12533: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 12534: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 12535: for(j=1; j <=ncovmodel; j++){
12536: printf("%12.7f ",p[jk]);
12537: fprintf(ficlog,"%12.7f ",p[jk]);
12538: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 12539: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 12540: jk++;
12541: }
12542: printf("\n");
12543: fprintf(ficlog,"\n");
12544: fprintf(ficres,"\n");
1.319 brouard 12545: fprintf(fichtm, "</tr>\n");
1.225 brouard 12546: }
1.126 brouard 12547: }
12548: }
1.319 brouard 12549: /* fprintf(fichtm,"</tr>\n"); */
12550: fprintf(fichtm,"</table>\n");
12551: fprintf(fichtm, "\n");
12552:
1.203 brouard 12553: if(mle != 0){
12554: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12555: ftolhess=ftol; /* Usually correct */
1.203 brouard 12556: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12557: 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");
12558: 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 12559: 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 12560: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
12561: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
12562: if(nagesqr==1){
12563: printf(" + age*age ");
12564: fprintf(ficres," + age*age ");
12565: fprintf(ficlog," + age*age ");
12566: fprintf(fichtm, "<th>+ age*age</th>");
12567: }
12568: for(j=1;j <=ncovmodel-2;j++){
12569: if(Typevar[j]==0) {
12570: printf(" + V%d ",Tvar[j]);
12571: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
12572: }else if(Typevar[j]==1) {
12573: printf(" + V%d*age ",Tvar[j]);
12574: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
12575: }else if(Typevar[j]==2) {
12576: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12577: }
12578: }
12579: fprintf(fichtm, "</tr>\n");
12580:
1.203 brouard 12581: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12582: for(k=1; k <=(nlstate+ndeath); k++){
12583: if (k != i) {
1.319 brouard 12584: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 12585: printf("%d%d ",i,k);
12586: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 12587: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 12588: for(j=1; j <=ncovmodel; j++){
1.319 brouard 12589: wald=p[jk]/sqrt(matcov[jk][jk]);
1.321 brouard 12590: printf("%12.7f(%12.7f) sqrt(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]));
12591: fprintf(ficlog,"%12.7f(%12.7f) sqrt(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 12592: if(fabs(wald) > 1.96){
1.321 brouard 12593: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 12594: }else{
12595: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
12596: }
1.321 brouard 12597: fprintf(fichtm,"sqrt(W)=%8.3f</br>",wald);
1.319 brouard 12598: 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 12599: jk++;
12600: }
12601: printf("\n");
12602: fprintf(ficlog,"\n");
1.319 brouard 12603: fprintf(fichtm, "</tr>\n");
1.225 brouard 12604: }
12605: }
1.193 brouard 12606: }
1.203 brouard 12607: } /* end of hesscov and Wald tests */
1.319 brouard 12608: fprintf(fichtm,"</table>\n");
1.225 brouard 12609:
1.203 brouard 12610: /* */
1.126 brouard 12611: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12612: printf("# Scales (for hessian or gradient estimation)\n");
12613: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12614: for(i=1,jk=1; i <=nlstate; i++){
12615: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12616: if (j!=i) {
12617: fprintf(ficres,"%1d%1d",i,j);
12618: printf("%1d%1d",i,j);
12619: fprintf(ficlog,"%1d%1d",i,j);
12620: for(k=1; k<=ncovmodel;k++){
12621: printf(" %.5e",delti[jk]);
12622: fprintf(ficlog," %.5e",delti[jk]);
12623: fprintf(ficres," %.5e",delti[jk]);
12624: jk++;
12625: }
12626: printf("\n");
12627: fprintf(ficlog,"\n");
12628: fprintf(ficres,"\n");
12629: }
1.126 brouard 12630: }
12631: }
12632:
12633: 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 12634: if(mle >= 1) /* To big for the screen */
1.126 brouard 12635: 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");
12636: 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");
12637: /* # 121 Var(a12)\n\ */
12638: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12639: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12640: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12641: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12642: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12643: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12644: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12645:
12646:
12647: /* Just to have a covariance matrix which will be more understandable
12648: even is we still don't want to manage dictionary of variables
12649: */
12650: for(itimes=1;itimes<=2;itimes++){
12651: jj=0;
12652: for(i=1; i <=nlstate; i++){
1.225 brouard 12653: for(j=1; j <=nlstate+ndeath; j++){
12654: if(j==i) continue;
12655: for(k=1; k<=ncovmodel;k++){
12656: jj++;
12657: ca[0]= k+'a'-1;ca[1]='\0';
