Annotation of imach/src/imach.c, revision 1.327
1.327 ! brouard 1: /* $Id: imach.c,v 1.326 2022/07/26 17:33:55 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.327 ! brouard 4: Revision 1.326 2022/07/26 17:33:55 brouard
! 5: Summary: some test with nres=1
! 6:
1.326 brouard 7: Revision 1.325 2022/07/25 14:27:23 brouard
8: Summary: r30
9:
10: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
11: coredumped, revealed by Feiuno, thank you.
12:
1.325 brouard 13: Revision 1.324 2022/07/23 17:44:26 brouard
14: *** empty log message ***
15:
1.324 brouard 16: Revision 1.323 2022/07/22 12:30:08 brouard
17: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
18:
1.323 brouard 19: Revision 1.322 2022/07/22 12:27:48 brouard
20: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
21:
1.322 brouard 22: Revision 1.321 2022/07/22 12:04:24 brouard
23: Summary: r28
24:
25: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
26:
1.321 brouard 27: Revision 1.320 2022/06/02 05:10:11 brouard
28: *** empty log message ***
29:
1.320 brouard 30: Revision 1.319 2022/06/02 04:45:11 brouard
31: * imach.c (Module): Adding the Wald tests from the log to the main
32: htm for better display of the maximum likelihood estimators.
33:
1.319 brouard 34: Revision 1.318 2022/05/24 08:10:59 brouard
35: * imach.c (Module): Some attempts to find a bug of wrong estimates
36: of confidencce intervals with product in the equation modelC
37:
1.318 brouard 38: Revision 1.317 2022/05/15 15:06:23 brouard
39: * imach.c (Module): Some minor improvements
40:
1.317 brouard 41: Revision 1.316 2022/05/11 15:11:31 brouard
42: Summary: r27
43:
1.316 brouard 44: Revision 1.315 2022/05/11 15:06:32 brouard
45: *** empty log message ***
46:
1.315 brouard 47: Revision 1.314 2022/04/13 17:43:09 brouard
48: * imach.c (Module): Adding link to text data files
49:
1.314 brouard 50: Revision 1.313 2022/04/11 15:57:42 brouard
51: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
52:
1.313 brouard 53: Revision 1.312 2022/04/05 21:24:39 brouard
54: *** empty log message ***
55:
1.312 brouard 56: Revision 1.311 2022/04/05 21:03:51 brouard
57: Summary: Fixed quantitative covariates
58:
59: Fixed covariates (dummy or quantitative)
60: with missing values have never been allowed but are ERRORS and
61: program quits. Standard deviations of fixed covariates were
62: wrongly computed. Mean and standard deviations of time varying
63: covariates are still not computed.
64:
1.311 brouard 65: Revision 1.310 2022/03/17 08:45:53 brouard
66: Summary: 99r25
67:
68: Improving detection of errors: result lines should be compatible with
69: the model.
70:
1.310 brouard 71: Revision 1.309 2021/05/20 12:39:14 brouard
72: Summary: Version 0.99r24
73:
1.309 brouard 74: Revision 1.308 2021/03/31 13:11:57 brouard
75: Summary: Version 0.99r23
76:
77:
78: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
79:
1.308 brouard 80: Revision 1.307 2021/03/08 18:11:32 brouard
81: Summary: 0.99r22 fixed bug on result:
82:
1.307 brouard 83: Revision 1.306 2021/02/20 15:44:02 brouard
84: Summary: Version 0.99r21
85:
86: * imach.c (Module): Fix bug on quitting after result lines!
87: (Module): Version 0.99r21
88:
1.306 brouard 89: Revision 1.305 2021/02/20 15:28:30 brouard
90: * imach.c (Module): Fix bug on quitting after result lines!
91:
1.305 brouard 92: Revision 1.304 2021/02/12 11:34:20 brouard
93: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
94:
1.304 brouard 95: Revision 1.303 2021/02/11 19:50:15 brouard
96: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
97:
1.303 brouard 98: Revision 1.302 2020/02/22 21:00:05 brouard
99: * (Module): imach.c Update mle=-3 (for computing Life expectancy
100: and life table from the data without any state)
101:
1.302 brouard 102: Revision 1.301 2019/06/04 13:51:20 brouard
103: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
104:
1.301 brouard 105: Revision 1.300 2019/05/22 19:09:45 brouard
106: Summary: version 0.99r19 of May 2019
107:
1.300 brouard 108: Revision 1.299 2019/05/22 18:37:08 brouard
109: Summary: Cleaned 0.99r19
110:
1.299 brouard 111: Revision 1.298 2019/05/22 18:19:56 brouard
112: *** empty log message ***
113:
1.298 brouard 114: Revision 1.297 2019/05/22 17:56:10 brouard
115: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
116:
1.297 brouard 117: Revision 1.296 2019/05/20 13:03:18 brouard
118: Summary: Projection syntax simplified
119:
120:
121: We can now start projections, forward or backward, from the mean date
122: of inteviews up to or down to a number of years of projection:
123: prevforecast=1 yearsfproj=15.3 mobil_average=0
124: or
125: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
126: or
127: prevbackcast=1 yearsbproj=12.3 mobil_average=1
128: or
129: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
130:
1.296 brouard 131: Revision 1.295 2019/05/18 09:52:50 brouard
132: Summary: doxygen tex bug
133:
1.295 brouard 134: Revision 1.294 2019/05/16 14:54:33 brouard
135: Summary: There was some wrong lines added
136:
1.294 brouard 137: Revision 1.293 2019/05/09 15:17:34 brouard
138: *** empty log message ***
139:
1.293 brouard 140: Revision 1.292 2019/05/09 14:17:20 brouard
141: Summary: Some updates
142:
1.292 brouard 143: Revision 1.291 2019/05/09 13:44:18 brouard
144: Summary: Before ncovmax
145:
1.291 brouard 146: Revision 1.290 2019/05/09 13:39:37 brouard
147: Summary: 0.99r18 unlimited number of individuals
148:
149: 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.
150:
1.290 brouard 151: Revision 1.289 2018/12/13 09:16:26 brouard
152: Summary: Bug for young ages (<-30) will be in r17
153:
1.289 brouard 154: Revision 1.288 2018/05/02 20:58:27 brouard
155: Summary: Some bugs fixed
156:
1.288 brouard 157: Revision 1.287 2018/05/01 17:57:25 brouard
158: Summary: Bug fixed by providing frequencies only for non missing covariates
159:
1.287 brouard 160: Revision 1.286 2018/04/27 14:27:04 brouard
161: Summary: some minor bugs
162:
1.286 brouard 163: Revision 1.285 2018/04/21 21:02:16 brouard
164: Summary: Some bugs fixed, valgrind tested
165:
1.285 brouard 166: Revision 1.284 2018/04/20 05:22:13 brouard
167: Summary: Computing mean and stdeviation of fixed quantitative variables
168:
1.284 brouard 169: Revision 1.283 2018/04/19 14:49:16 brouard
170: Summary: Some minor bugs fixed
171:
1.283 brouard 172: Revision 1.282 2018/02/27 22:50:02 brouard
173: *** empty log message ***
174:
1.282 brouard 175: Revision 1.281 2018/02/27 19:25:23 brouard
176: Summary: Adding second argument for quitting
177:
1.281 brouard 178: Revision 1.280 2018/02/21 07:58:13 brouard
179: Summary: 0.99r15
180:
181: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
182:
1.280 brouard 183: Revision 1.279 2017/07/20 13:35:01 brouard
184: Summary: temporary working
185:
1.279 brouard 186: Revision 1.278 2017/07/19 14:09:02 brouard
187: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
188:
1.278 brouard 189: Revision 1.277 2017/07/17 08:53:49 brouard
190: Summary: BOM files can be read now
191:
1.277 brouard 192: Revision 1.276 2017/06/30 15:48:31 brouard
193: Summary: Graphs improvements
194:
1.276 brouard 195: Revision 1.275 2017/06/30 13:39:33 brouard
196: Summary: Saito's color
197:
1.275 brouard 198: Revision 1.274 2017/06/29 09:47:08 brouard
199: Summary: Version 0.99r14
200:
1.274 brouard 201: Revision 1.273 2017/06/27 11:06:02 brouard
202: Summary: More documentation on projections
203:
1.273 brouard 204: Revision 1.272 2017/06/27 10:22:40 brouard
205: Summary: Color of backprojection changed from 6 to 5(yellow)
206:
1.272 brouard 207: Revision 1.271 2017/06/27 10:17:50 brouard
208: Summary: Some bug with rint
209:
1.271 brouard 210: Revision 1.270 2017/05/24 05:45:29 brouard
211: *** empty log message ***
212:
1.270 brouard 213: Revision 1.269 2017/05/23 08:39:25 brouard
214: Summary: Code into subroutine, cleanings
215:
1.269 brouard 216: Revision 1.268 2017/05/18 20:09:32 brouard
217: Summary: backprojection and confidence intervals of backprevalence
218:
1.268 brouard 219: Revision 1.267 2017/05/13 10:25:05 brouard
220: Summary: temporary save for backprojection
221:
1.267 brouard 222: Revision 1.266 2017/05/13 07:26:12 brouard
223: Summary: Version 0.99r13 (improvements and bugs fixed)
224:
1.266 brouard 225: Revision 1.265 2017/04/26 16:22:11 brouard
226: Summary: imach 0.99r13 Some bugs fixed
227:
1.265 brouard 228: Revision 1.264 2017/04/26 06:01:29 brouard
229: Summary: Labels in graphs
230:
1.264 brouard 231: Revision 1.263 2017/04/24 15:23:15 brouard
232: Summary: to save
233:
1.263 brouard 234: Revision 1.262 2017/04/18 16:48:12 brouard
235: *** empty log message ***
236:
1.262 brouard 237: Revision 1.261 2017/04/05 10:14:09 brouard
238: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
239:
1.261 brouard 240: Revision 1.260 2017/04/04 17:46:59 brouard
241: Summary: Gnuplot indexations fixed (humm)
242:
1.260 brouard 243: Revision 1.259 2017/04/04 13:01:16 brouard
244: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
245:
1.259 brouard 246: Revision 1.258 2017/04/03 10:17:47 brouard
247: Summary: Version 0.99r12
248:
249: Some cleanings, conformed with updated documentation.
250:
1.258 brouard 251: Revision 1.257 2017/03/29 16:53:30 brouard
252: Summary: Temp
253:
1.257 brouard 254: Revision 1.256 2017/03/27 05:50:23 brouard
255: Summary: Temporary
256:
1.256 brouard 257: Revision 1.255 2017/03/08 16:02:28 brouard
258: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
259:
1.255 brouard 260: Revision 1.254 2017/03/08 07:13:00 brouard
261: Summary: Fixing data parameter line
262:
1.254 brouard 263: Revision 1.253 2016/12/15 11:59:41 brouard
264: Summary: 0.99 in progress
265:
1.253 brouard 266: Revision 1.252 2016/09/15 21:15:37 brouard
267: *** empty log message ***
268:
1.252 brouard 269: Revision 1.251 2016/09/15 15:01:13 brouard
270: Summary: not working
271:
1.251 brouard 272: Revision 1.250 2016/09/08 16:07:27 brouard
273: Summary: continue
274:
1.250 brouard 275: Revision 1.249 2016/09/07 17:14:18 brouard
276: Summary: Starting values from frequencies
277:
1.249 brouard 278: Revision 1.248 2016/09/07 14:10:18 brouard
279: *** empty log message ***
280:
1.248 brouard 281: Revision 1.247 2016/09/02 11:11:21 brouard
282: *** empty log message ***
283:
1.247 brouard 284: Revision 1.246 2016/09/02 08:49:22 brouard
285: *** empty log message ***
286:
1.246 brouard 287: Revision 1.245 2016/09/02 07:25:01 brouard
288: *** empty log message ***
289:
1.245 brouard 290: Revision 1.244 2016/09/02 07:17:34 brouard
291: *** empty log message ***
292:
1.244 brouard 293: Revision 1.243 2016/09/02 06:45:35 brouard
294: *** empty log message ***
295:
1.243 brouard 296: Revision 1.242 2016/08/30 15:01:20 brouard
297: Summary: Fixing a lots
298:
1.242 brouard 299: Revision 1.241 2016/08/29 17:17:25 brouard
300: Summary: gnuplot problem in Back projection to fix
301:
1.241 brouard 302: Revision 1.240 2016/08/29 07:53:18 brouard
303: Summary: Better
304:
1.240 brouard 305: Revision 1.239 2016/08/26 15:51:03 brouard
306: Summary: Improvement in Powell output in order to copy and paste
307:
308: Author:
309:
1.239 brouard 310: Revision 1.238 2016/08/26 14:23:35 brouard
311: Summary: Starting tests of 0.99
312:
1.238 brouard 313: Revision 1.237 2016/08/26 09:20:19 brouard
314: Summary: to valgrind
315:
1.237 brouard 316: Revision 1.236 2016/08/25 10:50:18 brouard
317: *** empty log message ***
318:
1.236 brouard 319: Revision 1.235 2016/08/25 06:59:23 brouard
320: *** empty log message ***
321:
1.235 brouard 322: Revision 1.234 2016/08/23 16:51:20 brouard
323: *** empty log message ***
324:
1.234 brouard 325: Revision 1.233 2016/08/23 07:40:50 brouard
326: Summary: not working
327:
1.233 brouard 328: Revision 1.232 2016/08/22 14:20:21 brouard
329: Summary: not working
330:
1.232 brouard 331: Revision 1.231 2016/08/22 07:17:15 brouard
332: Summary: not working
333:
1.231 brouard 334: Revision 1.230 2016/08/22 06:55:53 brouard
335: Summary: Not working
336:
1.230 brouard 337: Revision 1.229 2016/07/23 09:45:53 brouard
338: Summary: Completing for func too
339:
1.229 brouard 340: Revision 1.228 2016/07/22 17:45:30 brouard
341: Summary: Fixing some arrays, still debugging
342:
1.227 brouard 343: Revision 1.226 2016/07/12 18:42:34 brouard
344: Summary: temp
345:
1.226 brouard 346: Revision 1.225 2016/07/12 08:40:03 brouard
347: Summary: saving but not running
348:
1.225 brouard 349: Revision 1.224 2016/07/01 13:16:01 brouard
350: Summary: Fixes
351:
1.224 brouard 352: Revision 1.223 2016/02/19 09:23:35 brouard
353: Summary: temporary
354:
1.223 brouard 355: Revision 1.222 2016/02/17 08:14:50 brouard
356: Summary: Probably last 0.98 stable version 0.98r6
357:
1.222 brouard 358: Revision 1.221 2016/02/15 23:35:36 brouard
359: Summary: minor bug
360:
1.220 brouard 361: Revision 1.219 2016/02/15 00:48:12 brouard
362: *** empty log message ***
363:
1.219 brouard 364: Revision 1.218 2016/02/12 11:29:23 brouard
365: Summary: 0.99 Back projections
366:
1.218 brouard 367: Revision 1.217 2015/12/23 17:18:31 brouard
368: Summary: Experimental backcast
369:
1.217 brouard 370: Revision 1.216 2015/12/18 17:32:11 brouard
371: Summary: 0.98r4 Warning and status=-2
372:
373: Version 0.98r4 is now:
374: - displaying an error when status is -1, date of interview unknown and date of death known;
375: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
376: Older changes concerning s=-2, dating from 2005 have been supersed.
377:
1.216 brouard 378: Revision 1.215 2015/12/16 08:52:24 brouard
379: Summary: 0.98r4 working
380:
1.215 brouard 381: Revision 1.214 2015/12/16 06:57:54 brouard
382: Summary: temporary not working
383:
1.214 brouard 384: Revision 1.213 2015/12/11 18:22:17 brouard
385: Summary: 0.98r4
386:
1.213 brouard 387: Revision 1.212 2015/11/21 12:47:24 brouard
388: Summary: minor typo
389:
1.212 brouard 390: Revision 1.211 2015/11/21 12:41:11 brouard
391: Summary: 0.98r3 with some graph of projected cross-sectional
392:
393: Author: Nicolas Brouard
394:
1.211 brouard 395: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 396: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 397: Summary: Adding ftolpl parameter
398: Author: N Brouard
399:
400: We had difficulties to get smoothed confidence intervals. It was due
401: to the period prevalence which wasn't computed accurately. The inner
402: parameter ftolpl is now an outer parameter of the .imach parameter
403: file after estepm. If ftolpl is small 1.e-4 and estepm too,
404: computation are long.
405:
1.209 brouard 406: Revision 1.208 2015/11/17 14:31:57 brouard
407: Summary: temporary
408:
1.208 brouard 409: Revision 1.207 2015/10/27 17:36:57 brouard
410: *** empty log message ***
411:
1.207 brouard 412: Revision 1.206 2015/10/24 07:14:11 brouard
413: *** empty log message ***
414:
1.206 brouard 415: Revision 1.205 2015/10/23 15:50:53 brouard
416: Summary: 0.98r3 some clarification for graphs on likelihood contributions
417:
1.205 brouard 418: Revision 1.204 2015/10/01 16:20:26 brouard
419: Summary: Some new graphs of contribution to likelihood
420:
1.204 brouard 421: Revision 1.203 2015/09/30 17:45:14 brouard
422: Summary: looking at better estimation of the hessian
423:
424: Also a better criteria for convergence to the period prevalence And
425: therefore adding the number of years needed to converge. (The
426: prevalence in any alive state shold sum to one
427:
1.203 brouard 428: Revision 1.202 2015/09/22 19:45:16 brouard
429: Summary: Adding some overall graph on contribution to likelihood. Might change
430:
1.202 brouard 431: Revision 1.201 2015/09/15 17:34:58 brouard
432: Summary: 0.98r0
433:
434: - Some new graphs like suvival functions
435: - Some bugs fixed like model=1+age+V2.
436:
1.201 brouard 437: Revision 1.200 2015/09/09 16:53:55 brouard
438: Summary: Big bug thanks to Flavia
439:
440: Even model=1+age+V2. did not work anymore
441:
1.200 brouard 442: Revision 1.199 2015/09/07 14:09:23 brouard
443: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
444:
1.199 brouard 445: Revision 1.198 2015/09/03 07:14:39 brouard
446: Summary: 0.98q5 Flavia
447:
1.198 brouard 448: Revision 1.197 2015/09/01 18:24:39 brouard
449: *** empty log message ***
450:
1.197 brouard 451: Revision 1.196 2015/08/18 23:17:52 brouard
452: Summary: 0.98q5
453:
1.196 brouard 454: Revision 1.195 2015/08/18 16:28:39 brouard
455: Summary: Adding a hack for testing purpose
456:
457: After reading the title, ftol and model lines, if the comment line has
458: a q, starting with #q, the answer at the end of the run is quit. It
459: permits to run test files in batch with ctest. The former workaround was
460: $ echo q | imach foo.imach
461:
1.195 brouard 462: Revision 1.194 2015/08/18 13:32:00 brouard
463: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
464:
1.194 brouard 465: Revision 1.193 2015/08/04 07:17:42 brouard
466: Summary: 0.98q4
467:
1.193 brouard 468: Revision 1.192 2015/07/16 16:49:02 brouard
469: Summary: Fixing some outputs
470:
1.192 brouard 471: Revision 1.191 2015/07/14 10:00:33 brouard
472: Summary: Some fixes
473:
1.191 brouard 474: Revision 1.190 2015/05/05 08:51:13 brouard
475: Summary: Adding digits in output parameters (7 digits instead of 6)
476:
477: Fix 1+age+.
478:
1.190 brouard 479: Revision 1.189 2015/04/30 14:45:16 brouard
480: Summary: 0.98q2
481:
1.189 brouard 482: Revision 1.188 2015/04/30 08:27:53 brouard
483: *** empty log message ***
484:
1.188 brouard 485: Revision 1.187 2015/04/29 09:11:15 brouard
486: *** empty log message ***
487:
1.187 brouard 488: Revision 1.186 2015/04/23 12:01:52 brouard
489: Summary: V1*age is working now, version 0.98q1
490:
491: Some codes had been disabled in order to simplify and Vn*age was
492: working in the optimization phase, ie, giving correct MLE parameters,
493: but, as usual, outputs were not correct and program core dumped.
494:
1.186 brouard 495: Revision 1.185 2015/03/11 13:26:42 brouard
496: Summary: Inclusion of compile and links command line for Intel Compiler
497:
1.185 brouard 498: Revision 1.184 2015/03/11 11:52:39 brouard
499: Summary: Back from Windows 8. Intel Compiler
500:
1.184 brouard 501: Revision 1.183 2015/03/10 20:34:32 brouard
502: Summary: 0.98q0, trying with directest, mnbrak fixed
503:
504: We use directest instead of original Powell test; probably no
505: incidence on the results, but better justifications;
506: We fixed Numerical Recipes mnbrak routine which was wrong and gave
507: wrong results.
508:
1.183 brouard 509: Revision 1.182 2015/02/12 08:19:57 brouard
510: Summary: Trying to keep directest which seems simpler and more general
511: Author: Nicolas Brouard
512:
1.182 brouard 513: Revision 1.181 2015/02/11 23:22:24 brouard
514: Summary: Comments on Powell added
515:
516: Author:
517:
1.181 brouard 518: Revision 1.180 2015/02/11 17:33:45 brouard
519: Summary: Finishing move from main to function (hpijx and prevalence_limit)
520:
1.180 brouard 521: Revision 1.179 2015/01/04 09:57:06 brouard
522: Summary: back to OS/X
523:
1.179 brouard 524: Revision 1.178 2015/01/04 09:35:48 brouard
525: *** empty log message ***
526:
1.178 brouard 527: Revision 1.177 2015/01/03 18:40:56 brouard
528: Summary: Still testing ilc32 on OSX
529:
1.177 brouard 530: Revision 1.176 2015/01/03 16:45:04 brouard
531: *** empty log message ***
532:
1.176 brouard 533: Revision 1.175 2015/01/03 16:33:42 brouard
534: *** empty log message ***
535:
1.175 brouard 536: Revision 1.174 2015/01/03 16:15:49 brouard
537: Summary: Still in cross-compilation
538:
1.174 brouard 539: Revision 1.173 2015/01/03 12:06:26 brouard
540: Summary: trying to detect cross-compilation
541:
1.173 brouard 542: Revision 1.172 2014/12/27 12:07:47 brouard
543: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
544:
1.172 brouard 545: Revision 1.171 2014/12/23 13:26:59 brouard
546: Summary: Back from Visual C
547:
548: Still problem with utsname.h on Windows
549:
1.171 brouard 550: Revision 1.170 2014/12/23 11:17:12 brouard
551: Summary: Cleaning some \%% back to %%
552:
553: The escape was mandatory for a specific compiler (which one?), but too many warnings.
554:
1.170 brouard 555: Revision 1.169 2014/12/22 23:08:31 brouard
556: Summary: 0.98p
557:
558: Outputs some informations on compiler used, OS etc. Testing on different platforms.
559:
1.169 brouard 560: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 561: Summary: update
1.169 brouard 562:
1.168 brouard 563: Revision 1.167 2014/12/22 13:50:56 brouard
564: Summary: Testing uname and compiler version and if compiled 32 or 64
565:
566: Testing on Linux 64
567:
1.167 brouard 568: Revision 1.166 2014/12/22 11:40:47 brouard
569: *** empty log message ***
570:
1.166 brouard 571: Revision 1.165 2014/12/16 11:20:36 brouard
572: Summary: After compiling on Visual C
573:
574: * imach.c (Module): Merging 1.61 to 1.162
575:
1.165 brouard 576: Revision 1.164 2014/12/16 10:52:11 brouard
577: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
578:
579: * imach.c (Module): Merging 1.61 to 1.162
580:
1.164 brouard 581: Revision 1.163 2014/12/16 10:30:11 brouard
582: * imach.c (Module): Merging 1.61 to 1.162
583:
1.163 brouard 584: Revision 1.162 2014/09/25 11:43:39 brouard
585: Summary: temporary backup 0.99!
586:
1.162 brouard 587: Revision 1.1 2014/09/16 11:06:58 brouard
588: Summary: With some code (wrong) for nlopt
589:
590: Author:
591:
592: Revision 1.161 2014/09/15 20:41:41 brouard
593: Summary: Problem with macro SQR on Intel compiler
594:
1.161 brouard 595: Revision 1.160 2014/09/02 09:24:05 brouard
596: *** empty log message ***
597:
1.160 brouard 598: Revision 1.159 2014/09/01 10:34:10 brouard
599: Summary: WIN32
600: Author: Brouard
601:
1.159 brouard 602: Revision 1.158 2014/08/27 17:11:51 brouard
603: *** empty log message ***
604:
1.158 brouard 605: Revision 1.157 2014/08/27 16:26:55 brouard
606: Summary: Preparing windows Visual studio version
607: Author: Brouard
608:
609: In order to compile on Visual studio, time.h is now correct and time_t
610: and tm struct should be used. difftime should be used but sometimes I
611: just make the differences in raw time format (time(&now).
612: Trying to suppress #ifdef LINUX
613: Add xdg-open for __linux in order to open default browser.
614:
1.157 brouard 615: Revision 1.156 2014/08/25 20:10:10 brouard
616: *** empty log message ***
617:
1.156 brouard 618: Revision 1.155 2014/08/25 18:32:34 brouard
619: Summary: New compile, minor changes
620: Author: Brouard
621:
1.155 brouard 622: Revision 1.154 2014/06/20 17:32:08 brouard
623: Summary: Outputs now all graphs of convergence to period prevalence
624:
1.154 brouard 625: Revision 1.153 2014/06/20 16:45:46 brouard
626: Summary: If 3 live state, convergence to period prevalence on same graph
627: Author: Brouard
628:
1.153 brouard 629: Revision 1.152 2014/06/18 17:54:09 brouard
630: Summary: open browser, use gnuplot on same dir than imach if not found in the path
631:
1.152 brouard 632: Revision 1.151 2014/06/18 16:43:30 brouard
633: *** empty log message ***
634:
1.151 brouard 635: Revision 1.150 2014/06/18 16:42:35 brouard
636: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
637: Author: brouard
638:
1.150 brouard 639: Revision 1.149 2014/06/18 15:51:14 brouard
640: Summary: Some fixes in parameter files errors
641: Author: Nicolas Brouard
642:
1.149 brouard 643: Revision 1.148 2014/06/17 17:38:48 brouard
644: Summary: Nothing new
645: Author: Brouard
646:
647: Just a new packaging for OS/X version 0.98nS
648:
1.148 brouard 649: Revision 1.147 2014/06/16 10:33:11 brouard
650: *** empty log message ***
651:
1.147 brouard 652: Revision 1.146 2014/06/16 10:20:28 brouard
653: Summary: Merge
654: Author: Brouard
655:
656: Merge, before building revised version.
657:
1.146 brouard 658: Revision 1.145 2014/06/10 21:23:15 brouard
659: Summary: Debugging with valgrind
660: Author: Nicolas Brouard
661:
662: Lot of changes in order to output the results with some covariates
663: After the Edimburgh REVES conference 2014, it seems mandatory to
664: improve the code.
665: No more memory valgrind error but a lot has to be done in order to
666: continue the work of splitting the code into subroutines.
667: Also, decodemodel has been improved. Tricode is still not
668: optimal. nbcode should be improved. Documentation has been added in
669: the source code.
670:
1.144 brouard 671: Revision 1.143 2014/01/26 09:45:38 brouard
672: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
673:
674: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
675: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
676:
1.143 brouard 677: Revision 1.142 2014/01/26 03:57:36 brouard
678: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
679:
680: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
681:
1.142 brouard 682: Revision 1.141 2014/01/26 02:42:01 brouard
683: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
684:
1.141 brouard 685: Revision 1.140 2011/09/02 10:37:54 brouard
686: Summary: times.h is ok with mingw32 now.
687:
1.140 brouard 688: Revision 1.139 2010/06/14 07:50:17 brouard
689: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
690: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
691:
1.139 brouard 692: Revision 1.138 2010/04/30 18:19:40 brouard
693: *** empty log message ***
694:
1.138 brouard 695: Revision 1.137 2010/04/29 18:11:38 brouard
696: (Module): Checking covariates for more complex models
697: than V1+V2. A lot of change to be done. Unstable.
698:
1.137 brouard 699: Revision 1.136 2010/04/26 20:30:53 brouard
700: (Module): merging some libgsl code. Fixing computation
701: of likelione (using inter/intrapolation if mle = 0) in order to
702: get same likelihood as if mle=1.
703: Some cleaning of code and comments added.
704:
1.136 brouard 705: Revision 1.135 2009/10/29 15:33:14 brouard
706: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
707:
1.135 brouard 708: Revision 1.134 2009/10/29 13:18:53 brouard
709: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
710:
1.134 brouard 711: Revision 1.133 2009/07/06 10:21:25 brouard
712: just nforces
713:
1.133 brouard 714: Revision 1.132 2009/07/06 08:22:05 brouard
715: Many tings
716:
1.132 brouard 717: Revision 1.131 2009/06/20 16:22:47 brouard
718: Some dimensions resccaled
719:
1.131 brouard 720: Revision 1.130 2009/05/26 06:44:34 brouard
721: (Module): Max Covariate is now set to 20 instead of 8. A
722: lot of cleaning with variables initialized to 0. Trying to make
723: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
724:
1.130 brouard 725: Revision 1.129 2007/08/31 13:49:27 lievre
726: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
727:
1.129 lievre 728: Revision 1.128 2006/06/30 13:02:05 brouard
729: (Module): Clarifications on computing e.j
730:
1.128 brouard 731: Revision 1.127 2006/04/28 18:11:50 brouard
732: (Module): Yes the sum of survivors was wrong since
733: imach-114 because nhstepm was no more computed in the age
734: loop. Now we define nhstepma in the age loop.
735: (Module): In order to speed up (in case of numerous covariates) we
736: compute health expectancies (without variances) in a first step
737: and then all the health expectancies with variances or standard
738: deviation (needs data from the Hessian matrices) which slows the
739: computation.
740: In the future we should be able to stop the program is only health
741: expectancies and graph are needed without standard deviations.
742:
1.127 brouard 743: Revision 1.126 2006/04/28 17:23:28 brouard
744: (Module): Yes the sum of survivors was wrong since
745: imach-114 because nhstepm was no more computed in the age
746: loop. Now we define nhstepma in the age loop.