12658: if(itimes==1){
12659: if(mle>=1)
12660: printf("#%1d%1d%d",i,j,k);
12661: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12662: fprintf(ficres,"#%1d%1d%d",i,j,k);
12663: }else{
12664: if(mle>=1)
12665: printf("%1d%1d%d",i,j,k);
12666: fprintf(ficlog,"%1d%1d%d",i,j,k);
12667: fprintf(ficres,"%1d%1d%d",i,j,k);
12668: }
12669: ll=0;
12670: for(li=1;li <=nlstate; li++){
12671: for(lj=1;lj <=nlstate+ndeath; lj++){
12672: if(lj==li) continue;
12673: for(lk=1;lk<=ncovmodel;lk++){
12674: ll++;
12675: if(ll<=jj){
12676: cb[0]= lk +'a'-1;cb[1]='\0';
12677: if(ll<jj){
12678: if(itimes==1){
12679: if(mle>=1)
12680: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12681: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12682: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12683: }else{
12684: if(mle>=1)
12685: printf(" %.5e",matcov[jj][ll]);
12686: fprintf(ficlog," %.5e",matcov[jj][ll]);
12687: fprintf(ficres," %.5e",matcov[jj][ll]);
12688: }
12689: }else{
12690: if(itimes==1){
12691: if(mle>=1)
12692: printf(" Var(%s%1d%1d)",ca,i,j);
12693: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12694: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12695: }else{
12696: if(mle>=1)
12697: printf(" %.7e",matcov[jj][ll]);
12698: fprintf(ficlog," %.7e",matcov[jj][ll]);
12699: fprintf(ficres," %.7e",matcov[jj][ll]);
12700: }
12701: }
12702: }
12703: } /* end lk */
12704: } /* end lj */
12705: } /* end li */
12706: if(mle>=1)
12707: printf("\n");
12708: fprintf(ficlog,"\n");
12709: fprintf(ficres,"\n");
12710: numlinepar++;
12711: } /* end k*/
12712: } /*end j */
1.126 brouard 12713: } /* end i */
12714: } /* end itimes */
12715:
12716: fflush(ficlog);
12717: fflush(ficres);
1.225 brouard 12718: while(fgets(line, MAXLINE, ficpar)) {
12719: /* If line starts with a # it is a comment */
12720: if (line[0] == '#') {
12721: numlinepar++;
12722: fputs(line,stdout);
12723: fputs(line,ficparo);
12724: fputs(line,ficlog);
1.299 brouard 12725: fputs(line,ficres);
1.225 brouard 12726: continue;
12727: }else
12728: break;
12729: }
12730:
1.209 brouard 12731: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12732: /* ungetc(c,ficpar); */
12733: /* fgets(line, MAXLINE, ficpar); */
12734: /* fputs(line,stdout); */
12735: /* fputs(line,ficparo); */
12736: /* } */
12737: /* ungetc(c,ficpar); */
1.126 brouard 12738:
12739: estepm=0;
1.209 brouard 12740: 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 12741:
12742: if (num_filled != 6) {
12743: 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);
12744: 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);
12745: goto end;
12746: }
12747: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12748: }
12749: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12750: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12751:
1.209 brouard 12752: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12753: if (estepm==0 || estepm < stepm) estepm=stepm;
12754: if (fage <= 2) {
12755: bage = ageminpar;
12756: fage = agemaxpar;
12757: }
12758:
12759: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12760: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12761: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12762:
1.186 brouard 12763: /* Other stuffs, more or less useful */
1.254 brouard 12764: while(fgets(line, MAXLINE, ficpar)) {
12765: /* If line starts with a # it is a comment */
12766: if (line[0] == '#') {
12767: numlinepar++;
12768: fputs(line,stdout);
12769: fputs(line,ficparo);
12770: fputs(line,ficlog);
1.299 brouard 12771: fputs(line,ficres);
1.254 brouard 12772: continue;
12773: }else
12774: break;
12775: }
12776:
12777: 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){
12778:
12779: if (num_filled != 7) {
12780: 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);
12781: 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);
12782: goto end;
12783: }
12784: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12785: 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);
12786: 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);
12787: 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 12788: }
1.254 brouard 12789:
12790: while(fgets(line, MAXLINE, ficpar)) {
12791: /* If line starts with a # it is a comment */
12792: if (line[0] == '#') {
12793: numlinepar++;
12794: fputs(line,stdout);
12795: fputs(line,ficparo);
12796: fputs(line,ficlog);
1.299 brouard 12797: fputs(line,ficres);
1.254 brouard 12798: continue;
12799: }else
12800: break;
1.126 brouard 12801: }
12802:
12803:
12804: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12805: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12806:
1.254 brouard 12807: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12808: if (num_filled != 1) {
12809: 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);
12810: 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);
12811: goto end;
12812: }
12813: printf("pop_based=%d\n",popbased);