747: Version 0.98h
748:
1.126 brouard 749: Revision 1.125 2006/04/04 15:20:31 lievre
750: Errors in calculation of health expectancies. Age was not initialized.
751: Forecasting file added.
752:
753: Revision 1.124 2006/03/22 17:13:53 lievre
754: Parameters are printed with %lf instead of %f (more numbers after the comma).
755: The log-likelihood is printed in the log file
756:
757: Revision 1.123 2006/03/20 10:52:43 brouard
758: * imach.c (Module): <title> changed, corresponds to .htm file
759: name. <head> headers where missing.
760:
761: * imach.c (Module): Weights can have a decimal point as for
762: English (a comma might work with a correct LC_NUMERIC environment,
763: otherwise the weight is truncated).
764: Modification of warning when the covariates values are not 0 or
765: 1.
766: Version 0.98g
767:
768: Revision 1.122 2006/03/20 09:45:41 brouard
769: (Module): Weights can have a decimal point as for
770: English (a comma might work with a correct LC_NUMERIC environment,
771: otherwise the weight is truncated).
772: Modification of warning when the covariates values are not 0 or
773: 1.
774: Version 0.98g
775:
776: Revision 1.121 2006/03/16 17:45:01 lievre
777: * imach.c (Module): Comments concerning covariates added
778:
779: * imach.c (Module): refinements in the computation of lli if
780: status=-2 in order to have more reliable computation if stepm is
781: not 1 month. Version 0.98f
782:
783: Revision 1.120 2006/03/16 15:10:38 lievre
784: (Module): refinements in the computation of lli if
785: status=-2 in order to have more reliable computation if stepm is
786: not 1 month. Version 0.98f
787:
788: Revision 1.119 2006/03/15 17:42:26 brouard
789: (Module): Bug if status = -2, the loglikelihood was
790: computed as likelihood omitting the logarithm. Version O.98e
791:
792: Revision 1.118 2006/03/14 18:20:07 brouard
793: (Module): varevsij Comments added explaining the second
794: table of variances if popbased=1 .
795: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
796: (Module): Function pstamp added
797: (Module): Version 0.98d
798:
799: Revision 1.117 2006/03/14 17:16:22 brouard
800: (Module): varevsij Comments added explaining the second
801: table of variances if popbased=1 .
802: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
803: (Module): Function pstamp added
804: (Module): Version 0.98d
805:
806: Revision 1.116 2006/03/06 10:29:27 brouard
807: (Module): Variance-covariance wrong links and
808: varian-covariance of ej. is needed (Saito).
809:
810: Revision 1.115 2006/02/27 12:17:45 brouard
811: (Module): One freematrix added in mlikeli! 0.98c
812:
813: Revision 1.114 2006/02/26 12:57:58 brouard
814: (Module): Some improvements in processing parameter
815: filename with strsep.
816:
817: Revision 1.113 2006/02/24 14:20:24 brouard
818: (Module): Memory leaks checks with valgrind and:
819: datafile was not closed, some imatrix were not freed and on matrix
820: allocation too.
821:
822: Revision 1.112 2006/01/30 09:55:26 brouard
823: (Module): Back to gnuplot.exe instead of wgnuplot.exe
824:
825: Revision 1.111 2006/01/25 20:38:18 brouard
826: (Module): Lots of cleaning and bugs added (Gompertz)
827: (Module): Comments can be added in data file. Missing date values
828: can be a simple dot '.'.
829:
830: Revision 1.110 2006/01/25 00:51:50 brouard
831: (Module): Lots of cleaning and bugs added (Gompertz)
832:
833: Revision 1.109 2006/01/24 19:37:15 brouard
834: (Module): Comments (lines starting with a #) are allowed in data.
835:
836: Revision 1.108 2006/01/19 18:05:42 lievre
837: Gnuplot problem appeared...
838: To be fixed
839:
840: Revision 1.107 2006/01/19 16:20:37 brouard
841: Test existence of gnuplot in imach path
842:
843: Revision 1.106 2006/01/19 13:24:36 brouard
844: Some cleaning and links added in html output
845:
846: Revision 1.105 2006/01/05 20:23:19 lievre
847: *** empty log message ***
848:
849: Revision 1.104 2005/09/30 16:11:43 lievre
850: (Module): sump fixed, loop imx fixed, and simplifications.
851: (Module): If the status is missing at the last wave but we know
852: that the person is alive, then we can code his/her status as -2
853: (instead of missing=-1 in earlier versions) and his/her
854: contributions to the likelihood is 1 - Prob of dying from last
855: health status (= 1-p13= p11+p12 in the easiest case of somebody in
856: the healthy state at last known wave). Version is 0.98
857:
858: Revision 1.103 2005/09/30 15:54:49 lievre
859: (Module): sump fixed, loop imx fixed, and simplifications.
860:
861: Revision 1.102 2004/09/15 17:31:30 brouard
862: Add the possibility to read data file including tab characters.
863:
864: Revision 1.101 2004/09/15 10:38:38 brouard
865: Fix on curr_time
866:
867: Revision 1.100 2004/07/12 18:29:06 brouard
868: Add version for Mac OS X. Just define UNIX in Makefile
869:
870: Revision 1.99 2004/06/05 08:57:40 brouard
871: *** empty log message ***
872:
873: Revision 1.98 2004/05/16 15:05:56 brouard
874: New version 0.97 . First attempt to estimate force of mortality
875: directly from the data i.e. without the need of knowing the health
876: state at each age, but using a Gompertz model: log u =a + b*age .
877: This is the basic analysis of mortality and should be done before any
878: other analysis, in order to test if the mortality estimated from the
879: cross-longitudinal survey is different from the mortality estimated
880: from other sources like vital statistic data.
881:
882: The same imach parameter file can be used but the option for mle should be -3.
883:
1.324 brouard 884: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 885: former routines in order to include the new code within the former code.
886:
887: The output is very simple: only an estimate of the intercept and of
888: the slope with 95% confident intervals.
889:
890: Current limitations:
891: A) Even if you enter covariates, i.e. with the
892: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
893: B) There is no computation of Life Expectancy nor Life Table.
894:
895: Revision 1.97 2004/02/20 13:25:42 lievre
896: Version 0.96d. Population forecasting command line is (temporarily)
897: suppressed.
898:
899: Revision 1.96 2003/07/15 15:38:55 brouard
900: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
901: rewritten within the same printf. Workaround: many printfs.
902:
903: Revision 1.95 2003/07/08 07:54:34 brouard
904: * imach.c (Repository):
905: (Repository): Using imachwizard code to output a more meaningful covariance
906: matrix (cov(a12,c31) instead of numbers.
907:
908: Revision 1.94 2003/06/27 13:00:02 brouard
909: Just cleaning
910:
911: Revision 1.93 2003/06/25 16:33:55 brouard
912: (Module): On windows (cygwin) function asctime_r doesn't
913: exist so I changed back to asctime which exists.
914: (Module): Version 0.96b
915:
916: Revision 1.92 2003/06/25 16:30:45 brouard
917: (Module): On windows (cygwin) function asctime_r doesn't
918: exist so I changed back to asctime which exists.
919:
920: Revision 1.91 2003/06/25 15:30:29 brouard
921: * imach.c (Repository): Duplicated warning errors corrected.
922: (Repository): Elapsed time after each iteration is now output. It
923: helps to forecast when convergence will be reached. Elapsed time
924: is stamped in powell. We created a new html file for the graphs
925: concerning matrix of covariance. It has extension -cov.htm.
926:
927: Revision 1.90 2003/06/24 12:34:15 brouard
928: (Module): Some bugs corrected for windows. Also, when
929: mle=-1 a template is output in file "or"mypar.txt with the design
930: of the covariance matrix to be input.
931:
932: Revision 1.89 2003/06/24 12:30:52 brouard
933: (Module): Some bugs corrected for windows. Also, when
934: mle=-1 a template is output in file "or"mypar.txt with the design
935: of the covariance matrix to be input.
936:
937: Revision 1.88 2003/06/23 17:54:56 brouard
938: * 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.
939:
940: Revision 1.87 2003/06/18 12:26:01 brouard
941: Version 0.96
942:
943: Revision 1.86 2003/06/17 20:04:08 brouard
944: (Module): Change position of html and gnuplot routines and added
945: routine fileappend.
946:
947: Revision 1.85 2003/06/17 13:12:43 brouard
948: * imach.c (Repository): Check when date of death was earlier that
949: current date of interview. It may happen when the death was just
950: prior to the death. In this case, dh was negative and likelihood
951: was wrong (infinity). We still send an "Error" but patch by
952: assuming that the date of death was just one stepm after the
953: interview.
954: (Repository): Because some people have very long ID (first column)
955: we changed int to long in num[] and we added a new lvector for
956: memory allocation. But we also truncated to 8 characters (left
957: truncation)
958: (Repository): No more line truncation errors.
959:
960: Revision 1.84 2003/06/13 21:44:43 brouard
961: * imach.c (Repository): Replace "freqsummary" at a correct
962: place. It differs from routine "prevalence" which may be called
963: many times. Probs is memory consuming and must be used with
964: parcimony.
965: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
966:
967: Revision 1.83 2003/06/10 13:39:11 lievre
968: *** empty log message ***
969:
970: Revision 1.82 2003/06/05 15:57:20 brouard
971: Add log in imach.c and fullversion number is now printed.
972:
973: */
974: /*
975: Interpolated Markov Chain
976:
977: Short summary of the programme:
978:
1.227 brouard 979: This program computes Healthy Life Expectancies or State-specific
980: (if states aren't health statuses) Expectancies from
981: cross-longitudinal data. Cross-longitudinal data consist in:
982:
983: -1- a first survey ("cross") where individuals from different ages
984: are interviewed on their health status or degree of disability (in
985: the case of a health survey which is our main interest)
986:
987: -2- at least a second wave of interviews ("longitudinal") which
988: measure each change (if any) in individual health status. Health
989: expectancies are computed from the time spent in each health state
990: according to a model. More health states you consider, more time is
991: necessary to reach the Maximum Likelihood of the parameters involved
992: in the model. The simplest model is the multinomial logistic model
993: where pij is the probability to be observed in state j at the second
994: wave conditional to be observed in state i at the first
995: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
996: etc , where 'age' is age and 'sex' is a covariate. If you want to
997: have a more complex model than "constant and age", you should modify
998: the program where the markup *Covariates have to be included here
999: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1000: convergence.
1001:
1002: The advantage of this computer programme, compared to a simple
1003: multinomial logistic model, is clear when the delay between waves is not
1004: identical for each individual. Also, if a individual missed an
1005: intermediate interview, the information is lost, but taken into
1006: account using an interpolation or extrapolation.
1007:
1008: hPijx is the probability to be observed in state i at age x+h
1009: conditional to the observed state i at age x. The delay 'h' can be
1010: split into an exact number (nh*stepm) of unobserved intermediate
1011: states. This elementary transition (by month, quarter,
1012: semester or year) is modelled as a multinomial logistic. The hPx
1013: matrix is simply the matrix product of nh*stepm elementary matrices
1014: and the contribution of each individual to the likelihood is simply
1015: hPijx.
1016:
1017: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1018: of the life expectancies. It also computes the period (stable) prevalence.
1019:
1020: Back prevalence and projections:
1.227 brouard 1021:
1022: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1023: double agemaxpar, double ftolpl, int *ncvyearp, double
1024: dateprev1,double dateprev2, int firstpass, int lastpass, int
1025: mobilavproj)
1026:
1027: Computes the back prevalence limit for any combination of
1028: covariate values k at any age between ageminpar and agemaxpar and
1029: returns it in **bprlim. In the loops,
1030:
1031: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1032: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1033:
1034: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1035: Computes for any combination of covariates k and any age between bage and fage
1036: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1037: oldm=oldms;savm=savms;
1.227 brouard 1038:
1.267 brouard 1039: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1040: Computes the transition matrix starting at age 'age' over
1041: 'nhstepm*hstepm*stepm' months (i.e. until
1042: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1043: nhstepm*hstepm matrices.
1044:
1045: Returns p3mat[i][j][h] after calling
1046: p3mat[i][j][h]=matprod2(newm,
1047: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1048: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1049: oldm);
1.226 brouard 1050:
1051: Important routines
1052:
1053: - func (or funcone), computes logit (pij) distinguishing
1054: o fixed variables (single or product dummies or quantitative);
1055: o varying variables by:
1056: (1) wave (single, product dummies, quantitative),
1057: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1058: % fixed dummy (treated) or quantitative (not done because time-consuming);
1059: % varying dummy (not done) or quantitative (not done);
1060: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1061: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1062: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1063: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1064: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1065:
1.226 brouard 1066:
1067:
1.324 brouard 1068: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1069: Institut national d'études démographiques, Paris.
1.126 brouard 1070: This software have been partly granted by Euro-REVES, a concerted action
1071: from the European Union.
1072: It is copyrighted identically to a GNU software product, ie programme and
1073: software can be distributed freely for non commercial use. Latest version
1074: can be accessed at http://euroreves.ined.fr/imach .
1075:
1076: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1077: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1078:
1079: **********************************************************************/
1080: /*
1081: main
1082: read parameterfile
1083: read datafile
1084: concatwav
1085: freqsummary
1086: if (mle >= 1)
1087: mlikeli
1088: print results files
1089: if mle==1
1090: computes hessian
1091: read end of parameter file: agemin, agemax, bage, fage, estepm
1092: begin-prev-date,...
1093: open gnuplot file
1094: open html file
1.145 brouard 1095: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1096: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1097: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1098: freexexit2 possible for memory heap.
1099:
1100: h Pij x | pij_nom ficrestpij
1101: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1102: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1103: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1104:
1105: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1106: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1107: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1108: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1109: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1110:
1.126 brouard 1111: forecasting if prevfcast==1 prevforecast call prevalence()
1112: health expectancies
1113: Variance-covariance of DFLE
1114: prevalence()
1115: movingaverage()
1116: varevsij()
1117: if popbased==1 varevsij(,popbased)
1118: total life expectancies
1119: Variance of period (stable) prevalence
1120: end
1121: */
1122:
1.187 brouard 1123: /* #define DEBUG */
1124: /* #define DEBUGBRENT */
1.203 brouard 1125: /* #define DEBUGLINMIN */
1126: /* #define DEBUGHESS */
1127: #define DEBUGHESSIJ
1.224 brouard 1128: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1129: #define POWELL /* Instead of NLOPT */
1.224 brouard 1130: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1131: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1132: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1133: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1134:
1135: #include <math.h>
1136: #include <stdio.h>
1137: #include <stdlib.h>
1138: #include <string.h>
1.226 brouard 1139: #include <ctype.h>
1.159 brouard 1140:
1141: #ifdef _WIN32
1142: #include <io.h>
1.172 brouard 1143: #include <windows.h>
1144: #include <tchar.h>
1.159 brouard 1145: #else
1.126 brouard 1146: #include <unistd.h>
1.159 brouard 1147: #endif
1.126 brouard 1148:
1149: #include <limits.h>
1150: #include <sys/types.h>
1.171 brouard 1151:
1152: #if defined(__GNUC__)
1153: #include <sys/utsname.h> /* Doesn't work on Windows */
1154: #endif
1155:
1.126 brouard 1156: #include <sys/stat.h>
1157: #include <errno.h>
1.159 brouard 1158: /* extern int errno; */
1.126 brouard 1159:
1.157 brouard 1160: /* #ifdef LINUX */
1161: /* #include <time.h> */
1162: /* #include "timeval.h" */
1163: /* #else */
1164: /* #include <sys/time.h> */
1165: /* #endif */
1166:
1.126 brouard 1167: #include <time.h>
1168:
1.136 brouard 1169: #ifdef GSL
1170: #include <gsl/gsl_errno.h>
1171: #include <gsl/gsl_multimin.h>
1172: #endif
1173:
1.167 brouard 1174:
1.162 brouard 1175: #ifdef NLOPT
1176: #include <nlopt.h>
1177: typedef struct {
1178: double (* function)(double [] );
1179: } myfunc_data ;
1180: #endif
1181:
1.126 brouard 1182: /* #include <libintl.h> */
1183: /* #define _(String) gettext (String) */
1184:
1.251 brouard 1185: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1186:
1187: #define GNUPLOTPROGRAM "gnuplot"
1188: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1189: #define FILENAMELENGTH 132
1190:
1191: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1192: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1193:
1.144 brouard 1194: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1195: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1196:
1197: #define NINTERVMAX 8
1.144 brouard 1198: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1199: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1200: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1201: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1202: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1203: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1204: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1205: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1206: /* #define AGESUP 130 */
1.288 brouard 1207: /* #define AGESUP 150 */
1208: #define AGESUP 200
1.268 brouard 1209: #define AGEINF 0
1.218 brouard 1210: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1211: #define AGEBASE 40
1.194 brouard 1212: #define AGEOVERFLOW 1.e20
1.164 brouard 1213: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1214: #ifdef _WIN32
1215: #define DIRSEPARATOR '\\'
1216: #define CHARSEPARATOR "\\"
1217: #define ODIRSEPARATOR '/'
1218: #else
1.126 brouard 1219: #define DIRSEPARATOR '/'
1220: #define CHARSEPARATOR "/"
1221: #define ODIRSEPARATOR '\\'
1222: #endif
1223:
1.327 ! brouard 1224: /* $Id: imach.c,v 1.326 2022/07/26 17:33:55 brouard Exp $ */
1.126 brouard 1225: /* $State: Exp $ */
1.196 brouard 1226: #include "version.h"
1227: char version[]=__IMACH_VERSION__;
1.323 brouard 1228: char copyright[]="July 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";
1.327 ! brouard 1229: char fullversion[]="$Revision: 1.326 $ $Date: 2022/07/26 17:33:55 $";
1.126 brouard 1230: char strstart[80];
1231: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1232: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1233: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1234: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1235: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1236: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1237: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1238: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1239: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1240: int cptcovprodnoage=0; /**< Number of covariate products without age */
1241: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1242: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1243: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1244: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1245: int nsd=0; /**< Total number of single dummy variables (output) */
1246: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1247: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1248: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1249: int ntveff=0; /**< ntveff number of effective time varying variables */
1250: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1251: int cptcov=0; /* Working variable */
1.290 brouard 1252: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1253: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1254: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1255: int nlstate=2; /* Number of live states */
1256: int ndeath=1; /* Number of dead states */
1.130 brouard 1257: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1258: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1259: int popbased=0;
1260:
1261: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1262: int maxwav=0; /* Maxim number of waves */
1263: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1264: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1265: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1266: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1267: int mle=1, weightopt=0;
1.126 brouard 1268: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1269: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1270: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1271: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1272: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1273: int selected(int kvar); /* Is covariate kvar selected for printing results */
1274:
1.130 brouard 1275: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1276: double **matprod2(); /* test */
1.126 brouard 1277: double **oldm, **newm, **savm; /* Working pointers to matrices */
1278: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1279: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1280:
1.136 brouard 1281: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1282: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1283: FILE *ficlog, *ficrespow;
1.130 brouard 1284: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1285: double fretone; /* Only one call to likelihood */
1.130 brouard 1286: long ipmx=0; /* Number of contributions */
1.126 brouard 1287: double sw; /* Sum of weights */
1288: char filerespow[FILENAMELENGTH];
1289: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1290: FILE *ficresilk;
1291: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1292: FILE *ficresprobmorprev;
1293: FILE *fichtm, *fichtmcov; /* Html File */
1294: FILE *ficreseij;
1295: char filerese[FILENAMELENGTH];
1296: FILE *ficresstdeij;
1297: char fileresstde[FILENAMELENGTH];
1298: FILE *ficrescveij;
1299: char filerescve[FILENAMELENGTH];
1300: FILE *ficresvij;
1301: char fileresv[FILENAMELENGTH];
1.269 brouard 1302:
1.126 brouard 1303: char title[MAXLINE];
1.234 brouard 1304: char model[MAXLINE]; /**< The model line */
1.217 brouard 1305: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1306: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1307: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1308: char command[FILENAMELENGTH];
1309: int outcmd=0;
1310:
1.217 brouard 1311: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1312: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1313: char filelog[FILENAMELENGTH]; /* Log file */
1314: char filerest[FILENAMELENGTH];
1315: char fileregp[FILENAMELENGTH];
1316: char popfile[FILENAMELENGTH];
1317:
1318: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1319:
1.157 brouard 1320: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1321: /* struct timezone tzp; */
1322: /* extern int gettimeofday(); */
1323: struct tm tml, *gmtime(), *localtime();
1324:
1325: extern time_t time();
1326:
1327: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1328: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1329: struct tm tm;
1330:
1.126 brouard 1331: char strcurr[80], strfor[80];
1332:
1333: char *endptr;
1334: long lval;
1335: double dval;
1336:
1337: #define NR_END 1
1338: #define FREE_ARG char*
1339: #define FTOL 1.0e-10
1340:
1341: #define NRANSI
1.240 brouard 1342: #define ITMAX 200
1343: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1344:
1345: #define TOL 2.0e-4
1346:
1347: #define CGOLD 0.3819660
1348: #define ZEPS 1.0e-10
1349: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1350:
1351: #define GOLD 1.618034
1352: #define GLIMIT 100.0
1353: #define TINY 1.0e-20
1354:
1355: static double maxarg1,maxarg2;
1356: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1357: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1358:
1359: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1360: #define rint(a) floor(a+0.5)
1.166 brouard 1361: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1362: #define mytinydouble 1.0e-16
1.166 brouard 1363: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1364: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1365: /* static double dsqrarg; */
1366: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1367: static double sqrarg;
1368: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1369: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1370: int agegomp= AGEGOMP;
1371:
1372: int imx;
1373: int stepm=1;
1374: /* Stepm, step in month: minimum step interpolation*/
1375:
1376: int estepm;
1377: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1378:
1379: int m,nb;
1380: long *num;
1.197 brouard 1381: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1382: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1383: covariate for which somebody answered excluding
1384: undefined. Usually 2: 0 and 1. */
1385: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1386: covariate for which somebody answered including
1387: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1388: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1389: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1390: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1391: double *ageexmed,*agecens;
1392: double dateintmean=0;
1.296 brouard 1393: double anprojd, mprojd, jprojd; /* For eventual projections */
1394: double anprojf, mprojf, jprojf;
1.126 brouard 1395:
1.296 brouard 1396: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1397: double anbackf, mbackf, jbackf;
1398: double jintmean,mintmean,aintmean;
1.126 brouard 1399: double *weight;
1400: int **s; /* Status */
1.141 brouard 1401: double *agedc;
1.145 brouard 1402: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1403: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1404: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1405: double **coqvar; /* Fixed quantitative covariate nqv */
1406: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1407: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1408: double idx;
1409: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1410: /* Some documentation */
1411: /* Design original data
1412: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1413: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1414: * ntv=3 nqtv=1
1415: * cptcovn number of covariates (not including constant and age) = # of + plus 1 = 10+1=11
1416: * For time varying covariate, quanti or dummies
1417: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1418: * cotvar[wav][ntv+iv][i]= [3+(1 to nqtv)][i]=(V12) quanti
1419: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1420: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1421: * covar[k,i], value of kth fixed covariate dummy or quanti :
1422: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1423: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1424: * k= 1 2 3 4 5 6 7 8 9 10 11
1425: */
1426: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1427: /* 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
1428: # States 1=Coresidence, 2 Living alone, 3 Institution
1429: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1430: */
1.319 brouard 1431: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1432: /* k 1 2 3 4 5 6 7 8 9 */
1433: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1434: /* fixed or varying), 1 for age product, 2 for*/
1435: /* product */
1436: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1437: /*(single or product without age), 2 dummy*/
1438: /* with age product, 3 quant with age product*/
1439: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1440: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1441: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1442: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1443: /* nsq 1 2 */ /* Counting single quantit tv */
1444: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1445: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1446: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1447: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1448: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1449: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1450: /* 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 1451: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1452: /* Type */
1453: /* V 1 2 3 4 5 */
1454: /* F F V V V */
1455: /* D Q D D Q */
1456: /* */
1457: int *TvarsD;
1458: int *TvarsDind;
1459: int *TvarsQ;
1460: int *TvarsQind;
1461:
1.318 brouard 1462: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1463: int nresult=0;
1.258 brouard 1464: int parameterline=0; /* # of the parameter (type) line */
1.318 brouard 1465: int TKresult[MAXRESULTLINESPONE];
1466: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable , value (output) */
1467: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable , value (output) */
1468: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For dummy variable , variable # (output) */
1469: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1470: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1471: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , variable # (output) */
1472:
1473: /* 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
1474: # States 1=Coresidence, 2 Living alone, 3 Institution
1475: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1476: */
1.234 brouard 1477: /* 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 1478: 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 */
1479: 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 */
1480: 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 */
1481: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1482: 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 */
1483: 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 1484: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1485: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1486: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1487: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1488: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1489: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1490: 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 */
1491: 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 */
1492:
1.230 brouard 1493: int *Tvarsel; /**< Selected covariates for output */
1494: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1495: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1496: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1497: 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 1498: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1499: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1500: int *Tage;
1.227 brouard 1501: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1502: 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 1503: 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*/
1504: 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 1505: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1506: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1507: int **Tvard;
1508: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1509: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1510: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1511: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1512: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1513: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1514: double *lsurv, *lpop, *tpop;
1515:
1.231 brouard 1516: #define FD 1; /* Fixed dummy covariate */
1517: #define FQ 2; /* Fixed quantitative covariate */
1518: #define FP 3; /* Fixed product covariate */
1519: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1520: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1521: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1522: #define VD 10; /* Varying dummy covariate */
1523: #define VQ 11; /* Varying quantitative covariate */
1524: #define VP 12; /* Varying product covariate */
1525: #define VPDD 13; /* Varying product dummy*dummy covariate */
1526: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1527: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1528: #define APFD 16; /* Age product * fixed dummy covariate */
1529: #define APFQ 17; /* Age product * fixed quantitative covariate */
1530: #define APVD 18; /* Age product * varying dummy covariate */
1531: #define APVQ 19; /* Age product * varying quantitative covariate */
1532:
1533: #define FTYPE 1; /* Fixed covariate */
1534: #define VTYPE 2; /* Varying covariate (loop in wave) */
1535: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1536:
1537: struct kmodel{
1538: int maintype; /* main type */
1539: int subtype; /* subtype */
1540: };
1541: struct kmodel modell[NCOVMAX];
1542:
1.143 brouard 1543: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1544: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1545:
1546: /**************** split *************************/
1547: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1548: {
1549: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1550: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1551: */
1552: char *ss; /* pointer */
1.186 brouard 1553: int l1=0, l2=0; /* length counters */
1.126 brouard 1554:
1555: l1 = strlen(path ); /* length of path */
1556: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1557: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1558: if ( ss == NULL ) { /* no directory, so determine current directory */
1559: strcpy( name, path ); /* we got the fullname name because no directory */
1560: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1561: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1562: /* get current working directory */
1563: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1564: #ifdef WIN32
1565: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1566: #else
1567: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1568: #endif
1.126 brouard 1569: return( GLOCK_ERROR_GETCWD );
1570: }
1571: /* got dirc from getcwd*/
1572: printf(" DIRC = %s \n",dirc);
1.205 brouard 1573: } else { /* strip directory from path */
1.126 brouard 1574: ss++; /* after this, the filename */
1575: l2 = strlen( ss ); /* length of filename */
1576: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1577: strcpy( name, ss ); /* save file name */
1578: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1579: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1580: printf(" DIRC2 = %s \n",dirc);
1581: }
1582: /* We add a separator at the end of dirc if not exists */
1583: l1 = strlen( dirc ); /* length of directory */
1584: if( dirc[l1-1] != DIRSEPARATOR ){
1585: dirc[l1] = DIRSEPARATOR;
1586: dirc[l1+1] = 0;
1587: printf(" DIRC3 = %s \n",dirc);
1588: }
1589: ss = strrchr( name, '.' ); /* find last / */
1590: if (ss >0){
1591: ss++;
1592: strcpy(ext,ss); /* save extension */
1593: l1= strlen( name);
1594: l2= strlen(ss)+1;
1595: strncpy( finame, name, l1-l2);
1596: finame[l1-l2]= 0;
1597: }
1598:
1599: return( 0 ); /* we're done */
1600: }
1601:
1602:
1603: /******************************************/
1604:
1605: void replace_back_to_slash(char *s, char*t)
1606: {
1607: int i;
1608: int lg=0;
1609: i=0;
1610: lg=strlen(t);
1611: for(i=0; i<= lg; i++) {
1612: (s[i] = t[i]);
1613: if (t[i]== '\\') s[i]='/';
1614: }
1615: }
1616:
1.132 brouard 1617: char *trimbb(char *out, char *in)
1.137 brouard 1618: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1619: char *s;
1620: s=out;
1621: while (*in != '\0'){
1.137 brouard 1622: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1623: in++;
1624: }
1625: *out++ = *in++;
1626: }
1627: *out='\0';
1628: return s;
1629: }
1630:
1.187 brouard 1631: /* char *substrchaine(char *out, char *in, char *chain) */
1632: /* { */
1633: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1634: /* char *s, *t; */
1635: /* t=in;s=out; */
1636: /* while ((*in != *chain) && (*in != '\0')){ */
1637: /* *out++ = *in++; */
1638: /* } */
1639:
1640: /* /\* *in matches *chain *\/ */
1641: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1642: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1643: /* } */
1644: /* in--; chain--; */
1645: /* while ( (*in != '\0')){ */
1646: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1647: /* *out++ = *in++; */
1648: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1649: /* } */
1650: /* *out='\0'; */
1651: /* out=s; */
1652: /* return out; */
1653: /* } */
1654: char *substrchaine(char *out, char *in, char *chain)
1655: {
1656: /* Substract chain 'chain' from 'in', return and output 'out' */
1657: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1658:
1659: char *strloc;
1660:
1661: strcpy (out, in);
1662: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1663: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1664: if(strloc != NULL){
1665: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1666: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1667: /* strcpy (strloc, strloc +strlen(chain));*/
1668: }
1669: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1670: return out;
1671: }
1672:
1673:
1.145 brouard 1674: char *cutl(char *blocc, char *alocc, char *in, char occ)
1675: {
1.187 brouard 1676: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1677: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1678: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1679: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1680: */
1.160 brouard 1681: char *s, *t;
1.145 brouard 1682: t=in;s=in;
1683: while ((*in != occ) && (*in != '\0')){
1684: *alocc++ = *in++;
1685: }
1686: if( *in == occ){
1687: *(alocc)='\0';
1688: s=++in;
1689: }
1690:
1691: if (s == t) {/* occ not found */
1692: *(alocc-(in-s))='\0';
1693: in=s;
1694: }
1695: while ( *in != '\0'){
1696: *blocc++ = *in++;
1697: }
1698:
1699: *blocc='\0';
1700: return t;
1701: }
1.137 brouard 1702: char *cutv(char *blocc, char *alocc, char *in, char occ)
1703: {
1.187 brouard 1704: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1705: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1706: gives blocc="abcdef2ghi" and alocc="j".
1707: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1708: */
1709: char *s, *t;
1710: t=in;s=in;
1711: while (*in != '\0'){
1712: while( *in == occ){
1713: *blocc++ = *in++;
1714: s=in;
1715: }
1716: *blocc++ = *in++;
1717: }
1718: if (s == t) /* occ not found */
1719: *(blocc-(in-s))='\0';
1720: else
1721: *(blocc-(in-s)-1)='\0';
1722: in=s;
1723: while ( *in != '\0'){
1724: *alocc++ = *in++;
1725: }
1726:
1727: *alocc='\0';
1728: return s;
1729: }
1730:
1.126 brouard 1731: int nbocc(char *s, char occ)
1732: {
1733: int i,j=0;
1734: int lg=20;
1735: i=0;
1736: lg=strlen(s);
1737: for(i=0; i<= lg; i++) {
1.234 brouard 1738: if (s[i] == occ ) j++;
1.126 brouard 1739: }
1740: return j;
1741: }
1742:
1.137 brouard 1743: /* void cutv(char *u,char *v, char*t, char occ) */
1744: /* { */
1745: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1746: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1747: /* gives u="abcdef2ghi" and v="j" *\/ */
1748: /* int i,lg,j,p=0; */
1749: /* i=0; */
1750: /* lg=strlen(t); */
1751: /* for(j=0; j<=lg-1; j++) { */
1752: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1753: /* } */
1.126 brouard 1754:
1.137 brouard 1755: /* for(j=0; j<p; j++) { */
1756: /* (u[j] = t[j]); */
1757: /* } */
1758: /* u[p]='\0'; */
1.126 brouard 1759:
1.137 brouard 1760: /* for(j=0; j<= lg; j++) { */
1761: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1762: /* } */
1763: /* } */
1.126 brouard 1764:
1.160 brouard 1765: #ifdef _WIN32
1766: char * strsep(char **pp, const char *delim)
1767: {
1768: char *p, *q;
1769:
1770: if ((p = *pp) == NULL)
1771: return 0;
1772: if ((q = strpbrk (p, delim)) != NULL)
1773: {
1774: *pp = q + 1;
1775: *q = '\0';
1776: }
1777: else
1778: *pp = 0;
1779: return p;
1780: }
1781: #endif
1782:
1.126 brouard 1783: /********************** nrerror ********************/
1784:
1785: void nrerror(char error_text[])
1786: {
1787: fprintf(stderr,"ERREUR ...\n");
1788: fprintf(stderr,"%s\n",error_text);
1789: exit(EXIT_FAILURE);
1790: }
1791: /*********************** vector *******************/
1792: double *vector(int nl, int nh)
1793: {
1794: double *v;
1795: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1796: if (!v) nrerror("allocation failure in vector");
1797: return v-nl+NR_END;
1798: }
1799:
1800: /************************ free vector ******************/
1801: void free_vector(double*v, int nl, int nh)
1802: {
1803: free((FREE_ARG)(v+nl-NR_END));
1804: }
1805:
1806: /************************ivector *******************************/
1807: int *ivector(long nl,long nh)
1808: {
1809: int *v;
1810: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1811: if (!v) nrerror("allocation failure in ivector");
1812: return v-nl+NR_END;
1813: }
1814:
1815: /******************free ivector **************************/
1816: void free_ivector(int *v, long nl, long nh)
1817: {
1818: free((FREE_ARG)(v+nl-NR_END));
1819: }
1820:
1821: /************************lvector *******************************/
1822: long *lvector(long nl,long nh)
1823: {
1824: long *v;
1825: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1826: if (!v) nrerror("allocation failure in ivector");
1827: return v-nl+NR_END;
1828: }
1829:
1830: /******************free lvector **************************/
1831: void free_lvector(long *v, long nl, long nh)
1832: {
1833: free((FREE_ARG)(v+nl-NR_END));
1834: }
1835:
1836: /******************* imatrix *******************************/
1837: int **imatrix(long nrl, long nrh, long ncl, long nch)
1838: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1839: {
1840: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1841: int **m;
1842:
1843: /* allocate pointers to rows */
1844: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1845: if (!m) nrerror("allocation failure 1 in matrix()");
1846: m += NR_END;
1847: m -= nrl;
1848:
1849:
1850: /* allocate rows and set pointers to them */
1851: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1852: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1853: m[nrl] += NR_END;
1854: m[nrl] -= ncl;
1855:
1856: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1857:
1858: /* return pointer to array of pointers to rows */
1859: return m;
1860: }
1861:
1862: /****************** free_imatrix *************************/
1863: void free_imatrix(m,nrl,nrh,ncl,nch)
1864: int **m;
1865: long nch,ncl,nrh,nrl;
1866: /* free an int matrix allocated by imatrix() */
1867: {
1868: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1869: free((FREE_ARG) (m+nrl-NR_END));
1870: }
1871:
1872: /******************* matrix *******************************/
1873: double **matrix(long nrl, long nrh, long ncl, long nch)
1874: {
1875: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1876: double **m;
1877:
1878: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1879: if (!m) nrerror("allocation failure 1 in matrix()");
1880: m += NR_END;
1881: m -= nrl;
1882:
1883: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1884: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1885: m[nrl] += NR_END;
1886: m[nrl] -= ncl;
1887:
1888: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1889: return m;
1.145 brouard 1890: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1891: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1892: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1893: */
1894: }
1895:
1896: /*************************free matrix ************************/
1897: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1898: {
1899: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1900: free((FREE_ARG)(m+nrl-NR_END));
1901: }
1902:
1903: /******************* ma3x *******************************/
1904: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1905: {
1906: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1907: double ***m;
1908:
1909: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1910: if (!m) nrerror("allocation failure 1 in matrix()");
1911: m += NR_END;
1912: m -= nrl;
1913:
1914: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1915: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1916: m[nrl] += NR_END;
1917: m[nrl] -= ncl;
1918:
1919: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1920:
1921: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1922: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1923: m[nrl][ncl] += NR_END;
1924: m[nrl][ncl] -= nll;
1925: for (j=ncl+1; j<=nch; j++)
1926: m[nrl][j]=m[nrl][j-1]+nlay;
1927:
1928: for (i=nrl+1; i<=nrh; i++) {
1929: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1930: for (j=ncl+1; j<=nch; j++)
1931: m[i][j]=m[i][j-1]+nlay;
1932: }
1933: return m;
1934: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1935: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1936: */
1937: }
1938:
1939: /*************************free ma3x ************************/
1940: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1941: {
1942: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1943: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1944: free((FREE_ARG)(m+nrl-NR_END));
1945: }
1946:
1947: /*************** function subdirf ***********/
1948: char *subdirf(char fileres[])
1949: {
1950: /* Caution optionfilefiname is hidden */
1951: strcpy(tmpout,optionfilefiname);
1952: strcat(tmpout,"/"); /* Add to the right */
1953: strcat(tmpout,fileres);
1954: return tmpout;
1955: }
1956:
1957: /*************** function subdirf2 ***********/
1958: char *subdirf2(char fileres[], char *preop)
1959: {
1.314 brouard 1960: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
1961: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 1962: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 1963: /* Caution optionfilefiname is hidden */
1964: strcpy(tmpout,optionfilefiname);
1965: strcat(tmpout,"/");
1966: strcat(tmpout,preop);
1967: strcat(tmpout,fileres);
1968: return tmpout;
1969: }
1970:
1971: /*************** function subdirf3 ***********/
1972: char *subdirf3(char fileres[], char *preop, char *preop2)
1973: {
1974:
1975: /* Caution optionfilefiname is hidden */
1976: strcpy(tmpout,optionfilefiname);
1977: strcat(tmpout,"/");
1978: strcat(tmpout,preop);
1979: strcat(tmpout,preop2);
1980: strcat(tmpout,fileres);
1981: return tmpout;
1982: }
1.213 brouard 1983:
1984: /*************** function subdirfext ***********/
1985: char *subdirfext(char fileres[], char *preop, char *postop)
1986: {
1987:
1988: strcpy(tmpout,preop);
1989: strcat(tmpout,fileres);
1990: strcat(tmpout,postop);
1991: return tmpout;
1992: }
1.126 brouard 1993:
1.213 brouard 1994: /*************** function subdirfext3 ***********/
1995: char *subdirfext3(char fileres[], char *preop, char *postop)
1996: {
1997:
1998: /* Caution optionfilefiname is hidden */
1999: strcpy(tmpout,optionfilefiname);
2000: strcat(tmpout,"/");
2001: strcat(tmpout,preop);
2002: strcat(tmpout,fileres);
2003: strcat(tmpout,postop);
2004: return tmpout;
2005: }
2006:
1.162 brouard 2007: char *asc_diff_time(long time_sec, char ascdiff[])
2008: {
2009: long sec_left, days, hours, minutes;
2010: days = (time_sec) / (60*60*24);
2011: sec_left = (time_sec) % (60*60*24);
2012: hours = (sec_left) / (60*60) ;
2013: sec_left = (sec_left) %(60*60);
2014: minutes = (sec_left) /60;
2015: sec_left = (sec_left) % (60);
2016: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2017: return ascdiff;
2018: }
2019:
1.126 brouard 2020: /***************** f1dim *************************/
2021: extern int ncom;
2022: extern double *pcom,*xicom;
2023: extern double (*nrfunc)(double []);
2024:
2025: double f1dim(double x)
2026: {
2027: int j;
2028: double f;
2029: double *xt;
2030:
2031: xt=vector(1,ncom);
2032: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2033: f=(*nrfunc)(xt);
2034: free_vector(xt,1,ncom);
2035: return f;
2036: }
2037:
2038: /*****************brent *************************/
2039: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2040: {
2041: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2042: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2043: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2044: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2045: * returned function value.
2046: */
1.126 brouard 2047: int iter;
2048: double a,b,d,etemp;
1.159 brouard 2049: double fu=0,fv,fw,fx;
1.164 brouard 2050: double ftemp=0.;
1.126 brouard 2051: double p,q,r,tol1,tol2,u,v,w,x,xm;
2052: double e=0.0;
2053:
2054: a=(ax < cx ? ax : cx);
2055: b=(ax > cx ? ax : cx);
2056: x=w=v=bx;
2057: fw=fv=fx=(*f)(x);
2058: for (iter=1;iter<=ITMAX;iter++) {
2059: xm=0.5*(a+b);
2060: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2061: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2062: printf(".");fflush(stdout);
2063: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2064: #ifdef DEBUGBRENT
1.126 brouard 2065: 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);
2066: 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);
2067: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2068: #endif
2069: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2070: *xmin=x;
2071: return fx;
2072: }
2073: ftemp=fu;
2074: if (fabs(e) > tol1) {
2075: r=(x-w)*(fx-fv);
2076: q=(x-v)*(fx-fw);
2077: p=(x-v)*q-(x-w)*r;
2078: q=2.0*(q-r);
2079: if (q > 0.0) p = -p;
2080: q=fabs(q);
2081: etemp=e;
2082: e=d;
2083: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2084: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2085: else {
1.224 brouard 2086: d=p/q;
2087: u=x+d;
2088: if (u-a < tol2 || b-u < tol2)
2089: d=SIGN(tol1,xm-x);
1.126 brouard 2090: }
2091: } else {
2092: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2093: }
2094: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2095: fu=(*f)(u);
2096: if (fu <= fx) {
2097: if (u >= x) a=x; else b=x;
2098: SHFT(v,w,x,u)
1.183 brouard 2099: SHFT(fv,fw,fx,fu)
2100: } else {
2101: if (u < x) a=u; else b=u;
2102: if (fu <= fw || w == x) {
1.224 brouard 2103: v=w;
2104: w=u;
2105: fv=fw;
2106: fw=fu;
1.183 brouard 2107: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2108: v=u;
2109: fv=fu;
1.183 brouard 2110: }
2111: }
1.126 brouard 2112: }
2113: nrerror("Too many iterations in brent");
2114: *xmin=x;
2115: return fx;
2116: }
2117:
2118: /****************** mnbrak ***********************/
2119:
2120: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2121: double (*func)(double))
1.183 brouard 2122: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2123: the downhill direction (defined by the function as evaluated at the initial points) and returns
2124: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2125: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2126: */
1.126 brouard 2127: double ulim,u,r,q, dum;
2128: double fu;
1.187 brouard 2129:
2130: double scale=10.;
2131: int iterscale=0;
2132:
2133: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2134: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2135:
2136:
2137: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2138: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2139: /* *bx = *ax - (*ax - *bx)/scale; */
2140: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2141: /* } */
2142:
1.126 brouard 2143: if (*fb > *fa) {
2144: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2145: SHFT(dum,*fb,*fa,dum)
2146: }
1.126 brouard 2147: *cx=(*bx)+GOLD*(*bx-*ax);
2148: *fc=(*func)(*cx);
1.183 brouard 2149: #ifdef DEBUG
1.224 brouard 2150: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2151: 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 2152: #endif
1.224 brouard 2153: 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 2154: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2155: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2156: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2157: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2158: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2159: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2160: fu=(*func)(u);
1.163 brouard 2161: #ifdef DEBUG
2162: /* f(x)=A(x-u)**2+f(u) */
2163: double A, fparabu;
2164: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2165: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2166: 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);
2167: 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 2168: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2169: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2170: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2171: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2172: #endif
1.184 brouard 2173: #ifdef MNBRAKORIGINAL
1.183 brouard 2174: #else
1.191 brouard 2175: /* if (fu > *fc) { */
2176: /* #ifdef DEBUG */
2177: /* printf("mnbrak4 fu > fc \n"); */
2178: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2179: /* #endif */
2180: /* /\* 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 *\\/ *\/ */
2181: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2182: /* dum=u; /\* Shifting c and u *\/ */
2183: /* u = *cx; */
2184: /* *cx = dum; */
2185: /* dum = fu; */
2186: /* fu = *fc; */
2187: /* *fc =dum; */
2188: /* } else { /\* end *\/ */
2189: /* #ifdef DEBUG */
2190: /* printf("mnbrak3 fu < fc \n"); */
2191: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2192: /* #endif */
2193: /* dum=u; /\* Shifting c and u *\/ */
2194: /* u = *cx; */
2195: /* *cx = dum; */
2196: /* dum = fu; */
2197: /* fu = *fc; */
2198: /* *fc =dum; */
2199: /* } */
1.224 brouard 2200: #ifdef DEBUGMNBRAK
2201: double A, fparabu;
2202: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2203: fparabu= *fa - A*(*ax-u)*(*ax-u);
2204: 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);
2205: 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 2206: #endif
1.191 brouard 2207: dum=u; /* Shifting c and u */
2208: u = *cx;
2209: *cx = dum;
2210: dum = fu;
2211: fu = *fc;
2212: *fc =dum;
1.183 brouard 2213: #endif
1.162 brouard 2214: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2215: #ifdef DEBUG
1.224 brouard 2216: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2217: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2218: #endif
1.126 brouard 2219: fu=(*func)(u);
2220: if (fu < *fc) {
1.183 brouard 2221: #ifdef DEBUG
1.224 brouard 2222: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2223: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2224: #endif
2225: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2226: SHFT(*fb,*fc,fu,(*func)(u))
2227: #ifdef DEBUG
2228: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2229: #endif
2230: }
1.162 brouard 2231: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2232: #ifdef DEBUG
1.224 brouard 2233: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2234: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2235: #endif
1.126 brouard 2236: u=ulim;
2237: fu=(*func)(u);
1.183 brouard 2238: } else { /* u could be left to b (if r > q parabola has a maximum) */
2239: #ifdef DEBUG
1.224 brouard 2240: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2241: 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 2242: #endif
1.126 brouard 2243: u=(*cx)+GOLD*(*cx-*bx);
2244: fu=(*func)(u);
1.224 brouard 2245: #ifdef DEBUG
2246: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2247: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2248: #endif
1.183 brouard 2249: } /* end tests */
1.126 brouard 2250: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2251: SHFT(*fa,*fb,*fc,fu)
2252: #ifdef DEBUG
1.224 brouard 2253: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2254: 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 2255: #endif
2256: } /* 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 2257: }
2258:
2259: /*************** linmin ************************/
1.162 brouard 2260: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2261: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2262: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2263: the value of func at the returned location p . This is actually all accomplished by calling the
2264: routines mnbrak and brent .*/
1.126 brouard 2265: int ncom;
2266: double *pcom,*xicom;
2267: double (*nrfunc)(double []);
2268:
1.224 brouard 2269: #ifdef LINMINORIGINAL
1.126 brouard 2270: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2271: #else
2272: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2273: #endif
1.126 brouard 2274: {
2275: double brent(double ax, double bx, double cx,
2276: double (*f)(double), double tol, double *xmin);
2277: double f1dim(double x);
2278: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2279: double *fc, double (*func)(double));
2280: int j;
2281: double xx,xmin,bx,ax;
2282: double fx,fb,fa;
1.187 brouard 2283:
1.203 brouard 2284: #ifdef LINMINORIGINAL
2285: #else
2286: double scale=10., axs, xxs; /* Scale added for infinity */
2287: #endif
2288:
1.126 brouard 2289: ncom=n;
2290: pcom=vector(1,n);
2291: xicom=vector(1,n);
2292: nrfunc=func;
2293: for (j=1;j<=n;j++) {
2294: pcom[j]=p[j];
1.202 brouard 2295: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2296: }
1.187 brouard 2297:
1.203 brouard 2298: #ifdef LINMINORIGINAL
2299: xx=1.;
2300: #else
2301: axs=0.0;
2302: xxs=1.;
2303: do{
2304: xx= xxs;
2305: #endif
1.187 brouard 2306: ax=0.;
2307: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2308: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2309: /* 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)) */
2310: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2311: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2312: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2313: /* 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 2314: #ifdef LINMINORIGINAL
2315: #else
2316: if (fx != fx){
1.224 brouard 2317: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2318: printf("|");
2319: fprintf(ficlog,"|");
1.203 brouard 2320: #ifdef DEBUGLINMIN
1.224 brouard 2321: 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 2322: #endif
2323: }
1.224 brouard 2324: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2325: #endif
2326:
1.191 brouard 2327: #ifdef DEBUGLINMIN
2328: 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 2329: 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 2330: #endif
1.224 brouard 2331: #ifdef LINMINORIGINAL
2332: #else
1.317 brouard 2333: if(fb == fx){ /* Flat function in the direction */
2334: xmin=xx;
1.224 brouard 2335: *flat=1;
1.317 brouard 2336: }else{
1.224 brouard 2337: *flat=0;
2338: #endif
2339: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2340: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2341: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2342: /* fmin = f(p[j] + xmin * xi[j]) */
2343: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2344: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2345: #ifdef DEBUG
1.224 brouard 2346: 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);
2347: 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);
2348: #endif
2349: #ifdef LINMINORIGINAL
2350: #else
2351: }
1.126 brouard 2352: #endif
1.191 brouard 2353: #ifdef DEBUGLINMIN
2354: printf("linmin end ");
1.202 brouard 2355: fprintf(ficlog,"linmin end ");
1.191 brouard 2356: #endif
1.126 brouard 2357: for (j=1;j<=n;j++) {
1.203 brouard 2358: #ifdef LINMINORIGINAL
2359: xi[j] *= xmin;
2360: #else
2361: #ifdef DEBUGLINMIN
2362: if(xxs <1.0)
2363: printf(" before xi[%d]=%12.8f", j,xi[j]);
2364: #endif
2365: 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) */
2366: #ifdef DEBUGLINMIN
2367: if(xxs <1.0)
2368: 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 );
2369: #endif
2370: #endif
1.187 brouard 2371: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2372: }
1.191 brouard 2373: #ifdef DEBUGLINMIN
1.203 brouard 2374: printf("\n");
1.191 brouard 2375: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2376: 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 2377: for (j=1;j<=n;j++) {
1.202 brouard 2378: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2379: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2380: if(j % ncovmodel == 0){
1.191 brouard 2381: printf("\n");
1.202 brouard 2382: fprintf(ficlog,"\n");
2383: }
1.191 brouard 2384: }
1.203 brouard 2385: #else
1.191 brouard 2386: #endif
1.126 brouard 2387: free_vector(xicom,1,n);
2388: free_vector(pcom,1,n);
2389: }
2390:
2391:
2392: /*************** powell ************************/
1.162 brouard 2393: /*
1.317 brouard 2394: Minimization of a function func of n variables. Input consists in an initial starting point
2395: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2396: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2397: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2398: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2399: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2400: */
1.224 brouard 2401: #ifdef LINMINORIGINAL
2402: #else
2403: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2404: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2405: #endif
1.126 brouard 2406: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2407: double (*func)(double []))
2408: {
1.224 brouard 2409: #ifdef LINMINORIGINAL
2410: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2411: double (*func)(double []));
1.224 brouard 2412: #else
1.241 brouard 2413: void linmin(double p[], double xi[], int n, double *fret,
2414: double (*func)(double []),int *flat);
1.224 brouard 2415: #endif
1.239 brouard 2416: int i,ibig,j,jk,k;
1.126 brouard 2417: double del,t,*pt,*ptt,*xit;
1.181 brouard 2418: double directest;
1.126 brouard 2419: double fp,fptt;
2420: double *xits;
2421: int niterf, itmp;
2422:
2423: pt=vector(1,n);
2424: ptt=vector(1,n);
2425: xit=vector(1,n);
2426: xits=vector(1,n);
2427: *fret=(*func)(p);
2428: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2429: rcurr_time = time(NULL);
1.126 brouard 2430: for (*iter=1;;++(*iter)) {
2431: ibig=0;
2432: del=0.0;
1.157 brouard 2433: rlast_time=rcurr_time;
2434: /* (void) gettimeofday(&curr_time,&tzp); */
2435: rcurr_time = time(NULL);
2436: curr_time = *localtime(&rcurr_time);
1.324 brouard 2437: printf("\nPowell iter=%d -2*LL=%.12f gain=%.12f=%.3g %ld sec. %ld sec.",*iter,*fret, fp-*fret,fp-*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2438: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f gain=%.12f=%.3g %ld sec. %ld sec.",*iter,*fret, fp-*fret,fp-*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
1.157 brouard 2439: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2440: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2441: for (i=1;i<=n;i++) {
1.126 brouard 2442: fprintf(ficrespow," %.12lf", p[i]);
2443: }
1.239 brouard 2444: fprintf(ficrespow,"\n");fflush(ficrespow);
2445: printf("\n#model= 1 + age ");
2446: fprintf(ficlog,"\n#model= 1 + age ");
2447: if(nagesqr==1){
1.241 brouard 2448: printf(" + age*age ");
2449: fprintf(ficlog," + age*age ");
1.239 brouard 2450: }
2451: for(j=1;j <=ncovmodel-2;j++){
2452: if(Typevar[j]==0) {
2453: printf(" + V%d ",Tvar[j]);
2454: fprintf(ficlog," + V%d ",Tvar[j]);
2455: }else if(Typevar[j]==1) {
2456: printf(" + V%d*age ",Tvar[j]);
2457: fprintf(ficlog," + V%d*age ",Tvar[j]);
2458: }else if(Typevar[j]==2) {
2459: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2460: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2461: }
2462: }
1.126 brouard 2463: printf("\n");
1.239 brouard 2464: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2465: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2466: fprintf(ficlog,"\n");
1.239 brouard 2467: for(i=1,jk=1; i <=nlstate; i++){
2468: for(k=1; k <=(nlstate+ndeath); k++){
2469: if (k != i) {
2470: printf("%d%d ",i,k);
2471: fprintf(ficlog,"%d%d ",i,k);
2472: for(j=1; j <=ncovmodel; j++){
2473: printf("%12.7f ",p[jk]);
2474: fprintf(ficlog,"%12.7f ",p[jk]);
2475: jk++;
2476: }
2477: printf("\n");
2478: fprintf(ficlog,"\n");
2479: }
2480: }
2481: }
1.241 brouard 2482: if(*iter <=3 && *iter >1){
1.157 brouard 2483: tml = *localtime(&rcurr_time);
2484: strcpy(strcurr,asctime(&tml));
2485: rforecast_time=rcurr_time;
1.126 brouard 2486: itmp = strlen(strcurr);
2487: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2488: strcurr[itmp-1]='\0';
1.162 brouard 2489: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2490: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2491: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2492: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2493: forecast_time = *localtime(&rforecast_time);
2494: strcpy(strfor,asctime(&forecast_time));
2495: itmp = strlen(strfor);
2496: if(strfor[itmp-1]=='\n')
2497: strfor[itmp-1]='\0';
2498: 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);
2499: 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 2500: }
2501: }
1.187 brouard 2502: for (i=1;i<=n;i++) { /* For each direction i */
2503: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2504: fptt=(*fret);
2505: #ifdef DEBUG
1.203 brouard 2506: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2507: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2508: #endif
1.203 brouard 2509: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2510: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2511: #ifdef LINMINORIGINAL
1.188 brouard 2512: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2513: #else
2514: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2515: flatdir[i]=flat; /* Function is vanishing in that direction i */
2516: #endif
2517: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2518: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2519: /* because that direction will be replaced unless the gain del is small */
2520: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2521: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2522: /* with the new direction. */
2523: del=fabs(fptt-(*fret));
2524: ibig=i;
1.126 brouard 2525: }
2526: #ifdef DEBUG
2527: printf("%d %.12e",i,(*fret));
2528: fprintf(ficlog,"%d %.12e",i,(*fret));
2529: for (j=1;j<=n;j++) {
1.224 brouard 2530: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2531: printf(" x(%d)=%.12e",j,xit[j]);
2532: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2533: }
2534: for(j=1;j<=n;j++) {
1.225 brouard 2535: printf(" p(%d)=%.12e",j,p[j]);
2536: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2537: }
2538: printf("\n");
2539: fprintf(ficlog,"\n");
2540: #endif
1.187 brouard 2541: } /* end loop on each direction i */
2542: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2543: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2544: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2545: for(j=1;j<=n;j++) {
2546: if(flatdir[j] >0){
2547: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2548: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2549: }
1.319 brouard 2550: /* printf("\n"); */
2551: /* fprintf(ficlog,"\n"); */
2552: }
1.243 brouard 2553: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2554: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2555: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2556: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2557: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2558: /* decreased of more than 3.84 */
2559: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2560: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2561: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2562:
1.188 brouard 2563: /* Starting the program with initial values given by a former maximization will simply change */
2564: /* the scales of the directions and the directions, because the are reset to canonical directions */
2565: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2566: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2567: #ifdef DEBUG
2568: int k[2],l;
2569: k[0]=1;
2570: k[1]=-1;
2571: printf("Max: %.12e",(*func)(p));
2572: fprintf(ficlog,"Max: %.12e",(*func)(p));
2573: for (j=1;j<=n;j++) {
2574: printf(" %.12e",p[j]);
2575: fprintf(ficlog," %.12e",p[j]);
2576: }
2577: printf("\n");
2578: fprintf(ficlog,"\n");
2579: for(l=0;l<=1;l++) {
2580: for (j=1;j<=n;j++) {
2581: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2582: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2583: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2584: }
2585: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2586: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2587: }
2588: #endif
2589:
2590: free_vector(xit,1,n);
2591: free_vector(xits,1,n);
2592: free_vector(ptt,1,n);
2593: free_vector(pt,1,n);
2594: return;
1.192 brouard 2595: } /* enough precision */
1.240 brouard 2596: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2597: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2598: ptt[j]=2.0*p[j]-pt[j];
2599: xit[j]=p[j]-pt[j];
2600: pt[j]=p[j];
2601: }
1.181 brouard 2602: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2603: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2604: if (*iter <=4) {
1.225 brouard 2605: #else
2606: #endif
1.224 brouard 2607: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2608: #else
1.161 brouard 2609: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2610: #endif
1.162 brouard 2611: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2612: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2613: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2614: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2615: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2616: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2617: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2618: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2619: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2620: /* Even if f3 <f1, directest can be negative and t >0 */
2621: /* mu² and del² are equal when f3=f1 */
2622: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2623: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2624: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2625: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2626: #ifdef NRCORIGINAL
2627: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2628: #else
2629: 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 2630: t= t- del*SQR(fp-fptt);
1.183 brouard 2631: #endif
1.202 brouard 2632: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2633: #ifdef DEBUG
1.181 brouard 2634: 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);
2635: 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 2636: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2637: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2638: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2639: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2640: 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);
2641: 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);
2642: #endif
1.183 brouard 2643: #ifdef POWELLORIGINAL
2644: if (t < 0.0) { /* Then we use it for new direction */
2645: #else
1.182 brouard 2646: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2647: 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 2648: 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 2649: 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 2650: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2651: }
1.181 brouard 2652: if (directest < 0.0) { /* Then we use it for new direction */
2653: #endif
1.191 brouard 2654: #ifdef DEBUGLINMIN
1.234 brouard 2655: printf("Before linmin in direction P%d-P0\n",n);
2656: for (j=1;j<=n;j++) {
2657: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2658: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2659: if(j % ncovmodel == 0){
2660: printf("\n");
2661: fprintf(ficlog,"\n");
2662: }
2663: }
1.224 brouard 2664: #endif
2665: #ifdef LINMINORIGINAL
1.234 brouard 2666: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2667: #else
1.234 brouard 2668: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2669: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2670: #endif
1.234 brouard 2671:
1.191 brouard 2672: #ifdef DEBUGLINMIN
1.234 brouard 2673: for (j=1;j<=n;j++) {
2674: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2675: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2676: if(j % ncovmodel == 0){
2677: printf("\n");
2678: fprintf(ficlog,"\n");
2679: }
2680: }
1.224 brouard 2681: #endif
1.234 brouard 2682: for (j=1;j<=n;j++) {
2683: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2684: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2685: }
1.224 brouard 2686: #ifdef LINMINORIGINAL
2687: #else
1.234 brouard 2688: for (j=1, flatd=0;j<=n;j++) {
2689: if(flatdir[j]>0)
2690: flatd++;
2691: }
2692: if(flatd >0){
1.255 brouard 2693: printf("%d flat directions: ",flatd);
2694: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2695: for (j=1;j<=n;j++) {
2696: if(flatdir[j]>0){
2697: printf("%d ",j);
2698: fprintf(ficlog,"%d ",j);
2699: }
2700: }
2701: printf("\n");
2702: fprintf(ficlog,"\n");
1.319 brouard 2703: #ifdef FLATSUP
2704: free_vector(xit,1,n);
2705: free_vector(xits,1,n);
2706: free_vector(ptt,1,n);
2707: free_vector(pt,1,n);
2708: return;
2709: #endif
1.234 brouard 2710: }
1.191 brouard 2711: #endif
1.234 brouard 2712: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2713: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2714:
1.126 brouard 2715: #ifdef DEBUG
1.234 brouard 2716: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2717: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2718: for(j=1;j<=n;j++){
2719: printf(" %lf",xit[j]);
2720: fprintf(ficlog," %lf",xit[j]);
2721: }
2722: printf("\n");
2723: fprintf(ficlog,"\n");
1.126 brouard 2724: #endif
1.192 brouard 2725: } /* end of t or directest negative */
1.224 brouard 2726: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2727: #else
1.234 brouard 2728: } /* end if (fptt < fp) */
1.192 brouard 2729: #endif
1.225 brouard 2730: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2731: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2732: #else
1.224 brouard 2733: #endif
1.234 brouard 2734: } /* loop iteration */
1.126 brouard 2735: }
1.234 brouard 2736:
1.126 brouard 2737: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2738:
1.235 brouard 2739: 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 2740: {
1.279 brouard 2741: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2742: * (and selected quantitative values in nres)
2743: * by left multiplying the unit
2744: * matrix by transitions matrix until convergence is reached with precision ftolpl
2745: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2746: * Wx is row vector: population in state 1, population in state 2, population dead
2747: * or prevalence in state 1, prevalence in state 2, 0
2748: * newm is the matrix after multiplications, its rows are identical at a factor.