12814: fprintf(ficlog,"pop_based=%d\n",popbased);
12815: fprintf(ficparo,"pop_based=%d\n",popbased);
12816: fprintf(ficres,"pop_based=%d\n",popbased);
12817: }
12818:
1.258 brouard 12819: /* Results */
1.307 brouard 12820: endishere=0;
1.258 brouard 12821: nresult=0;
1.308 brouard 12822: parameterline=0;
1.258 brouard 12823: do{
12824: if(!fgets(line, MAXLINE, ficpar)){
12825: endishere=1;
1.308 brouard 12826: parameterline=15;
1.258 brouard 12827: }else if (line[0] == '#') {
12828: /* If line starts with a # it is a comment */
1.254 brouard 12829: numlinepar++;
12830: fputs(line,stdout);
12831: fputs(line,ficparo);
12832: fputs(line,ficlog);
1.299 brouard 12833: fputs(line,ficres);
1.254 brouard 12834: continue;
1.258 brouard 12835: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12836: parameterline=11;
1.296 brouard 12837: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12838: parameterline=12;
1.307 brouard 12839: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 12840: parameterline=13;
1.307 brouard 12841: }
1.258 brouard 12842: else{
12843: parameterline=14;
1.254 brouard 12844: }
1.308 brouard 12845: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 12846: case 11:
1.296 brouard 12847: 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)){
12848: 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 12849: 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);
12850: 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);
12851: 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);
12852: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12853: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12854: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12855: prvforecast = 1;
12856: }
12857: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 12858: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12859: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12860: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 12861: prvforecast = 2;
12862: }
12863: else {
12864: 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);
12865: 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);
12866: goto end;
1.258 brouard 12867: }
1.254 brouard 12868: break;
1.258 brouard 12869: case 12:
1.296 brouard 12870: 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)){
12871: 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);
12872: 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);
12873: 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);
12874: 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);
12875: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12876: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12877: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12878: prvbackcast = 1;
12879: }
12880: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 12881: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12882: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12883: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 12884: prvbackcast = 2;
12885: }
12886: else {
12887: 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);
12888: 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);
12889: goto end;
1.258 brouard 12890: }
1.230 brouard 12891: break;
1.258 brouard 12892: case 13:
1.307 brouard 12893: num_filled=sscanf(line,"result:%[^\n]\n",resultline);
12894: nresult++; /* Sum of resultlines */
12895: printf("Result %d: result:%s\n",nresult, resultline);
1.318 brouard 12896: if(nresult > MAXRESULTLINESPONE-1){
12897: 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);
12898: 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 12899: goto end;
12900: }
1.310 brouard 12901: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 12902: fprintf(ficparo,"result: %s\n",resultline);
12903: fprintf(ficres,"result: %s\n",resultline);
12904: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 12905: } else
12906: goto end;
1.307 brouard 12907: break;
12908: case 14:
12909: printf("Error: Unknown command '%s'\n",line);
12910: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 12911: if(line[0] == ' ' || line[0] == '\n'){
12912: printf("It should not be an empty line '%s'\n",line);
12913: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
12914: }
1.307 brouard 12915: if(ncovmodel >=2 && nresult==0 ){
12916: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
12917: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12918: }
1.307 brouard 12919: /* goto end; */
12920: break;
1.308 brouard 12921: case 15:
12922: printf("End of resultlines.\n");
12923: fprintf(ficlog,"End of resultlines.\n");
12924: break;
12925: default: /* parameterline =0 */
1.307 brouard 12926: nresult=1;
12927: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 12928: } /* End switch parameterline */
12929: }while(endishere==0); /* End do */
1.126 brouard 12930:
1.230 brouard 12931: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12932: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12933:
12934: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12935: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12936: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12937: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12938: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12939: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12940: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12941: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12942: }else{
1.270 brouard 12943: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 12944: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
12945: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
12946: if(prvforecast==1){
12947: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
12948: jprojd=jproj1;
12949: mprojd=mproj1;
12950: anprojd=anproj1;
12951: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
12952: jprojf=jproj2;
12953: mprojf=mproj2;
12954: anprojf=anproj2;
12955: } else if(prvforecast == 2){
12956: dateprojd=dateintmean;
12957: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
12958: dateprojf=dateintmean+yrfproj;
12959: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
12960: }
12961: if(prvbackcast==1){
12962: datebackd=(jback1+12*mback1+365*anback1)/365;
12963: jbackd=jback1;
12964: mbackd=mback1;
12965: anbackd=anback1;
12966: datebackf=(jback2+12*mback2+365*anback2)/365;
12967: jbackf=jback2;
12968: mbackf=mback2;
12969: anbackf=anback2;
12970: } else if(prvbackcast == 2){
12971: datebackd=dateintmean;
12972: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
12973: datebackf=dateintmean-yrbproj;
12974: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
12975: }
12976:
12977: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 12978: }
12979: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 12980: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
12981: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 12982:
1.225 brouard 12983: /*------------ free_vector -------------*/
12984: /* chdir(path); */
1.220 brouard 12985:
1.215 brouard 12986: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12987: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12988: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12989: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 12990: free_lvector(num,firstobs,lastobs);
12991: free_vector(agedc,firstobs,lastobs);
1.126 brouard 12992: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12993: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12994: fclose(ficparo);
12995: fclose(ficres);
1.220 brouard 12996:
12997:
1.186 brouard 12998: /* Other results (useful)*/
1.220 brouard 12999:
13000:
1.126 brouard 13001: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 13002: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
13003: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 13004: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 13005: fclose(ficrespl);
13006:
13007: /*------------- h Pij x at various ages ------------*/
1.180 brouard 13008: /*#include "hpijx.h"*/
13009: hPijx(p, bage, fage);
1.145 brouard 13010: fclose(ficrespij);
1.227 brouard 13011:
1.220 brouard 13012: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 13013: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 13014: k=1;
1.126 brouard 13015: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 13016:
1.269 brouard 13017: /* Prevalence for each covariate combination in probs[age][status][cov] */
13018: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13019: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 13020: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 13021: for(k=1;k<=ncovcombmax;k++)
13022: probs[i][j][k]=0.;
1.269 brouard 13023: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
13024: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 13025: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 13026: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13027: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 13028: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 13029: for(k=1;k<=ncovcombmax;k++)
13030: mobaverages[i][j][k]=0.;
1.219 brouard 13031: mobaverage=mobaverages;
13032: if (mobilav!=0) {
1.235 brouard 13033: printf("Movingaveraging observed prevalence\n");
1.258 brouard 13034: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 13035: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
13036: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
13037: printf(" Error in movingaverage mobilav=%d\n",mobilav);
13038: }
1.269 brouard 13039: } else if (mobilavproj !=0) {
1.235 brouard 13040: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 13041: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 13042: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
13043: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
13044: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