2749: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2750: * Output is prlim.
2751: * Initial matrix pimij
2752: */
1.206 brouard 2753: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2754: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2755: /* 0, 0 , 1} */
2756: /*
2757: * and after some iteration: */
2758: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2759: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2760: /* 0, 0 , 1} */
2761: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2762: /* {0.51571254859325999, 0.4842874514067399, */
2763: /* 0.51326036147820708, 0.48673963852179264} */
2764: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2765:
1.126 brouard 2766: int i, ii,j,k;
1.209 brouard 2767: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2768: /* double **matprod2(); */ /* test */
1.218 brouard 2769: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2770: double **newm;
1.209 brouard 2771: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2772: int ncvloop=0;
1.288 brouard 2773: int first=0;
1.169 brouard 2774:
1.209 brouard 2775: min=vector(1,nlstate);
2776: max=vector(1,nlstate);
2777: meandiff=vector(1,nlstate);
2778:
1.218 brouard 2779: /* Starting with matrix unity */
1.126 brouard 2780: for (ii=1;ii<=nlstate+ndeath;ii++)
2781: for (j=1;j<=nlstate+ndeath;j++){
2782: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2783: }
1.169 brouard 2784:
2785: cov[1]=1.;
2786:
2787: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2788: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2789: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2790: ncvloop++;
1.126 brouard 2791: newm=savm;
2792: /* Covariates have to be included here again */
1.138 brouard 2793: cov[2]=agefin;
1.319 brouard 2794: if(nagesqr==1){
2795: cov[3]= agefin*agefin;
2796: }
1.234 brouard 2797: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2798: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2799: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.319 brouard 2800: /* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; */
1.235 brouard 2801: /* 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 2802: }
2803: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2804: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.319 brouard 2805: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2806: /* cov[++k1]=Tqresult[nres][k]; */
1.235 brouard 2807: /* 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 2808: }
1.237 brouard 2809: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.319 brouard 2810: if(Dummy[Tage[k]]==2){ /* dummy with age */
1.234 brouard 2811: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 2812: /* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
2813: } else if(Dummy[Tage[k]]==3){ /* quantitative with age */
2814: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2815: /* cov[++k1]=Tqresult[nres][k]; */
1.234 brouard 2816: }
1.235 brouard 2817: /* 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 2818: }
1.237 brouard 2819: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2820: /* 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 2821: if(Dummy[Tvard[k][1]==0]){
2822: if(Dummy[Tvard[k][2]==0]){
2823: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
1.319 brouard 2824: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.237 brouard 2825: }else{
2826: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
1.319 brouard 2827: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; */
1.237 brouard 2828: }
2829: }else{
2830: if(Dummy[Tvard[k][2]==0]){
2831: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
1.319 brouard 2832: /* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; */
1.237 brouard 2833: }else{
2834: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
1.319 brouard 2835: /* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
1.237 brouard 2836: }
2837: }
1.234 brouard 2838: }
1.138 brouard 2839: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2840: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2841: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2842: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2843: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 2844: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2845: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2846:
1.126 brouard 2847: savm=oldm;
2848: oldm=newm;
1.209 brouard 2849:
2850: for(j=1; j<=nlstate; j++){
2851: max[j]=0.;
2852: min[j]=1.;
2853: }
2854: for(i=1;i<=nlstate;i++){
2855: sumnew=0;
2856: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2857: for(j=1; j<=nlstate; j++){
2858: prlim[i][j]= newm[i][j]/(1-sumnew);
2859: max[j]=FMAX(max[j],prlim[i][j]);
2860: min[j]=FMIN(min[j],prlim[i][j]);
2861: }
2862: }
2863:
1.126 brouard 2864: maxmax=0.;
1.209 brouard 2865: for(j=1; j<=nlstate; j++){
2866: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2867: maxmax=FMAX(maxmax,meandiff[j]);
2868: /* 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 2869: } /* j loop */
1.203 brouard 2870: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2871: /* 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 2872: if(maxmax < ftolpl){
1.209 brouard 2873: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2874: free_vector(min,1,nlstate);
2875: free_vector(max,1,nlstate);
2876: free_vector(meandiff,1,nlstate);
1.126 brouard 2877: return prlim;
2878: }
1.288 brouard 2879: } /* agefin loop */
1.208 brouard 2880: /* After some age loop it doesn't converge */
1.288 brouard 2881: if(!first){
2882: first=1;
2883: 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 2884: 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);
2885: }else if (first >=1 && first <10){
2886: 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);
2887: first++;
2888: }else if (first ==10){
2889: 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);
2890: 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");
2891: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
2892: first++;
1.288 brouard 2893: }
2894:
1.209 brouard 2895: /* 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); */
2896: free_vector(min,1,nlstate);
2897: free_vector(max,1,nlstate);
2898: free_vector(meandiff,1,nlstate);
1.208 brouard 2899:
1.169 brouard 2900: return prlim; /* should not reach here */
1.126 brouard 2901: }
2902:
1.217 brouard 2903:
2904: /**** Back Prevalence limit (stable or period prevalence) ****************/
2905:
1.218 brouard 2906: /* 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) */
2907: /* 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 2908: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2909: {
1.264 brouard 2910: /* 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 2911: matrix by transitions matrix until convergence is reached with precision ftolpl */
2912: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2913: /* Wx is row vector: population in state 1, population in state 2, population dead */
2914: /* or prevalence in state 1, prevalence in state 2, 0 */
2915: /* newm is the matrix after multiplications, its rows are identical at a factor */
2916: /* Initial matrix pimij */
2917: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2918: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2919: /* 0, 0 , 1} */
2920: /*
2921: * and after some iteration: */
2922: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2923: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2924: /* 0, 0 , 1} */
2925: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2926: /* {0.51571254859325999, 0.4842874514067399, */
2927: /* 0.51326036147820708, 0.48673963852179264} */
2928: /* If we start from prlim again, prlim tends to a constant matrix */
2929:
2930: int i, ii,j,k;
1.247 brouard 2931: int first=0;
1.217 brouard 2932: double *min, *max, *meandiff, maxmax,sumnew=0.;
2933: /* double **matprod2(); */ /* test */
2934: double **out, cov[NCOVMAX+1], **bmij();
2935: double **newm;
1.218 brouard 2936: double **dnewm, **doldm, **dsavm; /* for use */
2937: double **oldm, **savm; /* for use */
2938:
1.217 brouard 2939: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2940: int ncvloop=0;
2941:
2942: min=vector(1,nlstate);
2943: max=vector(1,nlstate);
2944: meandiff=vector(1,nlstate);
2945:
1.266 brouard 2946: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2947: oldm=oldms; savm=savms;
2948:
2949: /* Starting with matrix unity */
2950: for (ii=1;ii<=nlstate+ndeath;ii++)
2951: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2952: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2953: }
2954:
2955: cov[1]=1.;
2956:
2957: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2958: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2959: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2960: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2961: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2962: ncvloop++;
1.218 brouard 2963: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2964: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2965: /* Covariates have to be included here again */
2966: cov[2]=agefin;
1.319 brouard 2967: if(nagesqr==1){
1.217 brouard 2968: cov[3]= agefin*agefin;;
1.319 brouard 2969: }
1.242 brouard 2970: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2971: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2972: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2973: /* 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 2974: }
2975: /* for (k=1; k<=cptcovn;k++) { */
2976: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2977: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2978: /* /\* 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])]); *\/ */
2979: /* } */
2980: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2981: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2982: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2983: /* 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]); */
2984: }
2985: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2986: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2987: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2988: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2989: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.319 brouard 2990: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age *\/ ERROR ???*/
2991: if(Dummy[Tage[k]]== 2){ /* dummy with age */
1.242 brouard 2992: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 2993: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
2994: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.242 brouard 2995: }
2996: /* 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]); */
2997: }
2998: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2999: /* 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]); */
3000: if(Dummy[Tvard[k][1]==0]){
3001: if(Dummy[Tvard[k][2]==0]){
3002: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3003: }else{
3004: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
3005: }
3006: }else{
3007: if(Dummy[Tvard[k][2]==0]){
3008: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
3009: }else{
3010: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
3011: }
3012: }
1.217 brouard 3013: }
3014:
3015: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3016: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3017: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3018: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3019: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3020: /* ij should be linked to the correct index of cov */
3021: /* age and covariate values ij are in 'cov', but we need to pass
3022: * ij for the observed prevalence at age and status and covariate
3023: * number: prevacurrent[(int)agefin][ii][ij]
3024: */
3025: /* 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 *\/ */
3026: /* 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 *\/ */
3027: 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 3028: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3029: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3030: /* for(i=1; i<=nlstate+ndeath; i++) { */
3031: /* printf("%d newm= ",i); */
3032: /* for(j=1;j<=nlstate+ndeath;j++) { */
3033: /* printf("%f ",newm[i][j]); */
3034: /* } */
3035: /* printf("oldm * "); */
3036: /* for(j=1;j<=nlstate+ndeath;j++) { */
3037: /* printf("%f ",oldm[i][j]); */
3038: /* } */
1.268 brouard 3039: /* printf(" bmmij "); */
1.266 brouard 3040: /* for(j=1;j<=nlstate+ndeath;j++) { */
3041: /* printf("%f ",pmmij[i][j]); */
3042: /* } */
3043: /* printf("\n"); */
3044: /* } */
3045: /* } */
1.217 brouard 3046: savm=oldm;
3047: oldm=newm;
1.266 brouard 3048:
1.217 brouard 3049: for(j=1; j<=nlstate; j++){
3050: max[j]=0.;
3051: min[j]=1.;
3052: }
3053: for(j=1; j<=nlstate; j++){
3054: for(i=1;i<=nlstate;i++){
1.234 brouard 3055: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3056: bprlim[i][j]= newm[i][j];
3057: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3058: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3059: }
3060: }
1.218 brouard 3061:
1.217 brouard 3062: maxmax=0.;
3063: for(i=1; i<=nlstate; i++){
1.318 brouard 3064: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3065: maxmax=FMAX(maxmax,meandiff[i]);
3066: /* 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 3067: } /* i loop */
1.217 brouard 3068: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3069: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3070: if(maxmax < ftolpl){
1.220 brouard 3071: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3072: free_vector(min,1,nlstate);
3073: free_vector(max,1,nlstate);
3074: free_vector(meandiff,1,nlstate);
3075: return bprlim;
3076: }
1.288 brouard 3077: } /* agefin loop */
1.217 brouard 3078: /* After some age loop it doesn't converge */
1.288 brouard 3079: if(!first){
1.247 brouard 3080: first=1;
3081: 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\
3082: 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);
3083: }
3084: 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 3085: 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);
3086: /* 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); */
3087: free_vector(min,1,nlstate);
3088: free_vector(max,1,nlstate);
3089: free_vector(meandiff,1,nlstate);
3090:
3091: return bprlim; /* should not reach here */
3092: }
3093:
1.126 brouard 3094: /*************** transition probabilities ***************/
3095:
3096: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3097: {
1.138 brouard 3098: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3099: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3100: model to the ncovmodel covariates (including constant and age).
3101: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3102: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3103: ncth covariate in the global vector x is given by the formula:
3104: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3105: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3106: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3107: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3108: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3109: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3110: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3111: */
3112: double s1, lnpijopii;
1.126 brouard 3113: /*double t34;*/
1.164 brouard 3114: int i,j, nc, ii, jj;
1.126 brouard 3115:
1.223 brouard 3116: for(i=1; i<= nlstate; i++){
3117: for(j=1; j<i;j++){
3118: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3119: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3120: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3121: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3122: }
3123: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3124: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3125: }
3126: for(j=i+1; j<=nlstate+ndeath;j++){
3127: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3128: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3129: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3130: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3131: }
3132: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3133: }
3134: }
1.218 brouard 3135:
1.223 brouard 3136: for(i=1; i<= nlstate; i++){
3137: s1=0;
3138: for(j=1; j<i; j++){
3139: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3140: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3141: }
3142: for(j=i+1; j<=nlstate+ndeath; j++){
3143: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3144: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3145: }
3146: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3147: ps[i][i]=1./(s1+1.);
3148: /* Computing other pijs */
3149: for(j=1; j<i; j++)
1.325 brouard 3150: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3151: for(j=i+1; j<=nlstate+ndeath; j++)
3152: ps[i][j]= exp(ps[i][j])*ps[i][i];
3153: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3154: } /* end i */
1.218 brouard 3155:
1.223 brouard 3156: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3157: for(jj=1; jj<= nlstate+ndeath; jj++){
3158: ps[ii][jj]=0;
3159: ps[ii][ii]=1;
3160: }
3161: }
1.294 brouard 3162:
3163:
1.223 brouard 3164: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3165: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3166: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3167: /* } */
3168: /* printf("\n "); */
3169: /* } */
3170: /* printf("\n ");printf("%lf ",cov[2]);*/
3171: /*
3172: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3173: goto end;*/
1.266 brouard 3174: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3175: }
3176:
1.218 brouard 3177: /*************** backward transition probabilities ***************/
3178:
3179: /* 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 ) */
3180: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3181: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3182: {
1.302 brouard 3183: /* 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 3184: * 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 3185: */
1.218 brouard 3186: int i, ii, j,k;
1.222 brouard 3187:
3188: double **out, **pmij();
3189: double sumnew=0.;
1.218 brouard 3190: double agefin;
1.292 brouard 3191: 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 3192: double **dnewm, **dsavm, **doldm;
3193: double **bbmij;
3194:
1.218 brouard 3195: doldm=ddoldms; /* global pointers */
1.222 brouard 3196: dnewm=ddnewms;
3197: dsavm=ddsavms;
1.318 brouard 3198:
3199: /* Debug */
3200: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3201: agefin=cov[2];
1.268 brouard 3202: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3203: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3204: the observed prevalence (with this covariate ij) at beginning of transition */
3205: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3206:
3207: /* P_x */
1.325 brouard 3208: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3209: /* outputs pmmij which is a stochastic matrix in row */
3210:
3211: /* Diag(w_x) */
1.292 brouard 3212: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3213: sumnew=0.;
1.269 brouard 3214: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3215: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3216: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3217: sumnew+=prevacurrent[(int)agefin][ii][ij];
3218: }
3219: if(sumnew >0.01){ /* At least some value in the prevalence */
3220: for (ii=1;ii<=nlstate+ndeath;ii++){
3221: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3222: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3223: }
3224: }else{
3225: for (ii=1;ii<=nlstate+ndeath;ii++){
3226: for (j=1;j<=nlstate+ndeath;j++)
3227: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3228: }
3229: /* if(sumnew <0.9){ */
3230: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3231: /* } */
3232: }
3233: k3=0.0; /* We put the last diagonal to 0 */
3234: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3235: doldm[ii][ii]= k3;
3236: }
3237: /* End doldm, At the end doldm is diag[(w_i)] */
3238:
1.292 brouard 3239: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3240: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3241:
1.292 brouard 3242: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3243: /* 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 3244: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3245: sumnew=0.;
1.222 brouard 3246: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3247: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3248: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3249: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3250: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3251: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3252: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3253: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3254: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3255: /* }else */
1.268 brouard 3256: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3257: } /*End ii */
3258: } /* 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 */
3259:
1.292 brouard 3260: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3261: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3262: /* end bmij */
1.266 brouard 3263: return ps; /*pointer is unchanged */
1.218 brouard 3264: }
1.217 brouard 3265: /*************** transition probabilities ***************/
3266:
1.218 brouard 3267: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3268: {
3269: /* According to parameters values stored in x and the covariate's values stored in cov,
3270: computes the probability to be observed in state j being in state i by appying the
3271: model to the ncovmodel covariates (including constant and age).
3272: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3273: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3274: ncth covariate in the global vector x is given by the formula:
3275: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3276: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3277: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3278: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3279: Outputs ps[i][j] the probability to be observed in j being in j according to
3280: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3281: */
3282: double s1, lnpijopii;
3283: /*double t34;*/
3284: int i,j, nc, ii, jj;
3285:
1.234 brouard 3286: for(i=1; i<= nlstate; i++){
3287: for(j=1; j<i;j++){
3288: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3289: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3290: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3291: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3292: }
3293: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3294: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3295: }
3296: for(j=i+1; j<=nlstate+ndeath;j++){
3297: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3298: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3299: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3300: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3301: }
3302: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3303: }
3304: }
3305:
3306: for(i=1; i<= nlstate; i++){
3307: s1=0;
3308: for(j=1; j<i; j++){
3309: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3310: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3311: }
3312: for(j=i+1; j<=nlstate+ndeath; j++){
3313: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3314: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3315: }
3316: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3317: ps[i][i]=1./(s1+1.);
3318: /* Computing other pijs */
3319: for(j=1; j<i; j++)
3320: ps[i][j]= exp(ps[i][j])*ps[i][i];
3321: for(j=i+1; j<=nlstate+ndeath; j++)
3322: ps[i][j]= exp(ps[i][j])*ps[i][i];
3323: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3324: } /* end i */
3325:
3326: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3327: for(jj=1; jj<= nlstate+ndeath; jj++){
3328: ps[ii][jj]=0;
3329: ps[ii][ii]=1;
3330: }
3331: }
1.296 brouard 3332: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3333: for(jj=1; jj<= nlstate+ndeath; jj++){
3334: s1=0.;
3335: for(ii=1; ii<= nlstate+ndeath; ii++){
3336: s1+=ps[ii][jj];
3337: }
3338: for(ii=1; ii<= nlstate; ii++){
3339: ps[ii][jj]=ps[ii][jj]/s1;
3340: }
3341: }
3342: /* Transposition */
3343: for(jj=1; jj<= nlstate+ndeath; jj++){
3344: for(ii=jj; ii<= nlstate+ndeath; ii++){
3345: s1=ps[ii][jj];
3346: ps[ii][jj]=ps[jj][ii];
3347: ps[jj][ii]=s1;
3348: }
3349: }
3350: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3351: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3352: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3353: /* } */
3354: /* printf("\n "); */
3355: /* } */
3356: /* printf("\n ");printf("%lf ",cov[2]);*/
3357: /*
3358: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3359: goto end;*/
3360: return ps;
1.217 brouard 3361: }
3362:
3363:
1.126 brouard 3364: /**************** Product of 2 matrices ******************/
3365:
1.145 brouard 3366: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3367: {
3368: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3369: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3370: /* in, b, out are matrice of pointers which should have been initialized
3371: before: only the contents of out is modified. The function returns
3372: a pointer to pointers identical to out */
1.145 brouard 3373: int i, j, k;
1.126 brouard 3374: for(i=nrl; i<= nrh; i++)
1.145 brouard 3375: for(k=ncolol; k<=ncoloh; k++){
3376: out[i][k]=0.;
3377: for(j=ncl; j<=nch; j++)
3378: out[i][k] +=in[i][j]*b[j][k];
3379: }
1.126 brouard 3380: return out;
3381: }
3382:
3383:
3384: /************* Higher Matrix Product ***************/
3385:
1.235 brouard 3386: 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 3387: {
1.218 brouard 3388: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3389: 'nhstepm*hstepm*stepm' months (i.e. until
3390: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3391: nhstepm*hstepm matrices.
3392: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3393: (typically every 2 years instead of every month which is too big
3394: for the memory).
3395: Model is determined by parameters x and covariates have to be
3396: included manually here.
3397:
3398: */
3399:
3400: int i, j, d, h, k;
1.131 brouard 3401: double **out, cov[NCOVMAX+1];
1.126 brouard 3402: double **newm;
1.187 brouard 3403: double agexact;
1.214 brouard 3404: double agebegin, ageend;
1.126 brouard 3405:
3406: /* Hstepm could be zero and should return the unit matrix */
3407: for (i=1;i<=nlstate+ndeath;i++)
3408: for (j=1;j<=nlstate+ndeath;j++){
3409: oldm[i][j]=(i==j ? 1.0 : 0.0);
3410: po[i][j][0]=(i==j ? 1.0 : 0.0);
3411: }
3412: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3413: for(h=1; h <=nhstepm; h++){
3414: for(d=1; d <=hstepm; d++){
3415: newm=savm;
3416: /* Covariates have to be included here again */
3417: cov[1]=1.;
1.214 brouard 3418: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3419: cov[2]=agexact;
1.319 brouard 3420: if(nagesqr==1){
1.227 brouard 3421: cov[3]= agexact*agexact;
1.319 brouard 3422: }
1.235 brouard 3423: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
1.319 brouard 3424: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3425: /* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 */
3426: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3427: /* k 1 2 3 4 5 6 7 8 9 */
3428: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
3429: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
3430: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
3431: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1.235 brouard 3432: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3433: /* 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)); */
3434: }
3435: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3436: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.319 brouard 3437: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
1.235 brouard 3438: /* 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]); */
3439: }
1.319 brouard 3440: for (k=1; k<=cptcovage;k++){ /* For product with age V1+V1*age +V4 +age*V3 */
3441: /* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*/
3442: /* */
3443: if(Dummy[Tage[k]]== 2){ /* dummy with age */
3444: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age *\/ */
1.235 brouard 3445: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 3446: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
3447: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.235 brouard 3448: }
3449: /* 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]); */
3450: }
1.319 brouard 3451: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 3452: /* 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 3453: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
3454: if(Dummy[Tvard[k][1]==0]){
3455: if(Dummy[Tvard[k][2]==0]){
3456: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3457: }else{
3458: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
3459: }
3460: }else{
3461: if(Dummy[Tvard[k][2]==0]){
3462: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
3463: }else{
3464: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
3465: }
3466: }
1.235 brouard 3467: }
3468: /* for (k=1; k<=cptcovn;k++) */
3469: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3470: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3471: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3472: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3473: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3474:
3475:
1.126 brouard 3476: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3477: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3478: /* right multiplication of oldm by the current matrix */
1.126 brouard 3479: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3480: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3481: /* if((int)age == 70){ */
3482: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3483: /* for(i=1; i<=nlstate+ndeath; i++) { */
3484: /* printf("%d pmmij ",i); */
3485: /* for(j=1;j<=nlstate+ndeath;j++) { */
3486: /* printf("%f ",pmmij[i][j]); */
3487: /* } */
3488: /* printf(" oldm "); */
3489: /* for(j=1;j<=nlstate+ndeath;j++) { */
3490: /* printf("%f ",oldm[i][j]); */
3491: /* } */
3492: /* printf("\n"); */
3493: /* } */
3494: /* } */
1.126 brouard 3495: savm=oldm;
3496: oldm=newm;
3497: }
3498: for(i=1; i<=nlstate+ndeath; i++)
3499: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3500: po[i][j][h]=newm[i][j];
3501: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3502: }
1.128 brouard 3503: /*printf("h=%d ",h);*/
1.126 brouard 3504: } /* end h */
1.267 brouard 3505: /* printf("\n H=%d \n",h); */
1.126 brouard 3506: return po;
3507: }
3508:
1.217 brouard 3509: /************* Higher Back Matrix Product ***************/
1.218 brouard 3510: /* 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 3511: 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 3512: {
1.266 brouard 3513: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3514: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3515: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3516: nhstepm*hstepm matrices.
3517: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3518: (typically every 2 years instead of every month which is too big
1.217 brouard 3519: for the memory).
1.218 brouard 3520: Model is determined by parameters x and covariates have to be
1.266 brouard 3521: included manually here. Then we use a call to bmij(x and cov)
3522: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3523: */
1.217 brouard 3524:
3525: int i, j, d, h, k;
1.266 brouard 3526: double **out, cov[NCOVMAX+1], **bmij();
3527: double **newm, ***newmm;
1.217 brouard 3528: double agexact;
3529: double agebegin, ageend;
1.222 brouard 3530: double **oldm, **savm;
1.217 brouard 3531:
1.266 brouard 3532: newmm=po; /* To be saved */
3533: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3534: /* Hstepm could be zero and should return the unit matrix */
3535: for (i=1;i<=nlstate+ndeath;i++)
3536: for (j=1;j<=nlstate+ndeath;j++){
3537: oldm[i][j]=(i==j ? 1.0 : 0.0);
3538: po[i][j][0]=(i==j ? 1.0 : 0.0);
3539: }
3540: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3541: for(h=1; h <=nhstepm; h++){
3542: for(d=1; d <=hstepm; d++){
3543: newm=savm;
3544: /* Covariates have to be included here again */
3545: cov[1]=1.;
1.271 brouard 3546: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3547: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3548: /* Debug */
3549: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3550: cov[2]=agexact;
3551: if(nagesqr==1)
1.222 brouard 3552: cov[3]= agexact*agexact;
1.325 brouard 3553: for (k=1; k<=nsd;k++){ /* For single dummy covariates only *//* cptcovn error */
1.266 brouard 3554: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3555: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
1.325 brouard 3556: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];/* Bug valgrind */
1.266 brouard 3557: /* 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)); */
3558: }
1.267 brouard 3559: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3560: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3561: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3562: /* 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]); */
3563: }
1.319 brouard 3564: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 *//* For product with age */
3565: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age error!!!*\/ */
3566: if(Dummy[Tage[k]]== 2){ /* dummy with age */
1.267 brouard 3567: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 3568: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
1.267 brouard 3569: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3570: }
3571: /* 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]); */
3572: }
3573: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3574: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.325 brouard 3575: if(Dummy[Tvard[k][1]==0]){
3576: if(Dummy[Tvard[k][2]==0]){
3577: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3578: }else{
3579: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
3580: }
3581: }else{
3582: if(Dummy[Tvard[k][2]==0]){
3583: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
3584: }else{
3585: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
3586: }
3587: }
1.267 brouard 3588: }
1.217 brouard 3589: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3590: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3591:
1.218 brouard 3592: /* Careful transposed matrix */
1.266 brouard 3593: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3594: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3595: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3596: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3597: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3598: /* if((int)age == 70){ */
3599: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3600: /* for(i=1; i<=nlstate+ndeath; i++) { */
3601: /* printf("%d pmmij ",i); */
3602: /* for(j=1;j<=nlstate+ndeath;j++) { */
3603: /* printf("%f ",pmmij[i][j]); */
3604: /* } */
3605: /* printf(" oldm "); */
3606: /* for(j=1;j<=nlstate+ndeath;j++) { */
3607: /* printf("%f ",oldm[i][j]); */
3608: /* } */
3609: /* printf("\n"); */
3610: /* } */
3611: /* } */
3612: savm=oldm;
3613: oldm=newm;
3614: }
3615: for(i=1; i<=nlstate+ndeath; i++)
3616: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3617: po[i][j][h]=newm[i][j];
1.268 brouard 3618: /* if(h==nhstepm) */
3619: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3620: }
1.268 brouard 3621: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3622: } /* end h */
1.268 brouard 3623: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3624: return po;
3625: }
3626:
3627:
1.162 brouard 3628: #ifdef NLOPT
3629: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3630: double fret;
3631: double *xt;
3632: int j;
3633: myfunc_data *d2 = (myfunc_data *) pd;
3634: /* xt = (p1-1); */
3635: xt=vector(1,n);
3636: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3637:
3638: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3639: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3640: printf("Function = %.12lf ",fret);
3641: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3642: printf("\n");
3643: free_vector(xt,1,n);
3644: return fret;
3645: }
3646: #endif
1.126 brouard 3647:
3648: /*************** log-likelihood *************/
3649: double func( double *x)
3650: {
1.226 brouard 3651: int i, ii, j, k, mi, d, kk;
3652: int ioffset=0;
3653: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3654: double **out;
3655: double lli; /* Individual log likelihood */
3656: int s1, s2;
1.228 brouard 3657: 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 3658: double bbh, survp;
3659: long ipmx;
3660: double agexact;
3661: /*extern weight */
3662: /* We are differentiating ll according to initial status */
3663: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3664: /*for(i=1;i<imx;i++)
3665: printf(" %d\n",s[4][i]);
3666: */
1.162 brouard 3667:
1.226 brouard 3668: ++countcallfunc;
1.162 brouard 3669:
1.226 brouard 3670: cov[1]=1.;
1.126 brouard 3671:
1.226 brouard 3672: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3673: ioffset=0;
1.226 brouard 3674: if(mle==1){
3675: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3676: /* Computes the values of the ncovmodel covariates of the model
3677: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3678: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3679: to be observed in j being in i according to the model.