13045: }
1.269 brouard 13046: }else{
13047: printf("Internal error moving average\n");
13048: fflush(stdout);
13049: exit(1);
1.219 brouard 13050: }
13051: }/* end if moving average */
1.227 brouard 13052:
1.126 brouard 13053: /*---------- Forecasting ------------------*/
1.296 brouard 13054: if(prevfcast==1){
13055: /* /\* if(stepm ==1){*\/ */
13056: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13057: /*This done previously after freqsummary.*/
13058: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
13059: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
13060:
13061: /* } else if (prvforecast==2){ */
13062: /* /\* if(stepm ==1){*\/ */
13063: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13064: /* } */
13065: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
13066: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 13067: }
1.269 brouard 13068:
1.296 brouard 13069: /* Prevbcasting */
13070: if(prevbcast==1){
1.219 brouard 13071: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13072: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13073: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13074:
13075: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
13076:
13077: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 13078:
1.219 brouard 13079: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
13080: fclose(ficresplb);
13081:
1.222 brouard 13082: hBijx(p, bage, fage, mobaverage);
13083: fclose(ficrespijb);
1.219 brouard 13084:
1.296 brouard 13085: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
13086: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
13087: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
13088: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
13089: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
13090: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
13091:
13092:
1.269 brouard 13093: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13094:
13095:
1.269 brouard 13096: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 13097: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13098: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13099: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 13100: } /* end Prevbcasting */
1.268 brouard 13101:
1.186 brouard 13102:
13103: /* ------ Other prevalence ratios------------ */
1.126 brouard 13104:
1.215 brouard 13105: free_ivector(wav,1,imx);
13106: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
13107: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
13108: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 13109:
13110:
1.127 brouard 13111: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 13112:
1.201 brouard 13113: strcpy(filerese,"E_");
13114: strcat(filerese,fileresu);
1.126 brouard 13115: if((ficreseij=fopen(filerese,"w"))==NULL) {
13116: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13117: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13118: }
1.208 brouard 13119: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
13120: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 13121:
13122: pstamp(ficreseij);
1.219 brouard 13123:
1.235 brouard 13124: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13125: if (cptcovn < 1){i1=1;}
13126:
13127: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13128: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13129: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13130: continue;
1.219 brouard 13131: fprintf(ficreseij,"\n#****** ");
1.235 brouard 13132: printf("\n#****** ");
1.225 brouard 13133: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 13134: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 13135: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13136: }
13137: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13138: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13139: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 13140: }
13141: fprintf(ficreseij,"******\n");
1.235 brouard 13142: printf("******\n");
1.219 brouard 13143:
13144: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13145: oldm=oldms;savm=savms;
1.235 brouard 13146: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 13147:
1.219 brouard 13148: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 13149: }
13150: fclose(ficreseij);
1.208 brouard 13151: printf("done evsij\n");fflush(stdout);
13152: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 13153:
1.218 brouard 13154:
1.227 brouard 13155: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 13156:
1.201 brouard 13157: strcpy(filerest,"T_");
13158: strcat(filerest,fileresu);
1.127 brouard 13159: if((ficrest=fopen(filerest,"w"))==NULL) {
13160: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
13161: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
13162: }
1.208 brouard 13163: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
13164: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 13165: strcpy(fileresstde,"STDE_");
13166: strcat(fileresstde,fileresu);
1.126 brouard 13167: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 13168: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
13169: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 13170: }
1.227 brouard 13171: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
13172: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 13173:
1.201 brouard 13174: strcpy(filerescve,"CVE_");
13175: strcat(filerescve,fileresu);
1.126 brouard 13176: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 13177: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
13178: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 13179: }
1.227 brouard 13180: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
13181: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 13182:
1.201 brouard 13183: strcpy(fileresv,"V_");
13184: strcat(fileresv,fileresu);
1.126 brouard 13185: if((ficresvij=fopen(fileresv,"w"))==NULL) {
13186: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
13187: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
13188: }
1.227 brouard 13189: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
13190: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 13191:
1.235 brouard 13192: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13193: if (cptcovn < 1){i1=1;}
13194:
13195: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13196: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13197: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13198: continue;
1.321 brouard 13199: printf("\n# model %s \n#****** Result for:", model);
13200: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
13201: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.227 brouard 13202: for(j=1;j<=cptcoveff;j++){
13203: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13204: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13205: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13206: }
1.235 brouard 13207: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13208: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13209: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13210: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13211: }
1.208 brouard 13212: fprintf(ficrest,"******\n");
1.227 brouard 13213: fprintf(ficlog,"******\n");
13214: printf("******\n");
1.208 brouard 13215:
13216: fprintf(ficresstdeij,"\n#****** ");
13217: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 13218: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 13219: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13220: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 13221: }
1.235 brouard 13222: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13223: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13224: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13225: }
1.208 brouard 13226: fprintf(ficresstdeij,"******\n");
13227: fprintf(ficrescveij,"******\n");
13228:
13229: fprintf(ficresvij,"\n#****** ");
1.238 brouard 13230: /* pstamp(ficresvij); */
1.225 brouard 13231: for(j=1;j<=cptcoveff;j++)
1.227 brouard 13232: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 13233: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13234: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13235: }
1.208 brouard 13236: fprintf(ficresvij,"******\n");
13237:
13238: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13239: oldm=oldms;savm=savms;
1.235 brouard 13240: printf(" cvevsij ");
13241: fprintf(ficlog, " cvevsij ");
13242: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 13243: printf(" end cvevsij \n ");
13244: fprintf(ficlog, " end cvevsij \n ");
13245:
13246: /*
13247: */
13248: /* goto endfree; */
13249:
13250: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13251: pstamp(ficrest);
13252:
1.269 brouard 13253: epj=vector(1,nlstate+1);
1.208 brouard 13254: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 13255: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
13256: cptcod= 0; /* To be deleted */
13257: printf("varevsij vpopbased=%d \n",vpopbased);
13258: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 13259: 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 13260: 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 ");
13261: if(vpopbased==1)
13262: 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);
13263: else
1.288 brouard 13264: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13265: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
13266: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
13267: fprintf(ficrest,"\n");
13268: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 13269: printf("Computing age specific forward period (stable) prevalences in each health state \n");
13270: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13271: for(age=bage; age <=fage ;age++){
1.235 brouard 13272: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 13273: if (vpopbased==1) {