3680: */
1.243 brouard 3681: ioffset=2+nagesqr ;
1.233 brouard 3682: /* Fixed */
1.319 brouard 3683: for (k=1; k<=ncovf;k++){ /* For each fixed covariate dummu or quant or prod */
3684: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3685: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3686: /* 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 3687: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.319 brouard 3688: 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)*/
3689: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3690: }
1.226 brouard 3691: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3692: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3693: has been calculated etc */
3694: /* For an individual i, wav[i] gives the number of effective waves */
3695: /* We compute the contribution to Likelihood of each effective transition
3696: mw[mi][i] is real wave of the mi th effectve wave */
3697: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3698: s2=s[mw[mi+1][i]][i];
3699: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3700: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3701: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3702: */
3703: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3704: 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*/
3705: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? */
1.242 brouard 3706: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3707: }
3708: for (ii=1;ii<=nlstate+ndeath;ii++)
3709: for (j=1;j<=nlstate+ndeath;j++){
3710: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3711: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3712: }
3713: for(d=0; d<dh[mi][i]; d++){
3714: newm=savm;
3715: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3716: cov[2]=agexact;
3717: if(nagesqr==1)
3718: cov[3]= agexact*agexact; /* Should be changed here */
3719: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 3720: if(!FixedV[Tvar[Tage[kk]]])
3721: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
3722: else
3723: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3724: }
3725: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3726: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3727: savm=oldm;
3728: oldm=newm;
3729: } /* end mult */
3730:
3731: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3732: /* But now since version 0.9 we anticipate for bias at large stepm.
3733: * If stepm is larger than one month (smallest stepm) and if the exact delay
3734: * (in months) between two waves is not a multiple of stepm, we rounded to
3735: * the nearest (and in case of equal distance, to the lowest) interval but now
3736: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3737: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3738: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3739: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3740: * -stepm/2 to stepm/2 .
3741: * For stepm=1 the results are the same as for previous versions of Imach.
3742: * For stepm > 1 the results are less biased than in previous versions.
3743: */
1.234 brouard 3744: s1=s[mw[mi][i]][i];
3745: s2=s[mw[mi+1][i]][i];
3746: bbh=(double)bh[mi][i]/(double)stepm;
3747: /* bias bh is positive if real duration
3748: * is higher than the multiple of stepm and negative otherwise.
3749: */
3750: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3751: if( s2 > nlstate){
3752: /* i.e. if s2 is a death state and if the date of death is known
3753: then the contribution to the likelihood is the probability to
3754: die between last step unit time and current step unit time,
3755: which is also equal to probability to die before dh
3756: minus probability to die before dh-stepm .
3757: In version up to 0.92 likelihood was computed
3758: as if date of death was unknown. Death was treated as any other
3759: health state: the date of the interview describes the actual state
3760: and not the date of a change in health state. The former idea was
3761: to consider that at each interview the state was recorded
3762: (healthy, disable or death) and IMaCh was corrected; but when we
3763: introduced the exact date of death then we should have modified
3764: the contribution of an exact death to the likelihood. This new
3765: contribution is smaller and very dependent of the step unit
3766: stepm. It is no more the probability to die between last interview
3767: and month of death but the probability to survive from last
3768: interview up to one month before death multiplied by the
3769: probability to die within a month. Thanks to Chris
3770: Jackson for correcting this bug. Former versions increased
3771: mortality artificially. The bad side is that we add another loop
3772: which slows down the processing. The difference can be up to 10%
3773: lower mortality.
3774: */
3775: /* If, at the beginning of the maximization mostly, the
3776: cumulative probability or probability to be dead is
3777: constant (ie = 1) over time d, the difference is equal to
3778: 0. out[s1][3] = savm[s1][3]: probability, being at state
3779: s1 at precedent wave, to be dead a month before current
3780: wave is equal to probability, being at state s1 at
3781: precedent wave, to be dead at mont of the current
3782: wave. Then the observed probability (that this person died)
3783: is null according to current estimated parameter. In fact,
3784: it should be very low but not zero otherwise the log go to
3785: infinity.
3786: */
1.183 brouard 3787: /* #ifdef INFINITYORIGINAL */
3788: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3789: /* #else */
3790: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3791: /* lli=log(mytinydouble); */
3792: /* else */
3793: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3794: /* #endif */
1.226 brouard 3795: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3796:
1.226 brouard 3797: } else if ( s2==-1 ) { /* alive */
3798: for (j=1,survp=0. ; j<=nlstate; j++)
3799: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3800: /*survp += out[s1][j]; */
3801: lli= log(survp);
3802: }
3803: else if (s2==-4) {
3804: for (j=3,survp=0. ; j<=nlstate; j++)
3805: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3806: lli= log(survp);
3807: }
3808: else if (s2==-5) {
3809: for (j=1,survp=0. ; j<=2; j++)
3810: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3811: lli= log(survp);
3812: }
3813: else{
3814: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3815: /* 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 */
3816: }
3817: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3818: /*if(lli ==000.0)*/
3819: /*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); */
3820: ipmx +=1;
3821: sw += weight[i];
3822: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3823: /* if (lli < log(mytinydouble)){ */
3824: /* 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); */
3825: /* 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]); */
3826: /* } */
3827: } /* end of wave */
3828: } /* end of individual */
3829: } else if(mle==2){
3830: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 3831: ioffset=2+nagesqr ;
3832: for (k=1; k<=ncovf;k++)
3833: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 3834: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3835: for(k=1; k <= ncovv ; k++){
3836: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3837: }
1.226 brouard 3838: for (ii=1;ii<=nlstate+ndeath;ii++)
3839: for (j=1;j<=nlstate+ndeath;j++){
3840: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3841: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3842: }
3843: for(d=0; d<=dh[mi][i]; d++){
3844: newm=savm;
3845: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3846: cov[2]=agexact;
3847: if(nagesqr==1)
3848: cov[3]= agexact*agexact;
3849: for (kk=1; kk<=cptcovage;kk++) {
3850: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3851: }
3852: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3853: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3854: savm=oldm;
3855: oldm=newm;
3856: } /* end mult */
3857:
3858: s1=s[mw[mi][i]][i];
3859: s2=s[mw[mi+1][i]][i];
3860: bbh=(double)bh[mi][i]/(double)stepm;
3861: 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 */
3862: ipmx +=1;
3863: sw += weight[i];
3864: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3865: } /* end of wave */
3866: } /* end of individual */
3867: } else if(mle==3){ /* exponential inter-extrapolation */
3868: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3869: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3870: for(mi=1; mi<= wav[i]-1; mi++){
3871: for (ii=1;ii<=nlstate+ndeath;ii++)
3872: for (j=1;j<=nlstate+ndeath;j++){
3873: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3874: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3875: }
3876: for(d=0; d<dh[mi][i]; d++){
3877: newm=savm;
3878: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3879: cov[2]=agexact;
3880: if(nagesqr==1)
3881: cov[3]= agexact*agexact;
3882: for (kk=1; kk<=cptcovage;kk++) {
3883: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3884: }
3885: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3886: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3887: savm=oldm;
3888: oldm=newm;
3889: } /* end mult */
3890:
3891: s1=s[mw[mi][i]][i];
3892: s2=s[mw[mi+1][i]][i];
3893: bbh=(double)bh[mi][i]/(double)stepm;
3894: 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 */
3895: ipmx +=1;
3896: sw += weight[i];
3897: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3898: } /* end of wave */
3899: } /* end of individual */
3900: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3901: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3902: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3903: for(mi=1; mi<= wav[i]-1; mi++){
3904: for (ii=1;ii<=nlstate+ndeath;ii++)
3905: for (j=1;j<=nlstate+ndeath;j++){
3906: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3907: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3908: }
3909: for(d=0; d<dh[mi][i]; d++){
3910: newm=savm;
3911: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3912: cov[2]=agexact;
3913: if(nagesqr==1)
3914: cov[3]= agexact*agexact;
3915: for (kk=1; kk<=cptcovage;kk++) {
3916: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3917: }
1.126 brouard 3918:
1.226 brouard 3919: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3920: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3921: savm=oldm;
3922: oldm=newm;
3923: } /* end mult */
3924:
3925: s1=s[mw[mi][i]][i];
3926: s2=s[mw[mi+1][i]][i];
3927: if( s2 > nlstate){
3928: lli=log(out[s1][s2] - savm[s1][s2]);
3929: } else if ( s2==-1 ) { /* alive */
3930: for (j=1,survp=0. ; j<=nlstate; j++)
3931: survp += out[s1][j];
3932: lli= log(survp);
3933: }else{
3934: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3935: }
3936: ipmx +=1;
3937: sw += weight[i];
3938: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3939: /* 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 3940: } /* end of wave */
3941: } /* end of individual */
3942: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3943: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3944: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3945: for(mi=1; mi<= wav[i]-1; mi++){
3946: for (ii=1;ii<=nlstate+ndeath;ii++)
3947: for (j=1;j<=nlstate+ndeath;j++){
3948: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3949: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3950: }
3951: for(d=0; d<dh[mi][i]; d++){
3952: newm=savm;
3953: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3954: cov[2]=agexact;
3955: if(nagesqr==1)
3956: cov[3]= agexact*agexact;
3957: for (kk=1; kk<=cptcovage;kk++) {
3958: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3959: }
1.126 brouard 3960:
1.226 brouard 3961: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3962: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3963: savm=oldm;
3964: oldm=newm;
3965: } /* end mult */
3966:
3967: s1=s[mw[mi][i]][i];
3968: s2=s[mw[mi+1][i]][i];
3969: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3970: ipmx +=1;
3971: sw += weight[i];
3972: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3973: /*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]);*/
3974: } /* end of wave */
3975: } /* end of individual */
3976: } /* End of if */
3977: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3978: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3979: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3980: return -l;
1.126 brouard 3981: }
3982:
3983: /*************** log-likelihood *************/
3984: double funcone( double *x)
3985: {
1.228 brouard 3986: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3987: int i, ii, j, k, mi, d, kk;
1.228 brouard 3988: int ioffset=0;
1.131 brouard 3989: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3990: double **out;
3991: double lli; /* Individual log likelihood */
3992: double llt;
3993: int s1, s2;
1.228 brouard 3994: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3995:
1.126 brouard 3996: double bbh, survp;
1.187 brouard 3997: double agexact;
1.214 brouard 3998: double agebegin, ageend;
1.126 brouard 3999: /*extern weight */
4000: /* We are differentiating ll according to initial status */
4001: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4002: /*for(i=1;i<imx;i++)
4003: printf(" %d\n",s[4][i]);
4004: */
4005: cov[1]=1.;
4006:
4007: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4008: ioffset=0;
4009: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 4010: /* ioffset=2+nagesqr+cptcovage; */
4011: ioffset=2+nagesqr;
1.232 brouard 4012: /* Fixed */
1.224 brouard 4013: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4014: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.311 brouard 4015: 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 4016: 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)*/
4017: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4018: /* cov[2+6]=covar[Tvar[6]][i]; */
4019: /* cov[2+6]=covar[2][i]; V2 */
4020: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4021: /* cov[2+7]=covar[Tvar[7]][i]; */
4022: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4023: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4024: /* cov[2+9]=covar[Tvar[9]][i]; */
4025: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4026: }
1.232 brouard 4027: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4028: /* 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?)*\/ */
4029: /* } */
1.231 brouard 4030: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4031: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4032: /* } */
1.225 brouard 4033:
1.233 brouard 4034:
4035: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 4036: /* Wave varying (but not age varying) */
4037: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 4038: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
4039: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4040: }
1.232 brouard 4041: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 4042: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4043: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
4044: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
4045: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
4046: /* 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 4047: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4048: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4049: /* /\* 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]); *\/ */
4050: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4051: /* } */
1.126 brouard 4052: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4053: for (j=1;j<=nlstate+ndeath;j++){
4054: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4055: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4056: }
1.214 brouard 4057:
4058: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4059: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4060: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4061: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4062: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4063: and mw[mi+1][i]. dh depends on stepm.*/
4064: newm=savm;
1.247 brouard 4065: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4066: cov[2]=agexact;
4067: if(nagesqr==1)
4068: cov[3]= agexact*agexact;
4069: for (kk=1; kk<=cptcovage;kk++) {
4070: if(!FixedV[Tvar[Tage[kk]]])
4071: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4072: else
4073: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
4074: }
4075: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4076: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4077: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4078: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4079: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4080: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4081: savm=oldm;
4082: oldm=newm;
1.126 brouard 4083: } /* end mult */
4084:
4085: s1=s[mw[mi][i]][i];
4086: s2=s[mw[mi+1][i]][i];
1.217 brouard 4087: /* if(s2==-1){ */
1.268 brouard 4088: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4089: /* /\* exit(1); *\/ */
4090: /* } */
1.126 brouard 4091: bbh=(double)bh[mi][i]/(double)stepm;
4092: /* bias is positive if real duration
4093: * is higher than the multiple of stepm and negative otherwise.
4094: */
4095: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4096: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4097: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4098: for (j=1,survp=0. ; j<=nlstate; j++)
4099: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4100: lli= log(survp);
1.126 brouard 4101: }else if (mle==1){
1.242 brouard 4102: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4103: } else if(mle==2){
1.242 brouard 4104: 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 4105: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4106: 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 4107: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4108: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4109: } else{ /* mle=0 back to 1 */
1.242 brouard 4110: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4111: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4112: } /* End of if */
4113: ipmx +=1;
4114: sw += weight[i];
4115: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 4116: /*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 4117: if(globpr){
1.246 brouard 4118: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4119: %11.6f %11.6f %11.6f ", \
1.242 brouard 4120: 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 4121: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 4122: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4123: llt +=ll[k]*gipmx/gsw;
4124: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
4125: }
4126: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 4127: }
1.232 brouard 4128: } /* end of wave */
4129: } /* end of individual */
4130: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4131: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4132: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4133: if(globpr==0){ /* First time we count the contributions and weights */
4134: gipmx=ipmx;
4135: gsw=sw;
4136: }
4137: return -l;
1.126 brouard 4138: }
4139:
4140:
4141: /*************** function likelione ***********/
1.292 brouard 4142: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4143: {
4144: /* This routine should help understanding what is done with
4145: the selection of individuals/waves and
4146: to check the exact contribution to the likelihood.
4147: Plotting could be done.
4148: */
4149: int k;
4150:
4151: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4152: strcpy(fileresilk,"ILK_");
1.202 brouard 4153: strcat(fileresilk,fileresu);
1.126 brouard 4154: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4155: printf("Problem with resultfile: %s\n", fileresilk);
4156: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4157: }
1.214 brouard 4158: 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");
4159: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4160: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4161: for(k=1; k<=nlstate; k++)
4162: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4163: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4164: }
4165:
1.292 brouard 4166: *fretone=(*func)(p);
1.126 brouard 4167: if(*globpri !=0){
4168: fclose(ficresilk);
1.205 brouard 4169: if (mle ==0)
4170: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4171: else if(mle >=1)
4172: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4173: 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 4174: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4175:
4176: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4177: 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 4178: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4179: }
1.207 brouard 4180: 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 4181: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4182: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4183: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4184: fflush(fichtm);
1.205 brouard 4185: }
1.126 brouard 4186: return;
4187: }
4188:
4189:
4190: /*********** Maximum Likelihood Estimation ***************/
4191:
4192: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4193: {
1.319 brouard 4194: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4195: double **xi;
4196: double fret;
4197: double fretone; /* Only one call to likelihood */
4198: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4199:
4200: #ifdef NLOPT
4201: int creturn;
4202: nlopt_opt opt;
4203: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4204: double *lb;
4205: double minf; /* the minimum objective value, upon return */
4206: double * p1; /* Shifted parameters from 0 instead of 1 */
4207: myfunc_data dinst, *d = &dinst;
4208: #endif
4209:
4210:
1.126 brouard 4211: xi=matrix(1,npar,1,npar);
4212: for (i=1;i<=npar;i++)
4213: for (j=1;j<=npar;j++)
4214: xi[i][j]=(i==j ? 1.0 : 0.0);
4215: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4216: strcpy(filerespow,"POW_");
1.126 brouard 4217: strcat(filerespow,fileres);
4218: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4219: printf("Problem with resultfile: %s\n", filerespow);
4220: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4221: }
4222: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4223: for (i=1;i<=nlstate;i++)
4224: for(j=1;j<=nlstate+ndeath;j++)
4225: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4226: fprintf(ficrespow,"\n");
1.162 brouard 4227: #ifdef POWELL
1.319 brouard 4228: #ifdef LINMINORIGINAL
4229: #else /* LINMINORIGINAL */
4230:
4231: flatdir=ivector(1,npar);
4232: for (j=1;j<=npar;j++) flatdir[j]=0;
4233: #endif /*LINMINORIGINAL */
4234:
4235: #ifdef FLATSUP
4236: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4237: /* reorganizing p by suppressing flat directions */
4238: for(i=1, jk=1; i <=nlstate; i++){
4239: for(k=1; k <=(nlstate+ndeath); k++){
4240: if (k != i) {
4241: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4242: if(flatdir[jk]==1){
4243: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4244: }
4245: for(j=1; j <=ncovmodel; j++){
4246: printf("%12.7f ",p[jk]);
4247: jk++;
4248: }
4249: printf("\n");
4250: }
4251: }
4252: }
4253: /* skipping */
4254: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4255: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4256: for(k=1; k <=(nlstate+ndeath); k++){
4257: if (k != i) {
4258: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4259: if(flatdir[jk]==1){
4260: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4261: for(j=1; j <=ncovmodel; jk++,j++){
4262: printf(" p[%d]=%12.7f",jk, p[jk]);
4263: /*q[jjk]=p[jk];*/
4264: }
4265: }else{
4266: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4267: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4268: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4269: /*q[jjk]=p[jk];*/
4270: }
4271: }
4272: printf("\n");
4273: }
4274: fflush(stdout);
4275: }
4276: }
4277: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4278: #else /* FLATSUP */
1.126 brouard 4279: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4280: #endif /* FLATSUP */
4281:
4282: #ifdef LINMINORIGINAL
4283: #else
4284: free_ivector(flatdir,1,npar);
4285: #endif /* LINMINORIGINAL*/
4286: #endif /* POWELL */
1.126 brouard 4287:
1.162 brouard 4288: #ifdef NLOPT
4289: #ifdef NEWUOA
4290: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4291: #else
4292: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4293: #endif
4294: lb=vector(0,npar-1);
4295: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4296: nlopt_set_lower_bounds(opt, lb);
4297: nlopt_set_initial_step1(opt, 0.1);
4298:
4299: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4300: d->function = func;
4301: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4302: nlopt_set_min_objective(opt, myfunc, d);
4303: nlopt_set_xtol_rel(opt, ftol);
4304: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4305: printf("nlopt failed! %d\n",creturn);
4306: }
4307: else {
4308: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4309: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4310: iter=1; /* not equal */
4311: }
4312: nlopt_destroy(opt);
4313: #endif
1.319 brouard 4314: #ifdef FLATSUP
4315: /* npared = npar -flatd/ncovmodel; */
4316: /* xired= matrix(1,npared,1,npared); */
4317: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4318: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4319: /* free_matrix(xire,1,npared,1,npared); */
4320: #else /* FLATSUP */
4321: #endif /* FLATSUP */
1.126 brouard 4322: free_matrix(xi,1,npar,1,npar);
4323: fclose(ficrespow);
1.203 brouard 4324: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4325: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4326: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4327:
4328: }
4329:
4330: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4331: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4332: {
4333: double **a,**y,*x,pd;
1.203 brouard 4334: /* double **hess; */
1.164 brouard 4335: int i, j;
1.126 brouard 4336: int *indx;
4337:
4338: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4339: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4340: void lubksb(double **a, int npar, int *indx, double b[]) ;
4341: void ludcmp(double **a, int npar, int *indx, double *d) ;
4342: double gompertz(double p[]);
1.203 brouard 4343: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4344:
4345: printf("\nCalculation of the hessian matrix. Wait...\n");
4346: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4347: for (i=1;i<=npar;i++){
1.203 brouard 4348: printf("%d-",i);fflush(stdout);
4349: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4350:
4351: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4352:
4353: /* printf(" %f ",p[i]);
4354: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4355: }
4356:
4357: for (i=1;i<=npar;i++) {
4358: for (j=1;j<=npar;j++) {
4359: if (j>i) {
1.203 brouard 4360: printf(".%d-%d",i,j);fflush(stdout);
4361: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4362: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4363:
4364: hess[j][i]=hess[i][j];
4365: /*printf(" %lf ",hess[i][j]);*/
4366: }
4367: }
4368: }
4369: printf("\n");
4370: fprintf(ficlog,"\n");
4371:
4372: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4373: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4374:
4375: a=matrix(1,npar,1,npar);
4376: y=matrix(1,npar,1,npar);
4377: x=vector(1,npar);
4378: indx=ivector(1,npar);
4379: for (i=1;i<=npar;i++)
4380: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4381: ludcmp(a,npar,indx,&pd);
4382:
4383: for (j=1;j<=npar;j++) {
4384: for (i=1;i<=npar;i++) x[i]=0;
4385: x[j]=1;
4386: lubksb(a,npar,indx,x);
4387: for (i=1;i<=npar;i++){
4388: matcov[i][j]=x[i];
4389: }
4390: }
4391:
4392: printf("\n#Hessian matrix#\n");
4393: fprintf(ficlog,"\n#Hessian matrix#\n");
4394: for (i=1;i<=npar;i++) {
4395: for (j=1;j<=npar;j++) {
1.203 brouard 4396: printf("%.6e ",hess[i][j]);
4397: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4398: }
4399: printf("\n");
4400: fprintf(ficlog,"\n");
4401: }
4402:
1.203 brouard 4403: /* printf("\n#Covariance matrix#\n"); */
4404: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4405: /* for (i=1;i<=npar;i++) { */
4406: /* for (j=1;j<=npar;j++) { */
4407: /* printf("%.6e ",matcov[i][j]); */
4408: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4409: /* } */
4410: /* printf("\n"); */
4411: /* fprintf(ficlog,"\n"); */
4412: /* } */
4413:
1.126 brouard 4414: /* Recompute Inverse */
1.203 brouard 4415: /* for (i=1;i<=npar;i++) */
4416: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4417: /* ludcmp(a,npar,indx,&pd); */
4418:
4419: /* printf("\n#Hessian matrix recomputed#\n"); */
4420:
4421: /* for (j=1;j<=npar;j++) { */
4422: /* for (i=1;i<=npar;i++) x[i]=0; */
4423: /* x[j]=1; */
4424: /* lubksb(a,npar,indx,x); */
4425: /* for (i=1;i<=npar;i++){ */
4426: /* y[i][j]=x[i]; */
4427: /* printf("%.3e ",y[i][j]); */
4428: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4429: /* } */
4430: /* printf("\n"); */
4431: /* fprintf(ficlog,"\n"); */
4432: /* } */
4433:
4434: /* Verifying the inverse matrix */
4435: #ifdef DEBUGHESS
4436: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4437:
1.203 brouard 4438: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4439: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4440:
4441: for (j=1;j<=npar;j++) {
4442: for (i=1;i<=npar;i++){
1.203 brouard 4443: printf("%.2f ",y[i][j]);
4444: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4445: }
4446: printf("\n");
4447: fprintf(ficlog,"\n");
4448: }
1.203 brouard 4449: #endif
1.126 brouard 4450:
4451: free_matrix(a,1,npar,1,npar);
4452: free_matrix(y,1,npar,1,npar);
4453: free_vector(x,1,npar);
4454: free_ivector(indx,1,npar);
1.203 brouard 4455: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4456:
4457:
4458: }
4459:
4460: /*************** hessian matrix ****************/
4461: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4462: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4463: int i;
4464: int l=1, lmax=20;
1.203 brouard 4465: double k1,k2, res, fx;
1.132 brouard 4466: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4467: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4468: int k=0,kmax=10;
4469: double l1;
4470:
4471: fx=func(x);
4472: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4473: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4474: l1=pow(10,l);
4475: delts=delt;
4476: for(k=1 ; k <kmax; k=k+1){
4477: delt = delta*(l1*k);
4478: p2[theta]=x[theta] +delt;
1.145 brouard 4479: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4480: p2[theta]=x[theta]-delt;
4481: k2=func(p2)-fx;
4482: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4483: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4484:
1.203 brouard 4485: #ifdef DEBUGHESSII
1.126 brouard 4486: 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);
4487: 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);
4488: #endif
4489: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4490: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4491: k=kmax;
4492: }
4493: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4494: k=kmax; l=lmax*10;
1.126 brouard 4495: }
4496: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4497: delts=delt;
4498: }
1.203 brouard 4499: } /* End loop k */
1.126 brouard 4500: }
4501: delti[theta]=delts;
4502: return res;
4503:
4504: }
4505:
1.203 brouard 4506: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4507: {
4508: int i;
1.164 brouard 4509: int l=1, lmax=20;
1.126 brouard 4510: double k1,k2,k3,k4,res,fx;
1.132 brouard 4511: double p2[MAXPARM+1];
1.203 brouard 4512: int k, kmax=1;
4513: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4514:
4515: int firstime=0;
1.203 brouard 4516:
1.126 brouard 4517: fx=func(x);
1.203 brouard 4518: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4519: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4520: p2[thetai]=x[thetai]+delti[thetai]*k;
4521: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4522: k1=func(p2)-fx;
4523:
1.203 brouard 4524: p2[thetai]=x[thetai]+delti[thetai]*k;
4525: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4526: k2=func(p2)-fx;
4527:
1.203 brouard 4528: p2[thetai]=x[thetai]-delti[thetai]*k;
4529: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4530: k3=func(p2)-fx;
4531:
1.203 brouard 4532: p2[thetai]=x[thetai]-delti[thetai]*k;
4533: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4534: k4=func(p2)-fx;
1.203 brouard 4535: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4536: if(k1*k2*k3*k4 <0.){
1.208 brouard 4537: firstime=1;
1.203 brouard 4538: kmax=kmax+10;
1.208 brouard 4539: }
4540: if(kmax >=10 || firstime ==1){
1.246 brouard 4541: 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);
4542: 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 4543: 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);
4544: 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);
4545: }
4546: #ifdef DEBUGHESSIJ
4547: v1=hess[thetai][thetai];
4548: v2=hess[thetaj][thetaj];
4549: cv12=res;
4550: /* Computing eigen value of Hessian matrix */
4551: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4552: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4553: if ((lc2 <0) || (lc1 <0) ){
4554: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4555: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4556: 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);
4557: 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);
4558: }
1.126 brouard 4559: #endif
4560: }
4561: return res;
4562: }
4563:
1.203 brouard 4564: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4565: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4566: /* { */
4567: /* int i; */
4568: /* int l=1, lmax=20; */
4569: /* double k1,k2,k3,k4,res,fx; */
4570: /* double p2[MAXPARM+1]; */
4571: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4572: /* int k=0,kmax=10; */
4573: /* double l1; */
4574:
4575: /* fx=func(x); */
4576: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4577: /* l1=pow(10,l); */
4578: /* delts=delt; */
4579: /* for(k=1 ; k <kmax; k=k+1){ */
4580: /* delt = delti*(l1*k); */
4581: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4582: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4583: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4584: /* k1=func(p2)-fx; */
4585:
4586: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4587: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4588: /* k2=func(p2)-fx; */
4589:
4590: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4591: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4592: /* k3=func(p2)-fx; */
4593:
4594: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4595: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4596: /* k4=func(p2)-fx; */
4597: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4598: /* #ifdef DEBUGHESSIJ */
4599: /* 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); */
4600: /* 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); */
4601: /* #endif */
4602: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4603: /* k=kmax; */
4604: /* } */
4605: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4606: /* k=kmax; l=lmax*10; */
4607: /* } */
4608: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4609: /* delts=delt; */
4610: /* } */
4611: /* } /\* End loop k *\/ */
4612: /* } */
4613: /* delti[theta]=delts; */
4614: /* return res; */
4615: /* } */
4616:
4617:
1.126 brouard 4618: /************** Inverse of matrix **************/
4619: void ludcmp(double **a, int n, int *indx, double *d)
4620: {
4621: int i,imax,j,k;
4622: double big,dum,sum,temp;
4623: double *vv;
4624:
4625: vv=vector(1,n);
4626: *d=1.0;
4627: for (i=1;i<=n;i++) {
4628: big=0.0;
4629: for (j=1;j<=n;j++)
4630: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4631: if (big == 0.0){
4632: printf(" Singular Hessian matrix at row %d:\n",i);
4633: for (j=1;j<=n;j++) {
4634: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4635: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4636: }
4637: fflush(ficlog);
4638: fclose(ficlog);
4639: nrerror("Singular matrix in routine ludcmp");
4640: }
1.126 brouard 4641: vv[i]=1.0/big;
4642: }
4643: for (j=1;j<=n;j++) {
4644: for (i=1;i<j;i++) {
4645: sum=a[i][j];
4646: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4647: a[i][j]=sum;
4648: }
4649: big=0.0;
4650: for (i=j;i<=n;i++) {
4651: sum=a[i][j];
4652: for (k=1;k<j;k++)
4653: sum -= a[i][k]*a[k][j];
4654: a[i][j]=sum;
4655: if ( (dum=vv[i]*fabs(sum)) >= big) {
4656: big=dum;
4657: imax=i;
4658: }
4659: }
4660: if (j != imax) {
4661: for (k=1;k<=n;k++) {
4662: dum=a[imax][k];
4663: a[imax][k]=a[j][k];
4664: a[j][k]=dum;
4665: }
4666: *d = -(*d);
4667: vv[imax]=vv[j];
4668: }
4669: indx[j]=imax;
4670: if (a[j][j] == 0.0) a[j][j]=TINY;
4671: if (j != n) {
4672: dum=1.0/(a[j][j]);
4673: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4674: }
4675: }
4676: free_vector(vv,1,n); /* Doesn't work */
4677: ;
4678: }
4679:
4680: void lubksb(double **a, int n, int *indx, double b[])
4681: {
4682: int i,ii=0,ip,j;
4683: double sum;
4684:
4685: for (i=1;i<=n;i++) {
4686: ip=indx[i];
4687: sum=b[ip];
4688: b[ip]=b[i];
4689: if (ii)
4690: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4691: else if (sum) ii=i;
4692: b[i]=sum;
4693: }
4694: for (i=n;i>=1;i--) {
4695: sum=b[i];
4696: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4697: b[i]=sum/a[i][i];
4698: }
4699: }
4700:
4701: void pstamp(FILE *fichier)
4702: {
1.196 brouard 4703: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4704: }
4705:
1.297 brouard 4706: void date2dmy(double date,double *day, double *month, double *year){
4707: double yp=0., yp1=0., yp2=0.;
4708:
4709: yp1=modf(date,&yp);/* extracts integral of date in yp and
4710: fractional in yp1 */
4711: *year=yp;
4712: yp2=modf((yp1*12),&yp);
4713: *month=yp;
4714: yp1=modf((yp2*30.5),&yp);
4715: *day=yp;
4716: if(*day==0) *day=1;
4717: if(*month==0) *month=1;
4718: }
4719:
1.253 brouard 4720:
4721:
1.126 brouard 4722: /************ Frequencies ********************/
1.251 brouard 4723: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4724: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4725: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4726: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4727:
1.265 brouard 4728: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4729: int iind=0, iage=0;
4730: int mi; /* Effective wave */
4731: int first;
4732: double ***freq; /* Frequencies */
1.268 brouard 4733: 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 */
4734: 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 4735: double *meanq, *stdq, *idq;
1.226 brouard 4736: double **meanqt;
4737: double *pp, **prop, *posprop, *pospropt;
4738: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4739: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4740: double agebegin, ageend;
4741:
4742: pp=vector(1,nlstate);
1.251 brouard 4743: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4744: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4745: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4746: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4747: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4748: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4749: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4750: meanqt=matrix(1,lastpass,1,nqtveff);
4751: strcpy(fileresp,"P_");
4752: strcat(fileresp,fileresu);
4753: /*strcat(fileresphtm,fileresu);*/
4754: if((ficresp=fopen(fileresp,"w"))==NULL) {
4755: printf("Problem with prevalence resultfile: %s\n", fileresp);
4756: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4757: exit(0);
4758: }
1.240 brouard 4759:
1.226 brouard 4760: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4761: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4762: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4763: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4764: fflush(ficlog);
4765: exit(70);
4766: }
4767: else{
4768: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4769: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4770: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4771: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4772: }
1.319 brouard 4773: 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 4774:
1.226 brouard 4775: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4776: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4777: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4778: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4779: fflush(ficlog);
4780: exit(70);
1.240 brouard 4781: } else{
1.226 brouard 4782: 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 4783: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4784: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4785: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4786: }
1.319 brouard 4787: 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 4788:
1.253 brouard 4789: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4790: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4791: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4792: j1=0;
1.126 brouard 4793:
1.227 brouard 4794: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4795: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4796: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4797:
4798:
1.226 brouard 4799: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4800: reference=low_education V1=0,V2=0
4801: med_educ V1=1 V2=0,
4802: high_educ V1=0 V2=1
4803: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4804: */
1.249 brouard 4805: dateintsum=0;
4806: k2cpt=0;
4807:
1.253 brouard 4808: if(cptcoveff == 0 )
1.265 brouard 4809: nl=1; /* Constant and age model only */
1.253 brouard 4810: else
4811: nl=2;
1.265 brouard 4812:
4813: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4814: /* Loop on nj=1 or 2 if dummy covariates j!=0
4815: * Loop on j1(1 to 2**cptcoveff) covariate combination
4816: * freq[s1][s2][iage] =0.