13274: if(mobilav ==0){
13275: for(i=1; i<=nlstate;i++)
13276: prlim[i][i]=probs[(int)age][i][k];
13277: }else{ /* mobilav */
13278: for(i=1; i<=nlstate;i++)
13279: prlim[i][i]=mobaverage[(int)age][i][k];
13280: }
13281: }
1.219 brouard 13282:
1.227 brouard 13283: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
13284: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
13285: /* printf(" age %4.0f ",age); */
13286: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
13287: for(i=1, epj[j]=0.;i <=nlstate;i++) {
13288: epj[j] += prlim[i][i]*eij[i][j][(int)age];
13289: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
13290: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
13291: }
13292: epj[nlstate+1] +=epj[j];
13293: }
13294: /* printf(" age %4.0f \n",age); */
1.219 brouard 13295:
1.227 brouard 13296: for(i=1, vepp=0.;i <=nlstate;i++)
13297: for(j=1;j <=nlstate;j++)
13298: vepp += vareij[i][j][(int)age];
13299: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
13300: for(j=1;j <=nlstate;j++){
13301: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
13302: }
13303: fprintf(ficrest,"\n");
13304: }
1.208 brouard 13305: } /* End vpopbased */
1.269 brouard 13306: free_vector(epj,1,nlstate+1);
1.208 brouard 13307: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
13308: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 13309: printf("done selection\n");fflush(stdout);
13310: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 13311:
1.235 brouard 13312: } /* End k selection */
1.227 brouard 13313:
13314: printf("done State-specific expectancies\n");fflush(stdout);
13315: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
13316:
1.288 brouard 13317: /* variance-covariance of forward period prevalence*/
1.269 brouard 13318: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13319:
1.227 brouard 13320:
1.290 brouard 13321: free_vector(weight,firstobs,lastobs);
1.227 brouard 13322: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 13323: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
13324: free_matrix(anint,1,maxwav,firstobs,lastobs);
13325: free_matrix(mint,1,maxwav,firstobs,lastobs);
13326: free_ivector(cod,firstobs,lastobs);
1.227 brouard 13327: free_ivector(tab,1,NCOVMAX);
13328: fclose(ficresstdeij);
13329: fclose(ficrescveij);
13330: fclose(ficresvij);
13331: fclose(ficrest);
13332: fclose(ficpar);
13333:
13334:
1.126 brouard 13335: /*---------- End : free ----------------*/
1.219 brouard 13336: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 13337: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
13338: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 13339: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
13340: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 13341: } /* mle==-3 arrives here for freeing */
1.227 brouard 13342: /* endfree:*/
13343: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
13344: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
13345: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 13346: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
13347: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
13348: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
13349: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 13350: free_matrix(matcov,1,npar,1,npar);
13351: free_matrix(hess,1,npar,1,npar);
13352: /*free_vector(delti,1,npar);*/
13353: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13354: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 13355: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 13356: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13357:
13358: free_ivector(ncodemax,1,NCOVMAX);
13359: free_ivector(ncodemaxwundef,1,NCOVMAX);
13360: free_ivector(Dummy,-1,NCOVMAX);
13361: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 13362: free_ivector(DummyV,1,NCOVMAX);
13363: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 13364: free_ivector(Typevar,-1,NCOVMAX);
13365: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 13366: free_ivector(TvarsQ,1,NCOVMAX);
13367: free_ivector(TvarsQind,1,NCOVMAX);
13368: free_ivector(TvarsD,1,NCOVMAX);
13369: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 13370: free_ivector(TvarFD,1,NCOVMAX);
13371: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 13372: free_ivector(TvarF,1,NCOVMAX);
13373: free_ivector(TvarFind,1,NCOVMAX);
13374: free_ivector(TvarV,1,NCOVMAX);
13375: free_ivector(TvarVind,1,NCOVMAX);
13376: free_ivector(TvarA,1,NCOVMAX);
13377: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 13378: free_ivector(TvarFQ,1,NCOVMAX);
13379: free_ivector(TvarFQind,1,NCOVMAX);
13380: free_ivector(TvarVD,1,NCOVMAX);
13381: free_ivector(TvarVDind,1,NCOVMAX);
13382: free_ivector(TvarVQ,1,NCOVMAX);
13383: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 13384: free_ivector(Tvarsel,1,NCOVMAX);
13385: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 13386: free_ivector(Tposprod,1,NCOVMAX);