4817: * Loop on iind
4818: * ++freq[s1][s2][iage] weighted
4819: * end iind
4820: * if covariate and j!0
4821: * headers Variable on one line
4822: * endif cov j!=0
4823: * header of frequency table by age
4824: * Loop on age
4825: * pp[s1]+=freq[s1][s2][iage] weighted
4826: * pos+=freq[s1][s2][iage] weighted
4827: * Loop on s1 initial state
4828: * fprintf(ficresp
4829: * end s1
4830: * end age
4831: * if j!=0 computes starting values
4832: * end compute starting values
4833: * end j1
4834: * end nl
4835: */
1.253 brouard 4836: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4837: if(nj==1)
4838: j=0; /* First pass for the constant */
1.265 brouard 4839: else{
1.253 brouard 4840: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4841: }
1.251 brouard 4842: first=1;
1.265 brouard 4843: 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 4844: posproptt=0.;
4845: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4846: scanf("%d", i);*/
4847: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4848: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4849: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4850: freq[i][s2][m]=0;
1.251 brouard 4851:
4852: for (i=1; i<=nlstate; i++) {
1.240 brouard 4853: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4854: prop[i][m]=0;
4855: posprop[i]=0;
4856: pospropt[i]=0;
4857: }
1.283 brouard 4858: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4859: idq[z1]=0.;
4860: meanq[z1]=0.;
4861: stdq[z1]=0.;
1.283 brouard 4862: }
4863: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4864: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4865: /* meanqt[m][z1]=0.; */
4866: /* } */
4867: /* } */
1.251 brouard 4868: /* dateintsum=0; */
4869: /* k2cpt=0; */
4870:
1.265 brouard 4871: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4872: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4873: bool=1;
4874: if(j !=0){
4875: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4876: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4877: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4878: /* if(Tvaraff[z1] ==-20){ */
4879: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4880: /* }else if(Tvaraff[z1] ==-10){ */
4881: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4882: /* }else */
4883: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4884: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4885: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4886: /* 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",
4887: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4888: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4889: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4890: } /* Onlyf fixed */
4891: } /* end z1 */
4892: } /* cptcovn > 0 */
4893: } /* end any */
4894: }/* end j==0 */
1.265 brouard 4895: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4896: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4897: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4898: m=mw[mi][iind];
4899: if(j!=0){
4900: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4901: for (z1=1; z1<=cptcoveff; z1++) {
4902: if( Fixed[Tmodelind[z1]]==1){
4903: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4904: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4905: value is -1, we don't select. It differs from the
4906: constant and age model which counts them. */
4907: bool=0; /* not selected */
4908: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4909: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4910: bool=0;
4911: }
4912: }
4913: }
4914: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4915: } /* end j==0 */
4916: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4917: if(bool==1){ /*Selected */
1.251 brouard 4918: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4919: and mw[mi+1][iind]. dh depends on stepm. */
4920: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4921: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4922: if(m >=firstpass && m <=lastpass){
4923: k2=anint[m][iind]+(mint[m][iind]/12.);
4924: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4925: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4926: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4927: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4928: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4929: if (m<lastpass) {
4930: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4931: /* 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]); */
4932: if(s[m][iind]==-1)
4933: 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.));
4934: 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 4935: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
4936: if(!isnan(covar[ncovcol+z1][iind])){
4937: idq[z1]=idq[z1]+weight[iind];
4938: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4939: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
4940: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4941: }
1.284 brouard 4942: }
1.251 brouard 4943: /* if((int)agev[m][iind] == 55) */
4944: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4945: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4946: 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 4947: }
1.251 brouard 4948: } /* end if between passes */
4949: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4950: dateintsum=dateintsum+k2; /* on all covariates ?*/
4951: k2cpt++;
4952: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4953: }
1.251 brouard 4954: }else{
4955: bool=1;
4956: }/* end bool 2 */
4957: } /* end m */
1.284 brouard 4958: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4959: /* idq[z1]=idq[z1]+weight[iind]; */
4960: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4961: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4962: /* } */
1.251 brouard 4963: } /* end bool */
4964: } /* end iind = 1 to imx */
1.319 brouard 4965: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 4966: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4967:
4968:
4969: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4970: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4971: pstamp(ficresp);
1.251 brouard 4972: if (cptcoveff>0 && j!=0){
1.265 brouard 4973: pstamp(ficresp);
1.251 brouard 4974: printf( "\n#********** Variable ");
4975: fprintf(ficresp, "\n#********** Variable ");
4976: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4977: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4978: fprintf(ficlog, "\n#********** Variable ");
4979: for (z1=1; z1<=cptcoveff; z1++){
4980: if(!FixedV[Tvaraff[z1]]){
4981: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4982: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4983: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4984: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4985: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4986: }else{
1.251 brouard 4987: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4988: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4989: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4990: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4991: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4992: }
4993: }
4994: printf( "**********\n#");
4995: fprintf(ficresp, "**********\n#");
4996: fprintf(ficresphtm, "**********</h3>\n");
4997: fprintf(ficresphtmfr, "**********</h3>\n");
4998: fprintf(ficlog, "**********\n");
4999: }
1.284 brouard 5000: /*
5001: Printing means of quantitative variables if any
5002: */
5003: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5004: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5005: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5006: if(weightopt==1){
5007: printf(" Weighted mean and standard deviation of");
5008: fprintf(ficlog," Weighted mean and standard deviation of");
5009: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5010: }
1.311 brouard 5011: /* mu = \frac{w x}{\sum w}
5012: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5013: */
5014: 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]));
5015: 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]));
5016: 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 5017: }
5018: /* for (z1=1; z1<= nqtveff; z1++) { */
5019: /* for(m=1;m<=lastpass;m++){ */
5020: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5021: /* } */
5022: /* } */
1.283 brouard 5023:
1.251 brouard 5024: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 5025: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
5026: fprintf(ficresp, " Age");
5027: 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 5028: for(i=1; i<=nlstate;i++) {
1.265 brouard 5029: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5030: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5031: }
1.265 brouard 5032: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5033: fprintf(ficresphtm, "\n");
5034:
5035: /* Header of frequency table by age */
5036: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5037: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5038: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5039: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5040: if(s2!=0 && m!=0)
5041: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5042: }
1.226 brouard 5043: }
1.251 brouard 5044: fprintf(ficresphtmfr, "\n");
5045:
5046: /* For each age */
5047: for(iage=iagemin; iage <= iagemax+3; iage++){
5048: fprintf(ficresphtm,"<tr>");
5049: if(iage==iagemax+1){
5050: fprintf(ficlog,"1");
5051: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5052: }else if(iage==iagemax+2){
5053: fprintf(ficlog,"0");
5054: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5055: }else if(iage==iagemax+3){
5056: fprintf(ficlog,"Total");
5057: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5058: }else{
1.240 brouard 5059: if(first==1){
1.251 brouard 5060: first=0;
5061: printf("See log file for details...\n");
5062: }
5063: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5064: fprintf(ficlog,"Age %d", iage);
5065: }
1.265 brouard 5066: for(s1=1; s1 <=nlstate ; s1++){
5067: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5068: pp[s1] += freq[s1][m][iage];
1.251 brouard 5069: }
1.265 brouard 5070: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5071: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5072: pos += freq[s1][m][iage];
5073: if(pp[s1]>=1.e-10){
1.251 brouard 5074: if(first==1){
1.265 brouard 5075: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5076: }
1.265 brouard 5077: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5078: }else{
5079: if(first==1)
1.265 brouard 5080: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5081: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5082: }
5083: }
5084:
1.265 brouard 5085: for(s1=1; s1 <=nlstate ; s1++){
5086: /* posprop[s1]=0; */
5087: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5088: pp[s1] += freq[s1][m][iage];
5089: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5090:
5091: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5092: pos += pp[s1]; /* pos is the total number of transitions until this age */
5093: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5094: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5095: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5096: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5097: }
5098:
5099: /* Writing ficresp */
5100: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
5101: if( iage <= iagemax){
5102: fprintf(ficresp," %d",iage);
5103: }
5104: }else if( nj==2){
5105: if( iage <= iagemax){
5106: fprintf(ficresp," %d",iage);
5107: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
5108: }
1.240 brouard 5109: }
1.265 brouard 5110: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5111: if(pos>=1.e-5){
1.251 brouard 5112: if(first==1)
1.265 brouard 5113: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5114: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5115: }else{
5116: if(first==1)
1.265 brouard 5117: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5118: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5119: }
5120: if( iage <= iagemax){
5121: if(pos>=1.e-5){
1.265 brouard 5122: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
5123: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5124: }else if( nj==2){
5125: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5126: }
5127: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5128: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5129: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5130: } else{
5131: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
5132: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5133: }
1.240 brouard 5134: }
1.265 brouard 5135: pospropt[s1] +=posprop[s1];
5136: } /* end loop s1 */
1.251 brouard 5137: /* pospropt=0.; */
1.265 brouard 5138: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5139: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5140: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5141: if(first==1){
1.265 brouard 5142: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5143: }
1.265 brouard 5144: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5145: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5146: }
1.265 brouard 5147: if(s1!=0 && m!=0)
5148: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5149: }
1.265 brouard 5150: } /* end loop s1 */
1.251 brouard 5151: posproptt=0.;
1.265 brouard 5152: for(s1=1; s1 <=nlstate; s1++){
5153: posproptt += pospropt[s1];
1.251 brouard 5154: }
5155: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5156: fprintf(ficresphtm,"</tr>\n");
5157: if((cptcoveff==0 && nj==1)|| nj==2 ) {
5158: if(iage <= iagemax)
5159: fprintf(ficresp,"\n");
1.240 brouard 5160: }
1.251 brouard 5161: if(first==1)
5162: printf("Others in log...\n");
5163: fprintf(ficlog,"\n");
5164: } /* end loop age iage */
1.265 brouard 5165:
1.251 brouard 5166: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5167: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5168: if(posproptt < 1.e-5){
1.265 brouard 5169: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5170: }else{
1.265 brouard 5171: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5172: }
1.226 brouard 5173: }
1.251 brouard 5174: fprintf(ficresphtm,"</tr>\n");
5175: fprintf(ficresphtm,"</table>\n");
5176: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5177: if(posproptt < 1.e-5){
1.251 brouard 5178: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5179: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5180: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5181: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5182: invalidvarcomb[j1]=1;
1.226 brouard 5183: }else{
1.251 brouard 5184: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
5185: invalidvarcomb[j1]=0;
1.226 brouard 5186: }
1.251 brouard 5187: fprintf(ficresphtmfr,"</table>\n");
5188: fprintf(ficlog,"\n");
5189: if(j!=0){
5190: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5191: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5192: for(k=1; k <=(nlstate+ndeath); k++){
5193: if (k != i) {
1.265 brouard 5194: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5195: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5196: if(j1==1){ /* All dummy covariates to zero */
5197: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5198: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5199: printf("%d%d ",i,k);
5200: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5201: 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]));
5202: 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]));
5203: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5204: }
1.253 brouard 5205: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5206: for(iage=iagemin; iage <= iagemax+3; iage++){
5207: x[iage]= (double)iage;
5208: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5209: /* 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 5210: }
1.268 brouard 5211: /* Some are not finite, but linreg will ignore these ages */
5212: no=0;
1.253 brouard 5213: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5214: pstart[s1]=b;
5215: pstart[s1-1]=a;
1.252 brouard 5216: }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 */
5217: 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]);
5218: 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 5219: 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 5220: printf("%d%d ",i,k);
5221: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5222: 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 5223: }else{ /* Other cases, like quantitative fixed or varying covariates */
5224: ;
5225: }
5226: /* printf("%12.7f )", param[i][jj][k]); */
5227: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5228: s1++;
1.251 brouard 5229: } /* end jj */
5230: } /* end k!= i */
5231: } /* end k */
1.265 brouard 5232: } /* end i, s1 */
1.251 brouard 5233: } /* end j !=0 */
5234: } /* end selected combination of covariate j1 */
5235: if(j==0){ /* We can estimate starting values from the occurences in each case */
5236: printf("#Freqsummary: Starting values for the constants:\n");
5237: fprintf(ficlog,"\n");
1.265 brouard 5238: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5239: for(k=1; k <=(nlstate+ndeath); k++){
5240: if (k != i) {
5241: printf("%d%d ",i,k);
5242: fprintf(ficlog,"%d%d ",i,k);
5243: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5244: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5245: if(jj==1){ /* Age has to be done */
1.265 brouard 5246: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5247: 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]));
5248: 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 5249: }
5250: /* printf("%12.7f )", param[i][jj][k]); */
5251: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5252: s1++;
1.250 brouard 5253: }
1.251 brouard 5254: printf("\n");
5255: fprintf(ficlog,"\n");
1.250 brouard 5256: }
5257: }
1.284 brouard 5258: } /* end of state i */
1.251 brouard 5259: printf("#Freqsummary\n");
5260: fprintf(ficlog,"\n");
1.265 brouard 5261: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5262: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5263: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5264: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5265: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5266: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5267: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5268: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5269: /* } */
5270: }
1.265 brouard 5271: } /* end loop s1 */
1.251 brouard 5272:
5273: printf("\n");
5274: fprintf(ficlog,"\n");
5275: } /* end j=0 */
1.249 brouard 5276: } /* end j */
1.252 brouard 5277:
1.253 brouard 5278: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5279: for(i=1, jk=1; i <=nlstate; i++){
5280: for(j=1; j <=nlstate+ndeath; j++){
5281: if(j!=i){
5282: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5283: printf("%1d%1d",i,j);
5284: fprintf(ficparo,"%1d%1d",i,j);
5285: for(k=1; k<=ncovmodel;k++){
5286: /* printf(" %lf",param[i][j][k]); */
5287: /* fprintf(ficparo," %lf",param[i][j][k]); */
5288: p[jk]=pstart[jk];
5289: printf(" %f ",pstart[jk]);
5290: fprintf(ficparo," %f ",pstart[jk]);
5291: jk++;
5292: }
5293: printf("\n");
5294: fprintf(ficparo,"\n");
5295: }
5296: }
5297: }
5298: } /* end mle=-2 */
1.226 brouard 5299: dateintmean=dateintsum/k2cpt;
1.296 brouard 5300: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5301:
1.226 brouard 5302: fclose(ficresp);
5303: fclose(ficresphtm);
5304: fclose(ficresphtmfr);
1.283 brouard 5305: free_vector(idq,1,nqfveff);
1.226 brouard 5306: free_vector(meanq,1,nqfveff);
1.284 brouard 5307: free_vector(stdq,1,nqfveff);
1.226 brouard 5308: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5309: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5310: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5311: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5312: free_vector(pospropt,1,nlstate);
5313: free_vector(posprop,1,nlstate);
1.251 brouard 5314: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5315: free_vector(pp,1,nlstate);
5316: /* End of freqsummary */
5317: }
1.126 brouard 5318:
1.268 brouard 5319: /* Simple linear regression */
5320: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5321:
5322: /* y=a+bx regression */
5323: double sumx = 0.0; /* sum of x */
5324: double sumx2 = 0.0; /* sum of x**2 */
5325: double sumxy = 0.0; /* sum of x * y */
5326: double sumy = 0.0; /* sum of y */
5327: double sumy2 = 0.0; /* sum of y**2 */
5328: double sume2 = 0.0; /* sum of square or residuals */
5329: double yhat;
5330:
5331: double denom=0;
5332: int i;
5333: int ne=*no;
5334:
5335: for ( i=ifi, ne=0;i<=ila;i++) {
5336: if(!isfinite(x[i]) || !isfinite(y[i])){
5337: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5338: continue;
5339: }
5340: ne=ne+1;
5341: sumx += x[i];
5342: sumx2 += x[i]*x[i];
5343: sumxy += x[i] * y[i];
5344: sumy += y[i];
5345: sumy2 += y[i]*y[i];
5346: denom = (ne * sumx2 - sumx*sumx);
5347: /* 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); */
5348: }
5349:
5350: denom = (ne * sumx2 - sumx*sumx);
5351: if (denom == 0) {
5352: // vertical, slope m is infinity
5353: *b = INFINITY;
5354: *a = 0;
5355: if (r) *r = 0;
5356: return 1;
5357: }
5358:
5359: *b = (ne * sumxy - sumx * sumy) / denom;
5360: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5361: if (r!=NULL) {
5362: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5363: sqrt((sumx2 - sumx*sumx/ne) *
5364: (sumy2 - sumy*sumy/ne));
5365: }
5366: *no=ne;
5367: for ( i=ifi, ne=0;i<=ila;i++) {
5368: if(!isfinite(x[i]) || !isfinite(y[i])){
5369: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5370: continue;
5371: }
5372: ne=ne+1;
5373: yhat = y[i] - *a -*b* x[i];
5374: sume2 += yhat * yhat ;
5375:
5376: denom = (ne * sumx2 - sumx*sumx);
5377: /* 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); */
5378: }
5379: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5380: *sa= *sb * sqrt(sumx2/ne);
5381:
5382: return 0;
5383: }
5384:
1.126 brouard 5385: /************ Prevalence ********************/
1.227 brouard 5386: 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)
5387: {
5388: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5389: in each health status at the date of interview (if between dateprev1 and dateprev2).
5390: We still use firstpass and lastpass as another selection.
5391: */
1.126 brouard 5392:
1.227 brouard 5393: int i, m, jk, j1, bool, z1,j, iv;
5394: int mi; /* Effective wave */
5395: int iage;
5396: double agebegin, ageend;
5397:
5398: double **prop;
5399: double posprop;
5400: double y2; /* in fractional years */
5401: int iagemin, iagemax;
5402: int first; /** to stop verbosity which is redirected to log file */
5403:
5404: iagemin= (int) agemin;
5405: iagemax= (int) agemax;
5406: /*pp=vector(1,nlstate);*/
1.251 brouard 5407: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5408: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5409: j1=0;
1.222 brouard 5410:
1.227 brouard 5411: /*j=cptcoveff;*/
5412: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5413:
1.288 brouard 5414: first=0;
1.227 brouard 5415: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5416: for (i=1; i<=nlstate; i++)
1.251 brouard 5417: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5418: prop[i][iage]=0.0;
5419: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5420: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5421: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5422:
5423: for (i=1; i<=imx; i++) { /* Each individual */
5424: bool=1;
5425: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5426: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5427: m=mw[mi][i];
5428: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5429: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5430: for (z1=1; z1<=cptcoveff; z1++){
5431: if( Fixed[Tmodelind[z1]]==1){
5432: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5433: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5434: bool=0;
5435: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5436: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5437: bool=0;
5438: }
5439: }
5440: if(bool==1){ /* Otherwise we skip that wave/person */
5441: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5442: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5443: if(m >=firstpass && m <=lastpass){
5444: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5445: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5446: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5447: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5448: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5449: 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);
5450: exit(1);
5451: }
5452: if (s[m][i]>0 && s[m][i]<=nlstate) {
5453: /*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]]);*/
5454: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5455: prop[s[m][i]][iagemax+3] += weight[i];
5456: } /* end valid statuses */
5457: } /* end selection of dates */
5458: } /* end selection of waves */
5459: } /* end bool */
5460: } /* end wave */
5461: } /* end individual */
5462: for(i=iagemin; i <= iagemax+3; i++){
5463: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5464: posprop += prop[jk][i];
5465: }
5466:
5467: for(jk=1; jk <=nlstate ; jk++){
5468: if( i <= iagemax){
5469: if(posprop>=1.e-5){
5470: probs[i][jk][j1]= prop[jk][i]/posprop;
5471: } else{
1.288 brouard 5472: if(!first){
5473: first=1;
1.266 brouard 5474: 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]);
5475: }else{
1.288 brouard 5476: 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 5477: }
5478: }
5479: }
5480: }/* end jk */
5481: }/* end i */
1.222 brouard 5482: /*} *//* end i1 */
1.227 brouard 5483: } /* end j1 */
1.222 brouard 5484:
1.227 brouard 5485: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5486: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5487: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5488: } /* End of prevalence */
1.126 brouard 5489:
5490: /************* Waves Concatenation ***************/
5491:
5492: 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)
5493: {
1.298 brouard 5494: /* 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 5495: Death is a valid wave (if date is known).
5496: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5497: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5498: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5499: */
1.126 brouard 5500:
1.224 brouard 5501: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5502: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5503: double sum=0., jmean=0.;*/
1.224 brouard 5504: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5505: int j, k=0,jk, ju, jl;
5506: double sum=0.;
5507: first=0;
1.214 brouard 5508: firstwo=0;
1.217 brouard 5509: firsthree=0;
1.218 brouard 5510: firstfour=0;
1.164 brouard 5511: jmin=100000;
1.126 brouard 5512: jmax=-1;
5513: jmean=0.;
1.224 brouard 5514:
5515: /* Treating live states */
1.214 brouard 5516: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5517: mi=0; /* First valid wave */
1.227 brouard 5518: mli=0; /* Last valid wave */
1.309 brouard 5519: m=firstpass; /* Loop on waves */
5520: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5521: 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 */
5522: mli=m-1;/* mw[++mi][i]=m-1; */
5523: }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 5524: 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 5525: mli=m;
1.224 brouard 5526: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5527: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5528: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5529: }
1.309 brouard 5530: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5531: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5532: break;
1.224 brouard 5533: #else
1.317 brouard 5534: 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 5535: if(firsthree == 0){
1.302 brouard 5536: 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 5537: firsthree=1;
1.317 brouard 5538: }else if(firsthree >=1 && firsthree < 10){
5539: 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);
5540: firsthree++;
5541: }else if(firsthree == 10){
5542: printf("Information, too many Information flags: no more reported to log either\n");
5543: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
5544: firsthree++;
5545: }else{
5546: firsthree++;
1.227 brouard 5547: }
1.309 brouard 5548: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5549: mli=m;
5550: }
5551: if(s[m][i]==-2){ /* Vital status is really unknown */
5552: nbwarn++;
1.309 brouard 5553: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5554: 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);
5555: 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);
5556: }
5557: break;
5558: }
5559: break;
1.224 brouard 5560: #endif
1.227 brouard 5561: }/* End m >= lastpass */
1.126 brouard 5562: }/* end while */
1.224 brouard 5563:
1.227 brouard 5564: /* 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 5565: /* After last pass */
1.224 brouard 5566: /* Treating death states */
1.214 brouard 5567: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5568: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5569: /* } */
1.126 brouard 5570: mi++; /* Death is another wave */
5571: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5572: /* Only death is a correct wave */
1.126 brouard 5573: mw[mi][i]=m;
1.257 brouard 5574: } /* else not in a death state */
1.224 brouard 5575: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5576: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5577: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5578: 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 5579: nbwarn++;
5580: if(firstfiv==0){
1.309 brouard 5581: 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 5582: firstfiv=1;
5583: }else{
1.309 brouard 5584: 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 5585: }
1.309 brouard 5586: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5587: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5588: nberr++;
5589: if(firstwo==0){
1.309 brouard 5590: 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 5591: firstwo=1;
5592: }
1.309 brouard 5593: 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 5594: }
1.257 brouard 5595: }else{ /* if date of interview is unknown */
1.227 brouard 5596: /* death is known but not confirmed by death status at any wave */
5597: if(firstfour==0){
1.309 brouard 5598: 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 5599: firstfour=1;
5600: }
1.309 brouard 5601: 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 5602: }
1.224 brouard 5603: } /* end if date of death is known */
5604: #endif
1.309 brouard 5605: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5606: /* wav[i]=mw[mi][i]; */
1.126 brouard 5607: if(mi==0){
5608: nbwarn++;
5609: if(first==0){
1.227 brouard 5610: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5611: first=1;
1.126 brouard 5612: }
5613: if(first==1){
1.227 brouard 5614: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5615: }
5616: } /* end mi==0 */
5617: } /* End individuals */
1.214 brouard 5618: /* wav and mw are no more changed */
1.223 brouard 5619:
1.317 brouard 5620: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
5621: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
5622:
5623:
1.126 brouard 5624: for(i=1; i<=imx; i++){
5625: for(mi=1; mi<wav[i];mi++){
5626: if (stepm <=0)
1.227 brouard 5627: dh[mi][i]=1;
1.126 brouard 5628: else{
1.260 brouard 5629: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5630: if (agedc[i] < 2*AGESUP) {
5631: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5632: if(j==0) j=1; /* Survives at least one month after exam */
5633: else if(j<0){
5634: nberr++;
5635: 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]);
5636: j=1; /* Temporary Dangerous patch */
5637: 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);
5638: 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]);
5639: 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);
5640: }
5641: k=k+1;
5642: if (j >= jmax){
5643: jmax=j;
5644: ijmax=i;
5645: }
5646: if (j <= jmin){
5647: jmin=j;
5648: ijmin=i;
5649: }
5650: sum=sum+j;
5651: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5652: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5653: }
5654: }
5655: else{
5656: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5657: /* 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 5658:
1.227 brouard 5659: k=k+1;
5660: if (j >= jmax) {
5661: jmax=j;
5662: ijmax=i;
5663: }
5664: else if (j <= jmin){
5665: jmin=j;
5666: ijmin=i;
5667: }
5668: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5669: /*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]);*/
5670: if(j<0){
5671: nberr++;
5672: 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]);
5673: 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]);
5674: }
5675: sum=sum+j;
5676: }
5677: jk= j/stepm;
5678: jl= j -jk*stepm;
5679: ju= j -(jk+1)*stepm;
5680: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5681: if(jl==0){
5682: dh[mi][i]=jk;
5683: bh[mi][i]=0;
5684: }else{ /* We want a negative bias in order to only have interpolation ie
5685: * to avoid the price of an extra matrix product in likelihood */
5686: dh[mi][i]=jk+1;
5687: bh[mi][i]=ju;
5688: }
5689: }else{
5690: if(jl <= -ju){
5691: dh[mi][i]=jk;
5692: bh[mi][i]=jl; /* bias is positive if real duration
5693: * is higher than the multiple of stepm and negative otherwise.
5694: */
5695: }
5696: else{
5697: dh[mi][i]=jk+1;
5698: bh[mi][i]=ju;
5699: }
5700: if(dh[mi][i]==0){
5701: dh[mi][i]=1; /* At least one step */
5702: bh[mi][i]=ju; /* At least one step */
5703: /* 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);*/
5704: }
5705: } /* end if mle */
1.126 brouard 5706: }
5707: } /* end wave */
5708: }
5709: jmean=sum/k;
5710: 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 5711: 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 5712: }
1.126 brouard 5713:
5714: /*********** Tricode ****************************/
1.220 brouard 5715: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5716: {
5717: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5718: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5719: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5720: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5721: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5722: */
1.130 brouard 5723:
1.242 brouard 5724: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5725: int modmaxcovj=0; /* Modality max of covariates j */
5726: int cptcode=0; /* Modality max of covariates j */
5727: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5728:
5729:
1.242 brouard 5730: /* cptcoveff=0; */
5731: /* *cptcov=0; */
1.126 brouard 5732:
1.242 brouard 5733: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5734: for (k=1; k <= maxncov; k++)
5735: for(j=1; j<=2; j++)
5736: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5737:
1.242 brouard 5738: /* Loop on covariates without age and products and no quantitative variable */
5739: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5740: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5741: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5742: switch(Fixed[k]) {
5743: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5744: modmaxcovj=0;
5745: modmincovj=0;
1.242 brouard 5746: 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*/
5747: ij=(int)(covar[Tvar[k]][i]);
5748: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5749: * If product of Vn*Vm, still boolean *:
5750: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5751: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5752: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5753: modality of the nth covariate of individual i. */
5754: if (ij > modmaxcovj)
5755: modmaxcovj=ij;
5756: else if (ij < modmincovj)
5757: modmincovj=ij;
1.287 brouard 5758: if (ij <0 || ij >1 ){
1.311 brouard 5759: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5760: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5761: fflush(ficlog);
5762: exit(1);
1.287 brouard 5763: }
5764: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5765: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5766: exit(1);
5767: }else
5768: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5769: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5770: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5771: /* getting the maximum value of the modality of the covariate
5772: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5773: female ies 1, then modmaxcovj=1.