13387: free_ivector(Tprod,1,NCOVMAX);
13388: free_ivector(Tvaraff,1,NCOVMAX);
13389: free_ivector(invalidvarcomb,1,ncovcombmax);
13390: free_ivector(Tage,1,NCOVMAX);
13391: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 13392: free_ivector(TmodelInvind,1,NCOVMAX);
13393: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 13394:
13395: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
13396: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 13397: fflush(fichtm);
13398: fflush(ficgp);
13399:
1.227 brouard 13400:
1.126 brouard 13401: if((nberr >0) || (nbwarn>0)){
1.216 brouard 13402: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
13403: 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 13404: }else{
13405: printf("End of Imach\n");
13406: fprintf(ficlog,"End of Imach\n");
13407: }
13408: printf("See log file on %s\n",filelog);
13409: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 13410: /*(void) gettimeofday(&end_time,&tzp);*/
13411: rend_time = time(NULL);
13412: end_time = *localtime(&rend_time);
13413: /* tml = *localtime(&end_time.tm_sec); */
13414: strcpy(strtend,asctime(&end_time));
1.126 brouard 13415: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
13416: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 13417: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 13418:
1.157 brouard 13419: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
13420: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
13421: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 13422: /* printf("Total time was %d uSec.\n", total_usecs);*/
13423: /* if(fileappend(fichtm,optionfilehtm)){ */
13424: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13425: fclose(fichtm);
13426: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13427: fclose(fichtmcov);
13428: fclose(ficgp);
13429: fclose(ficlog);
13430: /*------ End -----------*/
1.227 brouard 13431:
1.281 brouard 13432:
13433: /* Executes gnuplot */
1.227 brouard 13434:
13435: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 13436: #ifdef WIN32
1.227 brouard 13437: if (_chdir(pathcd) != 0)
13438: printf("Can't move to directory %s!\n",path);
13439: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 13440: #else
1.227 brouard 13441: if(chdir(pathcd) != 0)
13442: printf("Can't move to directory %s!\n", path);
13443: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 13444: #endif
1.126 brouard 13445: printf("Current directory %s!\n",pathcd);
13446: /*strcat(plotcmd,CHARSEPARATOR);*/
13447: sprintf(plotcmd,"gnuplot");
1.157 brouard 13448: #ifdef _WIN32
1.126 brouard 13449: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
13450: #endif
13451: if(!stat(plotcmd,&info)){
1.158 brouard 13452: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13453: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 13454: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 13455: }else
13456: strcpy(pplotcmd,plotcmd);
1.157 brouard 13457: #ifdef __unix
1.126 brouard 13458: strcpy(plotcmd,GNUPLOTPROGRAM);
13459: if(!stat(plotcmd,&info)){
1.158 brouard 13460: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13461: }else
13462: strcpy(pplotcmd,plotcmd);
13463: #endif
13464: }else
13465: strcpy(pplotcmd,plotcmd);
13466:
13467: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 13468: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 13469: strcpy(pplotcmd,plotcmd);
1.227 brouard 13470:
1.126 brouard 13471: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 13472: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 13473: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 13474: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 13475: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 13476: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 13477: strcpy(plotcmd,pplotcmd);
13478: }
1.126 brouard 13479: }
1.158 brouard 13480: printf(" Successful, please wait...");
1.126 brouard 13481: while (z[0] != 'q') {
13482: /* chdir(path); */
1.154 brouard 13483: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 13484: scanf("%s",z);
13485: /* if (z[0] == 'c') system("./imach"); */
13486: if (z[0] == 'e') {
1.158 brouard 13487: #ifdef __APPLE__
1.152 brouard 13488: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 13489: #elif __linux
13490: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 13491: #else
1.152 brouard 13492: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13493: #endif
13494: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13495: system(pplotcmd);
1.126 brouard 13496: }
13497: else if (z[0] == 'g') system(plotcmd);
13498: else if (z[0] == 'q') exit(0);
13499: }
1.227 brouard 13500: end:
1.126 brouard 13501: while (z[0] != 'q') {
1.195 brouard 13502: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13503: scanf("%s",z);
13504: }
1.283 brouard 13505: printf("End\n");
1.282 brouard 13506: exit(0);
1.126 brouard 13507: }
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