5774: */
5775: } /* end for loop on individuals i */
5776: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5777: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5778: cptcode=modmaxcovj;
5779: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5780: /*for (i=0; i<=cptcode; i++) {*/
5781: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5782: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5783: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5784: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5785: if( j != -1){
5786: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5787: covariate for which somebody answered excluding
5788: undefined. Usually 2: 0 and 1. */
5789: }
5790: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5791: covariate for which somebody answered including
5792: undefined. Usually 3: -1, 0 and 1. */
5793: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5794: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5795: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5796:
1.242 brouard 5797: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5798: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5799: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5800: /* modmincovj=3; modmaxcovj = 7; */
5801: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5802: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5803: /* defining two dummy variables: variables V1_1 and V1_2.*/
5804: /* nbcode[Tvar[j]][ij]=k; */
5805: /* nbcode[Tvar[j]][1]=0; */
5806: /* nbcode[Tvar[j]][2]=1; */
5807: /* nbcode[Tvar[j]][3]=2; */
5808: /* To be continued (not working yet). */
5809: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5810:
5811: /* 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*/
5812: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5813: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5814: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5815: /*, could be restored in the future */
5816: 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 5817: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5818: break;
5819: }
5820: ij++;
1.287 brouard 5821: 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 5822: cptcode = ij; /* New max modality for covar j */
5823: } /* end of loop on modality i=-1 to 1 or more */
5824: break;
5825: case 1: /* Testing on varying covariate, could be simple and
5826: * should look at waves or product of fixed *
5827: * varying. No time to test -1, assuming 0 and 1 only */
5828: ij=0;
5829: for(i=0; i<=1;i++){
5830: nbcode[Tvar[k]][++ij]=i;
5831: }
5832: break;
5833: default:
5834: break;
5835: } /* end switch */
5836: } /* end dummy test */
1.311 brouard 5837: if(Dummy[k]==1 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5838: 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*/
5839: if(isnan(covar[Tvar[k]][i])){
5840: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5841: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5842: fflush(ficlog);
5843: exit(1);
5844: }
5845: }
5846: }
1.287 brouard 5847: } /* 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 5848:
5849: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5850: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5851: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5852: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5853: 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 */
5854: 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 */
5855: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5856: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5857:
5858: ij=0;
5859: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5860: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5861: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5862: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5863: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5864: /* If product not in single variable we don't print results */
5865: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5866: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5867: 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*/
5868: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5869: 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 */
5870: if(Fixed[k]!=0)
5871: anyvaryingduminmodel=1;
5872: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5873: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5874: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5875: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5876: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5877: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5878: }
5879: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5880: /* ij--; */
5881: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5882: *cptcov=ij; /*Number of total real effective covariates: effective
5883: * because they can be excluded from the model and real
5884: * if in the model but excluded because missing values, but how to get k from ij?*/
5885: for(j=ij+1; j<= cptcovt; j++){
5886: Tvaraff[j]=0;
5887: Tmodelind[j]=0;
5888: }
5889: for(j=ntveff+1; j<= cptcovt; j++){
5890: TmodelInvind[j]=0;
5891: }
5892: /* To be sorted */
5893: ;
5894: }
1.126 brouard 5895:
1.145 brouard 5896:
1.126 brouard 5897: /*********** Health Expectancies ****************/
5898:
1.235 brouard 5899: 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 5900:
5901: {
5902: /* Health expectancies, no variances */
1.164 brouard 5903: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5904: int nhstepma, nstepma; /* Decreasing with age */
5905: double age, agelim, hf;
5906: double ***p3mat;
5907: double eip;
5908:
1.238 brouard 5909: /* pstamp(ficreseij); */
1.126 brouard 5910: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5911: fprintf(ficreseij,"# Age");
5912: for(i=1; i<=nlstate;i++){
5913: for(j=1; j<=nlstate;j++){
5914: fprintf(ficreseij," e%1d%1d ",i,j);
5915: }
5916: fprintf(ficreseij," e%1d. ",i);
5917: }
5918: fprintf(ficreseij,"\n");
5919:
5920:
5921: if(estepm < stepm){
5922: printf ("Problem %d lower than %d\n",estepm, stepm);
5923: }
5924: else hstepm=estepm;
5925: /* We compute the life expectancy from trapezoids spaced every estepm months
5926: * This is mainly to measure the difference between two models: for example
5927: * if stepm=24 months pijx are given only every 2 years and by summing them
5928: * we are calculating an estimate of the Life Expectancy assuming a linear
5929: * progression in between and thus overestimating or underestimating according
5930: * to the curvature of the survival function. If, for the same date, we
5931: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5932: * to compare the new estimate of Life expectancy with the same linear
5933: * hypothesis. A more precise result, taking into account a more precise
5934: * curvature will be obtained if estepm is as small as stepm. */
5935:
5936: /* For example we decided to compute the life expectancy with the smallest unit */
5937: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5938: nhstepm is the number of hstepm from age to agelim
5939: nstepm is the number of stepm from age to agelin.
1.270 brouard 5940: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5941: and note for a fixed period like estepm months */
5942: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5943: survival function given by stepm (the optimization length). Unfortunately it
5944: means that if the survival funtion is printed only each two years of age and if
5945: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5946: results. So we changed our mind and took the option of the best precision.
5947: */
5948: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5949:
5950: agelim=AGESUP;
5951: /* If stepm=6 months */
5952: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5953: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5954:
5955: /* nhstepm age range expressed in number of stepm */
5956: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5957: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5958: /* if (stepm >= YEARM) hstepm=1;*/
5959: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5960: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5961:
5962: for (age=bage; age<=fage; age ++){
5963: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5964: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5965: /* if (stepm >= YEARM) hstepm=1;*/
5966: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5967:
5968: /* If stepm=6 months */
5969: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5970: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5971:
1.235 brouard 5972: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5973:
5974: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5975:
5976: printf("%d|",(int)age);fflush(stdout);
5977: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5978:
5979: /* Computing expectancies */
5980: for(i=1; i<=nlstate;i++)
5981: for(j=1; j<=nlstate;j++)
5982: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5983: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5984:
5985: /* 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]);*/
5986:
5987: }
5988:
5989: fprintf(ficreseij,"%3.0f",age );
5990: for(i=1; i<=nlstate;i++){
5991: eip=0;
5992: for(j=1; j<=nlstate;j++){
5993: eip +=eij[i][j][(int)age];
5994: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5995: }
5996: fprintf(ficreseij,"%9.4f", eip );
5997: }
5998: fprintf(ficreseij,"\n");
5999:
6000: }
6001: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6002: printf("\n");
6003: fprintf(ficlog,"\n");
6004:
6005: }
6006:
1.235 brouard 6007: 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 6008:
6009: {
6010: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6011: to initial status i, ei. .
1.126 brouard 6012: */
6013: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6014: int nhstepma, nstepma; /* Decreasing with age */
6015: double age, agelim, hf;
6016: double ***p3matp, ***p3matm, ***varhe;
6017: double **dnewm,**doldm;
6018: double *xp, *xm;
6019: double **gp, **gm;
6020: double ***gradg, ***trgradg;
6021: int theta;
6022:
6023: double eip, vip;
6024:
6025: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6026: xp=vector(1,npar);
6027: xm=vector(1,npar);
6028: dnewm=matrix(1,nlstate*nlstate,1,npar);
6029: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6030:
6031: pstamp(ficresstdeij);
6032: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6033: fprintf(ficresstdeij,"# Age");
6034: for(i=1; i<=nlstate;i++){
6035: for(j=1; j<=nlstate;j++)
6036: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6037: fprintf(ficresstdeij," e%1d. ",i);
6038: }
6039: fprintf(ficresstdeij,"\n");
6040:
6041: pstamp(ficrescveij);
6042: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6043: fprintf(ficrescveij,"# Age");
6044: for(i=1; i<=nlstate;i++)
6045: for(j=1; j<=nlstate;j++){
6046: cptj= (j-1)*nlstate+i;
6047: for(i2=1; i2<=nlstate;i2++)
6048: for(j2=1; j2<=nlstate;j2++){
6049: cptj2= (j2-1)*nlstate+i2;
6050: if(cptj2 <= cptj)
6051: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6052: }
6053: }
6054: fprintf(ficrescveij,"\n");
6055:
6056: if(estepm < stepm){
6057: printf ("Problem %d lower than %d\n",estepm, stepm);
6058: }
6059: else hstepm=estepm;
6060: /* We compute the life expectancy from trapezoids spaced every estepm months
6061: * This is mainly to measure the difference between two models: for example
6062: * if stepm=24 months pijx are given only every 2 years and by summing them
6063: * we are calculating an estimate of the Life Expectancy assuming a linear
6064: * progression in between and thus overestimating or underestimating according
6065: * to the curvature of the survival function. If, for the same date, we
6066: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6067: * to compare the new estimate of Life expectancy with the same linear
6068: * hypothesis. A more precise result, taking into account a more precise
6069: * curvature will be obtained if estepm is as small as stepm. */
6070:
6071: /* For example we decided to compute the life expectancy with the smallest unit */
6072: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6073: nhstepm is the number of hstepm from age to agelim
6074: nstepm is the number of stepm from age to agelin.
6075: Look at hpijx to understand the reason of that which relies in memory size
6076: and note for a fixed period like estepm months */
6077: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6078: survival function given by stepm (the optimization length). Unfortunately it
6079: means that if the survival funtion is printed only each two years of age and if
6080: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6081: results. So we changed our mind and took the option of the best precision.
6082: */
6083: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6084:
6085: /* If stepm=6 months */
6086: /* nhstepm age range expressed in number of stepm */
6087: agelim=AGESUP;
6088: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6089: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6090: /* if (stepm >= YEARM) hstepm=1;*/
6091: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6092:
6093: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6094: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6095: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6096: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6097: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6098: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6099:
6100: for (age=bage; age<=fage; age ++){
6101: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6102: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6103: /* if (stepm >= YEARM) hstepm=1;*/
6104: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6105:
1.126 brouard 6106: /* If stepm=6 months */
6107: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6108: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6109:
6110: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6111:
1.126 brouard 6112: /* Computing Variances of health expectancies */
6113: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6114: decrease memory allocation */
6115: for(theta=1; theta <=npar; theta++){
6116: for(i=1; i<=npar; i++){
1.222 brouard 6117: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6118: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6119: }
1.235 brouard 6120: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6121: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6122:
1.126 brouard 6123: for(j=1; j<= nlstate; j++){
1.222 brouard 6124: for(i=1; i<=nlstate; i++){
6125: for(h=0; h<=nhstepm-1; h++){
6126: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6127: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6128: }
6129: }
1.126 brouard 6130: }
1.218 brouard 6131:
1.126 brouard 6132: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6133: for(h=0; h<=nhstepm-1; h++){
6134: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6135: }
1.126 brouard 6136: }/* End theta */
6137:
6138:
6139: for(h=0; h<=nhstepm-1; h++)
6140: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6141: for(theta=1; theta <=npar; theta++)
6142: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6143:
1.218 brouard 6144:
1.222 brouard 6145: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6146: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6147: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6148:
1.222 brouard 6149: printf("%d|",(int)age);fflush(stdout);
6150: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6151: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6152: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6153: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6154: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6155: for(ij=1;ij<=nlstate*nlstate;ij++)
6156: for(ji=1;ji<=nlstate*nlstate;ji++)
6157: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6158: }
6159: }
1.320 brouard 6160: /* if((int)age ==50){ */
6161: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6162: /* } */
1.126 brouard 6163: /* Computing expectancies */
1.235 brouard 6164: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6165: for(i=1; i<=nlstate;i++)
6166: for(j=1; j<=nlstate;j++)
1.222 brouard 6167: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6168: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6169:
1.222 brouard 6170: /* 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 6171:
1.222 brouard 6172: }
1.269 brouard 6173:
6174: /* Standard deviation of expectancies ij */
1.126 brouard 6175: fprintf(ficresstdeij,"%3.0f",age );
6176: for(i=1; i<=nlstate;i++){
6177: eip=0.;
6178: vip=0.;
6179: for(j=1; j<=nlstate;j++){
1.222 brouard 6180: eip += eij[i][j][(int)age];
6181: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6182: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6183: 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 6184: }
6185: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6186: }
6187: fprintf(ficresstdeij,"\n");
1.218 brouard 6188:
1.269 brouard 6189: /* Variance of expectancies ij */
1.126 brouard 6190: fprintf(ficrescveij,"%3.0f",age );
6191: for(i=1; i<=nlstate;i++)
6192: for(j=1; j<=nlstate;j++){
1.222 brouard 6193: cptj= (j-1)*nlstate+i;
6194: for(i2=1; i2<=nlstate;i2++)
6195: for(j2=1; j2<=nlstate;j2++){
6196: cptj2= (j2-1)*nlstate+i2;
6197: if(cptj2 <= cptj)
6198: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6199: }
1.126 brouard 6200: }
6201: fprintf(ficrescveij,"\n");
1.218 brouard 6202:
1.126 brouard 6203: }
6204: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6205: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6206: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6207: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6208: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6209: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6210: printf("\n");
6211: fprintf(ficlog,"\n");
1.218 brouard 6212:
1.126 brouard 6213: free_vector(xm,1,npar);
6214: free_vector(xp,1,npar);
6215: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6216: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6217: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6218: }
1.218 brouard 6219:
1.126 brouard 6220: /************ Variance ******************/
1.235 brouard 6221: 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 6222: {
1.279 brouard 6223: /** Variance of health expectancies
6224: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6225: * double **newm;
6226: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6227: */
1.218 brouard 6228:
6229: /* int movingaverage(); */
6230: double **dnewm,**doldm;
6231: double **dnewmp,**doldmp;
6232: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6233: int first=0;
1.218 brouard 6234: int k;
6235: double *xp;
1.279 brouard 6236: double **gp, **gm; /**< for var eij */
6237: double ***gradg, ***trgradg; /**< for var eij */
6238: double **gradgp, **trgradgp; /**< for var p point j */
6239: double *gpp, *gmp; /**< for var p point j */
6240: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6241: double ***p3mat;
6242: double age,agelim, hf;
6243: /* double ***mobaverage; */
6244: int theta;
6245: char digit[4];
6246: char digitp[25];
6247:
6248: char fileresprobmorprev[FILENAMELENGTH];
6249:
6250: if(popbased==1){
6251: if(mobilav!=0)
6252: strcpy(digitp,"-POPULBASED-MOBILAV_");
6253: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6254: }
6255: else
6256: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6257:
1.218 brouard 6258: /* if (mobilav!=0) { */
6259: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6260: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6261: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6262: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6263: /* } */
6264: /* } */
6265:
6266: strcpy(fileresprobmorprev,"PRMORPREV-");
6267: sprintf(digit,"%-d",ij);
6268: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6269: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6270: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6271: strcat(fileresprobmorprev,fileresu);
6272: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6273: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6274: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6275: }
6276: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6277: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6278: pstamp(ficresprobmorprev);
6279: 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 6280: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
6281: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
6282: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
6283: }
6284: for(j=1;j<=cptcoveff;j++)
6285: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
6286: fprintf(ficresprobmorprev,"\n");
6287:
1.218 brouard 6288: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6289: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6290: fprintf(ficresprobmorprev," p.%-d SE",j);
6291: for(i=1; i<=nlstate;i++)
6292: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6293: }
6294: fprintf(ficresprobmorprev,"\n");
6295:
6296: fprintf(ficgp,"\n# Routine varevsij");
6297: fprintf(ficgp,"\nunset title \n");
6298: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6299: 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");
6300: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6301:
1.218 brouard 6302: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6303: pstamp(ficresvij);
6304: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6305: if(popbased==1)
6306: 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);
6307: else
6308: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6309: fprintf(ficresvij,"# Age");
6310: for(i=1; i<=nlstate;i++)
6311: for(j=1; j<=nlstate;j++)
6312: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6313: fprintf(ficresvij,"\n");
6314:
6315: xp=vector(1,npar);
6316: dnewm=matrix(1,nlstate,1,npar);
6317: doldm=matrix(1,nlstate,1,nlstate);
6318: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6319: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6320:
6321: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6322: gpp=vector(nlstate+1,nlstate+ndeath);
6323: gmp=vector(nlstate+1,nlstate+ndeath);
6324: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6325:
1.218 brouard 6326: if(estepm < stepm){
6327: printf ("Problem %d lower than %d\n",estepm, stepm);
6328: }
6329: else hstepm=estepm;
6330: /* For example we decided to compute the life expectancy with the smallest unit */
6331: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6332: nhstepm is the number of hstepm from age to agelim
6333: nstepm is the number of stepm from age to agelim.
6334: Look at function hpijx to understand why because of memory size limitations,
6335: we decided (b) to get a life expectancy respecting the most precise curvature of the
6336: survival function given by stepm (the optimization length). Unfortunately it
6337: means that if the survival funtion is printed every two years of age and if
6338: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6339: results. So we changed our mind and took the option of the best precision.
6340: */
6341: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6342: agelim = AGESUP;
6343: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6344: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6345: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6346: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6347: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6348: gp=matrix(0,nhstepm,1,nlstate);
6349: gm=matrix(0,nhstepm,1,nlstate);
6350:
6351:
6352: for(theta=1; theta <=npar; theta++){
6353: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6354: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6355: }
1.279 brouard 6356: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6357: * returns into prlim .
1.288 brouard 6358: */
1.242 brouard 6359: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6360:
6361: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6362: if (popbased==1) {
6363: if(mobilav ==0){
6364: for(i=1; i<=nlstate;i++)
6365: prlim[i][i]=probs[(int)age][i][ij];
6366: }else{ /* mobilav */
6367: for(i=1; i<=nlstate;i++)
6368: prlim[i][i]=mobaverage[(int)age][i][ij];
6369: }
6370: }
1.295 brouard 6371: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6372: */
6373: 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 6374: /**< 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 6375: * at horizon h in state j including mortality.
6376: */
1.218 brouard 6377: for(j=1; j<= nlstate; j++){
6378: for(h=0; h<=nhstepm; h++){
6379: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6380: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6381: }
6382: }
1.279 brouard 6383: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6384: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6385: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6386: */
6387: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6388: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6389: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6390: }
6391:
6392: /* Again with minus shift */
1.218 brouard 6393:
6394: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6395: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6396:
1.242 brouard 6397: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6398:
6399: if (popbased==1) {
6400: if(mobilav ==0){
6401: for(i=1; i<=nlstate;i++)
6402: prlim[i][i]=probs[(int)age][i][ij];
6403: }else{ /* mobilav */
6404: for(i=1; i<=nlstate;i++)
6405: prlim[i][i]=mobaverage[(int)age][i][ij];
6406: }
6407: }
6408:
1.235 brouard 6409: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6410:
6411: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6412: for(h=0; h<=nhstepm; h++){
6413: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6414: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6415: }
6416: }
6417: /* This for computing probability of death (h=1 means
6418: computed over hstepm matrices product = hstepm*stepm months)
6419: as a weighted average of prlim.
6420: */
6421: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6422: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6423: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6424: }
1.279 brouard 6425: /* end shifting computations */
6426:
6427: /**< Computing gradient matrix at horizon h
6428: */
1.218 brouard 6429: for(j=1; j<= nlstate; j++) /* vareij */
6430: for(h=0; h<=nhstepm; h++){
6431: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6432: }
1.279 brouard 6433: /**< Gradient of overall mortality p.3 (or p.j)
6434: */
6435: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6436: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6437: }
6438:
6439: } /* End theta */
1.279 brouard 6440:
6441: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6442: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6443:
6444: for(h=0; h<=nhstepm; h++) /* veij */
6445: for(j=1; j<=nlstate;j++)
6446: for(theta=1; theta <=npar; theta++)
6447: trgradg[h][j][theta]=gradg[h][theta][j];
6448:
6449: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6450: for(theta=1; theta <=npar; theta++)
6451: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6452: /**< as well as its transposed matrix
6453: */
1.218 brouard 6454:
6455: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6456: for(i=1;i<=nlstate;i++)
6457: for(j=1;j<=nlstate;j++)
6458: vareij[i][j][(int)age] =0.;
1.279 brouard 6459:
6460: /* Computing trgradg by matcov by gradg at age and summing over h
6461: * and k (nhstepm) formula 15 of article
6462: * Lievre-Brouard-Heathcote
6463: */
6464:
1.218 brouard 6465: for(h=0;h<=nhstepm;h++){
6466: for(k=0;k<=nhstepm;k++){
6467: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6468: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6469: for(i=1;i<=nlstate;i++)
6470: for(j=1;j<=nlstate;j++)
6471: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6472: }
6473: }
6474:
1.279 brouard 6475: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6476: * p.j overall mortality formula 49 but computed directly because
6477: * we compute the grad (wix pijx) instead of grad (pijx),even if
6478: * wix is independent of theta.
6479: */
1.218 brouard 6480: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6481: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6482: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6483: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6484: varppt[j][i]=doldmp[j][i];
6485: /* end ppptj */
6486: /* x centered again */
6487:
1.242 brouard 6488: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6489:
6490: if (popbased==1) {
6491: if(mobilav ==0){
6492: for(i=1; i<=nlstate;i++)
6493: prlim[i][i]=probs[(int)age][i][ij];
6494: }else{ /* mobilav */
6495: for(i=1; i<=nlstate;i++)
6496: prlim[i][i]=mobaverage[(int)age][i][ij];
6497: }
6498: }
6499:
6500: /* This for computing probability of death (h=1 means
6501: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6502: as a weighted average of prlim.
6503: */
1.235 brouard 6504: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6505: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6506: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6507: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6508: }
6509: /* end probability of death */
6510:
6511: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6512: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6513: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6514: for(i=1; i<=nlstate;i++){
6515: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6516: }
6517: }
6518: fprintf(ficresprobmorprev,"\n");
6519:
6520: fprintf(ficresvij,"%.0f ",age );
6521: for(i=1; i<=nlstate;i++)
6522: for(j=1; j<=nlstate;j++){
6523: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6524: }
6525: fprintf(ficresvij,"\n");
6526: free_matrix(gp,0,nhstepm,1,nlstate);
6527: free_matrix(gm,0,nhstepm,1,nlstate);
6528: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6529: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6530: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6531: } /* End age */
6532: free_vector(gpp,nlstate+1,nlstate+ndeath);
6533: free_vector(gmp,nlstate+1,nlstate+ndeath);
6534: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6535: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6536: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6537: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6538: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6539: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6540: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6541: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6542: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6543: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6544: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6545: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6546: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6547: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6548: 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);
6549: /* 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 6550: */
1.218 brouard 6551: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6552: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6553:
1.218 brouard 6554: free_vector(xp,1,npar);
6555: free_matrix(doldm,1,nlstate,1,nlstate);
6556: free_matrix(dnewm,1,nlstate,1,npar);
6557: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6558: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6559: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6560: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6561: fclose(ficresprobmorprev);
6562: fflush(ficgp);
6563: fflush(fichtm);
6564: } /* end varevsij */
1.126 brouard 6565:
6566: /************ Variance of prevlim ******************/
1.269 brouard 6567: 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 6568: {
1.205 brouard 6569: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6570: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6571:
1.268 brouard 6572: double **dnewmpar,**doldm;
1.126 brouard 6573: int i, j, nhstepm, hstepm;
6574: double *xp;
6575: double *gp, *gm;
6576: double **gradg, **trgradg;
1.208 brouard 6577: double **mgm, **mgp;
1.126 brouard 6578: double age,agelim;
6579: int theta;
6580:
6581: pstamp(ficresvpl);
1.288 brouard 6582: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6583: fprintf(ficresvpl,"# Age ");
6584: if(nresult >=1)
6585: fprintf(ficresvpl," Result# ");
1.126 brouard 6586: for(i=1; i<=nlstate;i++)
6587: fprintf(ficresvpl," %1d-%1d",i,i);
6588: fprintf(ficresvpl,"\n");
6589:
6590: xp=vector(1,npar);
1.268 brouard 6591: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6592: doldm=matrix(1,nlstate,1,nlstate);
6593:
6594: hstepm=1*YEARM; /* Every year of age */
6595: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6596: agelim = AGESUP;
6597: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6598: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6599: if (stepm >= YEARM) hstepm=1;
6600: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6601: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6602: mgp=matrix(1,npar,1,nlstate);
6603: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6604: gp=vector(1,nlstate);
6605: gm=vector(1,nlstate);
6606:
6607: for(theta=1; theta <=npar; theta++){
6608: for(i=1; i<=npar; i++){ /* Computes gradient */
6609: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6610: }
1.288 brouard 6611: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6612: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6613: /* else */
6614: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6615: for(i=1;i<=nlstate;i++){
1.126 brouard 6616: gp[i] = prlim[i][i];
1.208 brouard 6617: mgp[theta][i] = prlim[i][i];
6618: }
1.126 brouard 6619: for(i=1; i<=npar; i++) /* Computes gradient */
6620: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6621: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6622: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6623: /* else */
6624: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6625: for(i=1;i<=nlstate;i++){
1.126 brouard 6626: gm[i] = prlim[i][i];
1.208 brouard 6627: mgm[theta][i] = prlim[i][i];
6628: }
1.126 brouard 6629: for(i=1;i<=nlstate;i++)
6630: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6631: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6632: } /* End theta */
6633:
6634: trgradg =matrix(1,nlstate,1,npar);
6635:
6636: for(j=1; j<=nlstate;j++)
6637: for(theta=1; theta <=npar; theta++)
6638: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6639: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6640: /* printf("\nmgm mgp %d ",(int)age); */
6641: /* for(j=1; j<=nlstate;j++){ */
6642: /* printf(" %d ",j); */
6643: /* for(theta=1; theta <=npar; theta++) */
6644: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6645: /* printf("\n "); */
6646: /* } */
6647: /* } */
6648: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6649: /* printf("\n gradg %d ",(int)age); */
6650: /* for(j=1; j<=nlstate;j++){ */
6651: /* printf("%d ",j); */
6652: /* for(theta=1; theta <=npar; theta++) */
6653: /* printf("%d %lf ",theta,gradg[theta][j]); */
6654: /* printf("\n "); */
6655: /* } */
6656: /* } */
1.126 brouard 6657:
6658: for(i=1;i<=nlstate;i++)
6659: varpl[i][(int)age] =0.;
1.209 brouard 6660: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6661: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6662: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6663: }else{
1.268 brouard 6664: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6665: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6666: }
1.126 brouard 6667: for(i=1;i<=nlstate;i++)
6668: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6669:
6670: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6671: if(nresult >=1)
6672: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6673: for(i=1; i<=nlstate;i++){
1.126 brouard 6674: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6675: /* for(j=1;j<=nlstate;j++) */
6676: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6677: }
1.126 brouard 6678: fprintf(ficresvpl,"\n");
6679: free_vector(gp,1,nlstate);
6680: free_vector(gm,1,nlstate);
1.208 brouard 6681: free_matrix(mgm,1,npar,1,nlstate);
6682: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6683: free_matrix(gradg,1,npar,1,nlstate);
6684: free_matrix(trgradg,1,nlstate,1,npar);
6685: } /* End age */
6686:
6687: free_vector(xp,1,npar);
6688: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6689: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6690:
6691: }
6692:
6693:
6694: /************ Variance of backprevalence limit ******************/
1.269 brouard 6695: 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 6696: {
6697: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6698: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6699:
6700: double **dnewmpar,**doldm;
6701: int i, j, nhstepm, hstepm;
6702: double *xp;
6703: double *gp, *gm;
6704: double **gradg, **trgradg;
6705: double **mgm, **mgp;
6706: double age,agelim;
6707: int theta;
6708:
6709: pstamp(ficresvbl);
6710: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6711: fprintf(ficresvbl,"# Age ");
6712: if(nresult >=1)
6713: fprintf(ficresvbl," Result# ");
6714: for(i=1; i<=nlstate;i++)
6715: fprintf(ficresvbl," %1d-%1d",i,i);
6716: fprintf(ficresvbl,"\n");
6717:
6718: xp=vector(1,npar);
6719: dnewmpar=matrix(1,nlstate,1,npar);
6720: doldm=matrix(1,nlstate,1,nlstate);
6721:
6722: hstepm=1*YEARM; /* Every year of age */
6723: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6724: agelim = AGEINF;
6725: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6726: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6727: if (stepm >= YEARM) hstepm=1;
6728: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6729: gradg=matrix(1,npar,1,nlstate);
6730: mgp=matrix(1,npar,1,nlstate);
6731: mgm=matrix(1,npar,1,nlstate);
6732: gp=vector(1,nlstate);
6733: gm=vector(1,nlstate);
6734:
6735: for(theta=1; theta <=npar; theta++){
6736: for(i=1; i<=npar; i++){ /* Computes gradient */
6737: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6738: }
6739: if(mobilavproj > 0 )
6740: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6741: else
6742: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6743: for(i=1;i<=nlstate;i++){
6744: gp[i] = bprlim[i][i];
6745: mgp[theta][i] = bprlim[i][i];
6746: }
6747: for(i=1; i<=npar; i++) /* Computes gradient */
6748: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6749: if(mobilavproj > 0 )
6750: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6751: else
6752: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6753: for(i=1;i<=nlstate;i++){
6754: gm[i] = bprlim[i][i];
6755: mgm[theta][i] = bprlim[i][i];
6756: }
6757: for(i=1;i<=nlstate;i++)
6758: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6759: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6760: } /* End theta */
6761:
6762: trgradg =matrix(1,nlstate,1,npar);
6763:
6764: for(j=1; j<=nlstate;j++)
6765: for(theta=1; theta <=npar; theta++)
6766: trgradg[j][theta]=gradg[theta][j];
6767: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6768: /* printf("\nmgm mgp %d ",(int)age); */
6769: /* for(j=1; j<=nlstate;j++){ */
6770: /* printf(" %d ",j); */
6771: /* for(theta=1; theta <=npar; theta++) */
6772: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6773: /* printf("\n "); */
6774: /* } */
6775: /* } */
6776: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6777: /* printf("\n gradg %d ",(int)age); */
6778: /* for(j=1; j<=nlstate;j++){ */
6779: /* printf("%d ",j); */
6780: /* for(theta=1; theta <=npar; theta++) */
6781: /* printf("%d %lf ",theta,gradg[theta][j]); */
6782: /* printf("\n "); */
6783: /* } */
6784: /* } */
6785:
6786: for(i=1;i<=nlstate;i++)
6787: varbpl[i][(int)age] =0.;
6788: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6789: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6790: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6791: }else{
6792: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6793: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6794: }
6795: for(i=1;i<=nlstate;i++)
6796: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6797:
6798: fprintf(ficresvbl,"%.0f ",age );
6799: if(nresult >=1)
6800: fprintf(ficresvbl,"%d ",nres );
6801: for(i=1; i<=nlstate;i++)
6802: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6803: fprintf(ficresvbl,"\n");
6804: free_vector(gp,1,nlstate);
6805: free_vector(gm,1,nlstate);
6806: free_matrix(mgm,1,npar,1,nlstate);
6807: free_matrix(mgp,1,npar,1,nlstate);
6808: free_matrix(gradg,1,npar,1,nlstate);
6809: free_matrix(trgradg,1,nlstate,1,npar);
6810: } /* End age */
6811:
6812: free_vector(xp,1,npar);
6813: free_matrix(doldm,1,nlstate,1,npar);
6814: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6815:
6816: }
6817:
6818: /************ Variance of one-step probabilities ******************/
6819: 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 6820: {
6821: int i, j=0, k1, l1, tj;
6822: int k2, l2, j1, z1;
6823: int k=0, l;
6824: int first=1, first1, first2;
1.326 brouard 6825: int nres=0; /* New */
1.222 brouard 6826: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6827: double **dnewm,**doldm;
6828: double *xp;
6829: double *gp, *gm;
6830: double **gradg, **trgradg;
6831: double **mu;
6832: double age, cov[NCOVMAX+1];
6833: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6834: int theta;
6835: char fileresprob[FILENAMELENGTH];
6836: char fileresprobcov[FILENAMELENGTH];
6837: char fileresprobcor[FILENAMELENGTH];
6838: double ***varpij;
6839:
6840: strcpy(fileresprob,"PROB_");
6841: strcat(fileresprob,fileres);
6842: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6843: printf("Problem with resultfile: %s\n", fileresprob);
6844: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6845: }
6846: strcpy(fileresprobcov,"PROBCOV_");
6847: strcat(fileresprobcov,fileresu);
6848: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6849: printf("Problem with resultfile: %s\n", fileresprobcov);
6850: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6851: }
6852: strcpy(fileresprobcor,"PROBCOR_");
6853: strcat(fileresprobcor,fileresu);
6854: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6855: printf("Problem with resultfile: %s\n", fileresprobcor);
6856: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6857: }
6858: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6859: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6860: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6861: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6862: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6863: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6864: pstamp(ficresprob);
6865: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6866: fprintf(ficresprob,"# Age");
6867: pstamp(ficresprobcov);
6868: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6869: fprintf(ficresprobcov,"# Age");
6870: pstamp(ficresprobcor);
6871: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6872: fprintf(ficresprobcor,"# Age");
1.126 brouard 6873:
6874:
1.222 brouard 6875: for(i=1; i<=nlstate;i++)
6876: for(j=1; j<=(nlstate+ndeath);j++){
6877: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6878: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6879: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6880: }
6881: /* fprintf(ficresprob,"\n");
6882: fprintf(ficresprobcov,"\n");
6883: fprintf(ficresprobcor,"\n");
6884: */
6885: xp=vector(1,npar);
6886: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6887: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6888: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6889: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6890: first=1;
6891: fprintf(ficgp,"\n# Routine varprob");
6892: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6893: fprintf(fichtm,"\n");
6894:
1.288 brouard 6895: 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 6896: 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);
6897: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6898: and drawn. It helps understanding how is the covariance between two incidences.\
6899: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6900: 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 6901: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6902: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6903: standard deviations wide on each axis. <br>\
6904: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6905: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6906: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6907:
1.222 brouard 6908: cov[1]=1;
6909: /* tj=cptcoveff; */
1.225 brouard 6910: tj = (int) pow(2,cptcoveff);
1.222 brouard 6911: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6912: j1=0;
1.224 brouard 6913: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.326 brouard 6914: for(nres=1;nres <=1; nres++){ /* For each resultline */
6915: /* for(nres=1;nres <=nresult; nres++){ /\* For each resultline *\/ */
1.222 brouard 6916: if (cptcovn>0) {
6917: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6918: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6919: fprintf(ficresprob, "**********\n#\n");
6920: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6921: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6922: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6923:
1.222 brouard 6924: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6925: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6926: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6927:
6928:
1.222 brouard 6929: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.319 brouard 6930: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]); */
6931: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtmcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6932: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6933:
1.222 brouard 6934: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6935: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6936: fprintf(ficresprobcor, "**********\n#");
6937: if(invalidvarcomb[j1]){
6938: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6939: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6940: continue;
6941: }
6942: }
6943: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6944: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6945: gp=vector(1,(nlstate)*(nlstate+ndeath));
6946: gm=vector(1,(nlstate)*(nlstate+ndeath));
6947: for (age=bage; age<=fage; age ++){
6948: cov[2]=age;
6949: if(nagesqr==1)
6950: cov[3]= age*age;
1.326 brouard 6951: /* for (k=1; k<=cptcovn;k++) { */
6952: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; */
6953: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
6954: /* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates */
6955: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,k)];
1.222 brouard 6956: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6957: * 1 1 1 1 1
6958: * 2 2 1 1 1
6959: * 3 1 2 1 1
6960: */
6961: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6962: }
1.319 brouard 6963: /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 */
6964: /* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] */
6965: /*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
1.326 brouard 6966: for (k=1; k<=cptcovage;k++){ /* For product with age */
6967: if(Dummy[Tage[k]]==2){ /* dummy with age */
6968: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,k)]*cov[2];
6969: /* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
6970: } else if(Dummy[Tage[k]]==3){ /* quantitative with age */
1.327 ! brouard 6971: printf("Internal IMaCh error, don't know which value for quantitative covariate with age, Tage[k]%d, k=%d, Tvar[Tage[k]]=V%d, age=%d\n",Tage[k],k ,Tvar[Tage[k]], (int)cov[2]);
! 6972: exit(1);
! 6973: /* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\* Using the mean of quantitative variable Tvar[Tage[k]] /\* Tqresult[nres][k]; *\/ */
1.326 brouard 6974: /* cov[++k1]=Tqresult[nres][k]; */
6975: }
6976: /* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
6977: }
6978: for (k=1; k<=cptcovprod;k++){/* For product without age */
6979: if(Dummy[Tvard[k][1]==0]){
6980: if(Dummy[Tvard[k][2]==0]){
6981: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,k)] * nbcode[Tvard[k][2]][codtabm(j1,k)];
6982: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
6983: }else{ /* Should we use the mean of the quantitative variables? */
6984: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,k)] * Tqresult[nres][k];
6985: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; */
6986: }
6987: }else{
6988: if(Dummy[Tvard[k][2]==0]){
6989: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,k)] * Tqinvresult[nres][Tvard[k][1]];
6990: /* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; */
6991: }else{
6992: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
6993: /* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
6994: }
6995: }
6996: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
6997: }
6998: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 6999: for(theta=1; theta <=npar; theta++){
7000: for(i=1; i<=npar; i++)
7001: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7002:
1.222 brouard 7003: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7004:
1.222 brouard 7005: k=0;
7006: for(i=1; i<= (nlstate); i++){
7007: for(j=1; j<=(nlstate+ndeath);j++){
7008: k=k+1;
7009: gp[k]=pmmij[i][j];
7010: }
7011: }
1.220 brouard 7012:
1.222 brouard 7013: for(i=1; i<=npar; i++)
7014: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7015:
1.222 brouard 7016: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7017: k=0;
7018: for(i=1; i<=(nlstate); i++){
7019: for(j=1; j<=(nlstate+ndeath);j++){
7020: k=k+1;
7021: gm[k]=pmmij[i][j];
7022: }
7023: }
1.220 brouard 7024:
1.222 brouard 7025: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7026: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7027: }
1.126 brouard 7028:
1.222 brouard 7029: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7030: for(theta=1; theta <=npar; theta++)
7031: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7032:
1.222 brouard 7033: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7034: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7035:
1.222 brouard 7036: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7037:
1.222 brouard 7038: k=0;
7039: for(i=1; i<=(nlstate); i++){
7040: for(j=1; j<=(nlstate+ndeath);j++){
7041: k=k+1;
7042: mu[k][(int) age]=pmmij[i][j];
7043: }
7044: }
7045: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7046: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7047: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7048:
1.222 brouard 7049: /*printf("\n%d ",(int)age);
7050: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7051: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7052: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7053: }*/
1.220 brouard 7054:
1.222 brouard 7055: fprintf(ficresprob,"\n%d ",(int)age);
7056: fprintf(ficresprobcov,"\n%d ",(int)age);
7057: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7058:
1.222 brouard 7059: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7060: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7061: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7062: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7063: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7064: }
7065: i=0;
7066: for (k=1; k<=(nlstate);k++){
7067: for (l=1; l<=(nlstate+ndeath);l++){
7068: i++;
7069: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7070: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7071: for (j=1; j<=i;j++){
7072: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7073: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7074: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7075: }
7076: }
7077: }/* end of loop for state */
7078: } /* end of loop for age */
7079: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7080: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7081: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7082: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7083:
7084: /* Confidence intervalle of pij */
7085: /*
7086: fprintf(ficgp,"\nunset parametric;unset label");
7087: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7088: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7089: 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);
7090: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7091: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7092: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7093: */
7094:
7095: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7096: first1=1;first2=2;
7097: for (k2=1; k2<=(nlstate);k2++){
7098: for (l2=1; l2<=(nlstate+ndeath);l2++){
7099: if(l2==k2) continue;
7100: j=(k2-1)*(nlstate+ndeath)+l2;
7101: for (k1=1; k1<=(nlstate);k1++){
7102: for (l1=1; l1<=(nlstate+ndeath);l1++){
7103: if(l1==k1) continue;
7104: i=(k1-1)*(nlstate+ndeath)+l1;
7105: if(i<=j) continue;
7106: for (age=bage; age<=fage; age ++){
7107: if ((int)age %5==0){
7108: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7109: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7110: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7111: mu1=mu[i][(int) age]/stepm*YEARM ;
7112: mu2=mu[j][(int) age]/stepm*YEARM;
7113: c12=cv12/sqrt(v1*v2);
7114: /* Computing eigen value of matrix of covariance */
7115: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7116: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7117: if ((lc2 <0) || (lc1 <0) ){
7118: if(first2==1){
7119: first1=0;
7120: 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);
7121: }
7122: 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);
7123: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7124: /* lc2=fabs(lc2); */
7125: }
1.220 brouard 7126:
1.222 brouard 7127: /* Eigen vectors */
1.280 brouard 7128: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7129: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7130: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7131: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7132: }else
7133: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7134: /*v21=sqrt(1.-v11*v11); *//* error */
7135: v21=(lc1-v1)/cv12*v11;
7136: v12=-v21;
7137: v22=v11;
7138: tnalp=v21/v11;
7139: if(first1==1){
7140: first1=0;
7141: 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);
7142: }
7143: 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);
7144: /*printf(fignu*/
7145: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7146: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7147: if(first==1){
7148: first=0;
7149: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7150: fprintf(ficgp,"\nset parametric;unset label");
7151: 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);
7152: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7153: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7154: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7155: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7156: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7157: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7158: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7159: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7160: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7161: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7162: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7163: 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 7164: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7165: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7166: }else{
7167: first=0;
7168: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7169: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7170: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7171: 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 7172: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7173: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7174: }/* if first */
7175: } /* age mod 5 */
7176: } /* end loop age */
7177: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7178: first=1;
7179: } /*l12 */
7180: } /* k12 */
7181: } /*l1 */
7182: }/* k1 */
1.326 brouard 7183: } /* loop on nres */
1.222 brouard 7184: } /* loop on combination of covariates j1 */
7185: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7186: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7187: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7188: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7189: free_vector(xp,1,npar);
7190: fclose(ficresprob);
7191: fclose(ficresprobcov);
7192: fclose(ficresprobcor);
7193: fflush(ficgp);
7194: fflush(fichtmcov);
7195: }
1.126 brouard 7196:
7197:
7198: /******************* Printing html file ***********/
1.201 brouard 7199: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7200: int lastpass, int stepm, int weightopt, char model[],\
7201: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7202: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7203: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7204: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7205: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7206: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7207: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7208: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7209: </ul>");
1.319 brouard 7210: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7211: /* </ul>", model); */
1.214 brouard 7212: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7213: 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",
7214: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
7215: 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 7216: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7217: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7218: fprintf(fichtm,"\
7219: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7220: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7221: fprintf(fichtm,"\
1.217 brouard 7222: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7223: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7224: fprintf(fichtm,"\
1.288 brouard 7225: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7226: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7227: fprintf(fichtm,"\
1.288 brouard 7228: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7229: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7230: fprintf(fichtm,"\
1.211 brouard 7231: - (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 7232: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7233: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7234: if(prevfcast==1){
7235: fprintf(fichtm,"\
7236: - Prevalence projections by age and states: \
1.201 brouard 7237: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7238: }
1.126 brouard 7239:
7240:
1.225 brouard 7241: m=pow(2,cptcoveff);
1.222 brouard 7242: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7243:
1.317 brouard 7244: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7245:
7246: jj1=0;
7247:
7248: fprintf(fichtm," \n<ul>");
7249: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7250: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7251: if(m != 1 && TKresult[nres]!= k1)
7252: continue;
7253: jj1++;
7254: if (cptcovn > 0) {
7255: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
7256: for (cpt=1; cpt<=cptcoveff;cpt++){
7257: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7258: }
7259: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7260: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7261: }
7262: fprintf(fichtm,"\">");
7263:
7264: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7265: fprintf(fichtm,"************ Results for covariates");
7266: for (cpt=1; cpt<=cptcoveff;cpt++){
7267: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7268: }
7269: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7270: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7271: }
7272: if(invalidvarcomb[k1]){
7273: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7274: continue;
7275: }
7276: fprintf(fichtm,"</a></li>");
7277: } /* cptcovn >0 */
7278: }
1.317 brouard 7279: fprintf(fichtm," \n</ul>");
1.264 brouard 7280:
1.222 brouard 7281: jj1=0;
1.237 brouard 7282:
7283: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 7284: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 7285: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7286: continue;
1.220 brouard 7287:
1.222 brouard 7288: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7289: jj1++;
7290: if (cptcovn > 0) {
1.264 brouard 7291: fprintf(fichtm,"\n<p><a name=\"rescov");
7292: for (cpt=1; cpt<=cptcoveff;cpt++){
7293: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7294: }
7295: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7296: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7297: }
7298: fprintf(fichtm,"\"</a>");
7299:
1.222 brouard 7300: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7301: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 7302: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7303: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
7304: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7305: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7306: }
1.237 brouard 7307: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7308: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7309: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
7310: }
7311:
1.230 brouard 7312: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.321 brouard 7313: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7314: if(invalidvarcomb[k1]){
7315: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7316: printf("\nCombination (%d) ignored because no cases \n",k1);
7317: continue;
7318: }
7319: }
7320: /* aij, bij */
1.259 brouard 7321: 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 7322: <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 7323: /* Pij */
1.241 brouard 7324: 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> \
7325: <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 7326: /* Quasi-incidences */
7327: 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 7328: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7329: 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 7330: 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> \
7331: <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 7332: /* Survival functions (period) in state j */
7333: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7334: 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 7335: <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 7336: }
7337: /* State specific survival functions (period) */
7338: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7339: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7340: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 7341: <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 7342: }
1.288 brouard 7343: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7344: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7345: 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> \
7346: <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 7347: }
1.296 brouard 7348: if(prevbcast==1){
1.288 brouard 7349: /* Backward prevalence in each health state */
1.222 brouard 7350: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7351: 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 7352: <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 7353: }
1.217 brouard 7354: }
1.222 brouard 7355: if(prevfcast==1){
1.288 brouard 7356: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7357: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7358: 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);
7359: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7360: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7361: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7362: }
7363: }
1.296 brouard 7364: if(prevbcast==1){
1.268 brouard 7365: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7366: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7367: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7368: 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 \
7369: 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 7370: 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);
7371: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7372: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7373: }
7374: }
1.220 brouard 7375:
1.222 brouard 7376: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7377: 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);
7378: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7379: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7380: }
7381: /* } /\* end i1 *\/ */
7382: }/* End k1 */
7383: fprintf(fichtm,"</ul>");
1.126 brouard 7384:
1.222 brouard 7385: fprintf(fichtm,"\
1.126 brouard 7386: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7387: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7388: - 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 7389: But because parameters are usually highly correlated (a higher incidence of disability \
7390: and a higher incidence of recovery can give very close observed transition) it might \
7391: be very useful to look not only at linear confidence intervals estimated from the \
7392: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7393: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7394: covariance matrix of the one-step probabilities. \
7395: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7396:
1.222 brouard 7397: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7398: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7399: fprintf(fichtm,"\
1.126 brouard 7400: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7401: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7402:
1.222 brouard 7403: fprintf(fichtm,"\
1.126 brouard 7404: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7405: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7406: fprintf(fichtm,"\
1.126 brouard 7407: - 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): \
7408: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7409: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7410: fprintf(fichtm,"\
1.126 brouard 7411: - (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): \
7412: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7413: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7414: fprintf(fichtm,"\
1.288 brouard 7415: - 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 7416: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7417: fprintf(fichtm,"\
1.128 brouard 7418: - 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 7419: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7420: fprintf(fichtm,"\
1.288 brouard 7421: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7422: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7423:
7424: /* if(popforecast==1) fprintf(fichtm,"\n */
7425: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7426: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7427: /* <br>",fileres,fileres,fileres,fileres); */
7428: /* else */
7429: /* 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 7430: fflush(fichtm);
1.126 brouard 7431:
1.225 brouard 7432: m=pow(2,cptcoveff);
1.222 brouard 7433: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7434:
1.317 brouard 7435: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
7436:
7437: jj1=0;
7438:
7439: fprintf(fichtm," \n<ul>");
7440: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7441: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7442: if(m != 1 && TKresult[nres]!= k1)
7443: continue;
7444: jj1++;
7445: if (cptcovn > 0) {
7446: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
7447: for (cpt=1; cpt<=cptcoveff;cpt++){
7448: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7449: }
7450: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7451: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7452: }
7453: fprintf(fichtm,"\">");
7454:
7455: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7456: fprintf(fichtm,"************ Results for covariates");
7457: for (cpt=1; cpt<=cptcoveff;cpt++){
7458: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7459: }
7460: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7461: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7462: }
7463: if(invalidvarcomb[k1]){
7464: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7465: continue;
7466: }
7467: fprintf(fichtm,"</a></li>");
7468: } /* cptcovn >0 */
7469: }
7470: fprintf(fichtm," \n</ul>");
7471:
1.222 brouard 7472: jj1=0;
1.237 brouard 7473:
1.241 brouard 7474: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7475: for(k1=1; k1<=m;k1++){
1.253 brouard 7476: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7477: continue;
1.222 brouard 7478: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7479: jj1++;
1.126 brouard 7480: if (cptcovn > 0) {
1.317 brouard 7481: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
7482: for (cpt=1; cpt<=cptcoveff;cpt++){
7483: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7484: }
7485: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7486: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7487: }
7488: fprintf(fichtm,"\"</a>");
7489:
1.126 brouard 7490: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.317 brouard 7491: for (cpt=1; cpt<=cptcoveff;cpt++){ /**< cptcoveff number of variables */
1.237 brouard 7492: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
1.317 brouard 7493: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
1.237 brouard 7494: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 7495: }
1.237 brouard 7496: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7497: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7498: }
7499:
1.321 brouard 7500: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 7501:
1.222 brouard 7502: if(invalidvarcomb[k1]){
7503: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7504: continue;
7505: }
1.126 brouard 7506: }
7507: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7508: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 7509: 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);
7510: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
7511: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 7512: }
7513: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 7514: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 7515: true period expectancies (those weighted with period prevalences are also\
7516: drawn in addition to the population based expectancies computed using\
1.314 brouard 7517: 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);
7518: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
7519: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7520: /* } /\* end i1 *\/ */
7521: }/* End k1 */
1.241 brouard 7522: }/* End nres */
1.222 brouard 7523: fprintf(fichtm,"</ul>");
7524: fflush(fichtm);
1.126 brouard 7525: }
7526:
7527: /******************* Gnuplot file **************/
1.296 brouard 7528: 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 7529:
7530: char dirfileres[132],optfileres[132];
1.264 brouard 7531: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7532: 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 7533: int lv=0, vlv=0, kl=0;
1.130 brouard 7534: int ng=0;
1.201 brouard 7535: int vpopbased;
1.223 brouard 7536: int ioffset; /* variable offset for columns */
1.270 brouard 7537: int iyearc=1; /* variable column for year of projection */
7538: int iagec=1; /* variable column for age of projection */
1.235 brouard 7539: int nres=0; /* Index of resultline */
1.266 brouard 7540: int istart=1; /* For starting graphs in projections */
1.219 brouard 7541:
1.126 brouard 7542: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7543: /* printf("Problem with file %s",optionfilegnuplot); */
7544: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7545: /* } */
7546:
7547: /*#ifdef windows */
7548: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7549: /*#endif */
1.225 brouard 7550: m=pow(2,cptcoveff);
1.126 brouard 7551:
1.274 brouard 7552: /* diagram of the model */
7553: fprintf(ficgp,"\n#Diagram of the model \n");
7554: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7555: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7556: 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);
7557:
7558: 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);
7559: fprintf(ficgp,"\n#show arrow\nunset label\n");
7560: 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);
7561: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7562: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7563: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7564: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7565:
1.202 brouard 7566: /* Contribution to likelihood */
7567: /* Plot the probability implied in the likelihood */
1.223 brouard 7568: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7569: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7570: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7571: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7572: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7573: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7574: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7575: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7576: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7577: 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));
7578: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7579: 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));
7580: for (i=1; i<= nlstate ; i ++) {
7581: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7582: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7583: 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);
7584: for (j=2; j<= nlstate+ndeath ; j ++) {
7585: 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);
7586: }
7587: fprintf(ficgp,";\nset out; unset ylabel;\n");
7588: }
7589: /* 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 */
7590: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7591: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7592: fprintf(ficgp,"\nset out;unset log\n");
7593: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7594:
1.126 brouard 7595: strcpy(dirfileres,optionfilefiname);
7596: strcpy(optfileres,"vpl");
1.223 brouard 7597: /* 1eme*/
1.238 brouard 7598: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7599: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7600: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7601: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7602: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7603: continue;
7604: /* We are interested in selected combination by the resultline */
1.246 brouard 7605: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7606: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7607: strcpy(gplotlabel,"(");
1.238 brouard 7608: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7609: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7610: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7611: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7612: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7613: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7614: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7615: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7616: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7617: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7618: }
7619: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7620: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7621: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7622: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7623: }
7624: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7625: /* printf("\n#\n"); */
1.238 brouard 7626: fprintf(ficgp,"\n#\n");
7627: if(invalidvarcomb[k1]){
1.260 brouard 7628: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7629: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7630: continue;
7631: }
1.235 brouard 7632:
1.241 brouard 7633: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7634: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7635: /* 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 7636: fprintf(ficgp,"set title \"Alive state %d %s model=%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 7637: 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);
7638: /* 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); */
7639: /* k1-1 error should be nres-1*/
1.238 brouard 7640: for (i=1; i<= nlstate ; i ++) {
7641: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7642: else fprintf(ficgp," %%*lf (%%*lf)");
7643: }
1.288 brouard 7644: 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 7645: for (i=1; i<= nlstate ; i ++) {
7646: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7647: else fprintf(ficgp," %%*lf (%%*lf)");
7648: }
1.260 brouard 7649: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7650: for (i=1; i<= nlstate ; i ++) {
7651: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7652: else fprintf(ficgp," %%*lf (%%*lf)");
7653: }
1.265 brouard 7654: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7655:
7656: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7657: if(cptcoveff ==0){
1.271 brouard 7658: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7659: }else{
7660: kl=0;
7661: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7662: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7663: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7664: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7665: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7666: vlv= nbcode[Tvaraff[k]][lv];
7667: kl++;
7668: /* 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 *\/ */
7669: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7670: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7671: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7672: if(k==cptcoveff){
7673: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7674: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7675: }else{
7676: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7677: kl++;
7678: }
7679: } /* end covariate */
7680: } /* end if no covariate */
7681:
1.296 brouard 7682: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7683: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7684: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7685: if(cptcoveff ==0){
1.245 brouard 7686: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7687: }else{
7688: kl=0;
7689: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7690: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7691: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7692: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7693: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7694: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7695: kl++;
1.238 brouard 7696: /* 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 *\/ */
7697: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7698: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7699: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7700: if(k==cptcoveff){
1.245 brouard 7701: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7702: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7703: }else{
7704: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7705: kl++;
7706: }
7707: } /* end covariate */
7708: } /* end if no covariate */
1.296 brouard 7709: if(prevbcast == 1){
1.268 brouard 7710: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7711: /* k1-1 error should be nres-1*/
7712: for (i=1; i<= nlstate ; i ++) {
7713: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7714: else fprintf(ficgp," %%*lf (%%*lf)");
7715: }
1.271 brouard 7716: 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 7717: for (i=1; i<= nlstate ; i ++) {
7718: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7719: else fprintf(ficgp," %%*lf (%%*lf)");
7720: }
1.276 brouard 7721: 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 7722: for (i=1; i<= nlstate ; i ++) {
7723: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7724: else fprintf(ficgp," %%*lf (%%*lf)");
7725: }
1.274 brouard 7726: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7727: } /* end if backprojcast */
1.296 brouard 7728: } /* end if prevbcast */
1.276 brouard 7729: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7730: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7731: } /* nres */
1.201 brouard 7732: } /* k1 */
7733: } /* cpt */
1.235 brouard 7734:
7735:
1.126 brouard 7736: /*2 eme*/
1.238 brouard 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: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7742: strcpy(gplotlabel,"(");
1.238 brouard 7743: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7744: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7745: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7746: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7747: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7748: vlv= nbcode[Tvaraff[k]][lv];
7749: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7750: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7751: }
1.237 brouard 7752: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7753: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7754: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 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),")");
1.211 brouard 7759: fprintf(ficgp,"\n#\n");
1.223 brouard 7760: if(invalidvarcomb[k1]){
7761: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7762: continue;
7763: }
1.219 brouard 7764:
1.241 brouard 7765: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7766: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7767: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7768: if(vpopbased==0){
1.238 brouard 7769: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7770: }else
1.238 brouard 7771: fprintf(ficgp,"\nreplot ");
7772: for (i=1; i<= nlstate+1 ; i ++) {
7773: k=2*i;
1.261 brouard 7774: 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 7775: for (j=1; j<= nlstate+1 ; j ++) {
7776: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7777: else fprintf(ficgp," %%*lf (%%*lf)");
7778: }
7779: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7780: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7781: 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 7782: for (j=1; j<= nlstate+1 ; j ++) {
7783: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7784: else fprintf(ficgp," %%*lf (%%*lf)");
7785: }
7786: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7787: 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 7788: for (j=1; j<= nlstate+1 ; j ++) {
7789: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7790: else fprintf(ficgp," %%*lf (%%*lf)");
7791: }
7792: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7793: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7794: } /* state */
7795: } /* vpopbased */
1.264 brouard 7796: 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 7797: } /* end nres */
7798: } /* k1 end 2 eme*/
7799:
7800:
7801: /*3eme*/
7802: for (k1=1; k1<= m ; k1 ++){
7803: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7804: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7805: continue;
7806:
7807: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7808: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7809: strcpy(gplotlabel,"(");
1.238 brouard 7810: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7811: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7812: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7813: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7814: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7815: vlv= nbcode[Tvaraff[k]][lv];
7816: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7817: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7818: }
7819: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7820: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7821: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7822: }
1.264 brouard 7823: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7824: fprintf(ficgp,"\n#\n");
7825: if(invalidvarcomb[k1]){
7826: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7827: continue;
7828: }
7829:
7830: /* k=2+nlstate*(2*cpt-2); */
7831: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7832: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7833: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7834: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7835: 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 7836: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7837: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7838: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7839: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7840: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7841: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7842:
1.238 brouard 7843: */
7844: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7845: 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 7846: /* 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 7847:
1.238 brouard 7848: }
1.261 brouard 7849: 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 7850: }
1.264 brouard 7851: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7852: } /* end nres */
7853: } /* end kl 3eme */
1.126 brouard 7854:
1.223 brouard 7855: /* 4eme */
1.201 brouard 7856: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7857: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7858: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7859: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7860: continue;
1.238 brouard 7861: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7862: strcpy(gplotlabel,"(");
1.238 brouard 7863: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7864: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7865: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7866: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7867: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7868: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7869: vlv= nbcode[Tvaraff[k]][lv];
7870: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7871: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7872: }
7873: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7874: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7875: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7876: }
1.264 brouard 7877: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7878: fprintf(ficgp,"\n#\n");
7879: if(invalidvarcomb[k1]){
7880: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7881: continue;
1.223 brouard 7882: }
1.238 brouard 7883:
1.241 brouard 7884: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7885: 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 7886: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7887: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7888: k=3;
7889: for (i=1; i<= nlstate ; i ++){
7890: if(i==1){
7891: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7892: }else{
7893: fprintf(ficgp,", '' ");
7894: }
7895: l=(nlstate+ndeath)*(i-1)+1;
7896: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7897: for (j=2; j<= nlstate+ndeath ; j ++)
7898: fprintf(ficgp,"+$%d",k+l+j-1);
7899: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7900: } /* nlstate */
1.264 brouard 7901: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7902: } /* end cpt state*/
7903: } /* end nres */
7904: } /* end covariate k1 */
7905:
1.220 brouard 7906: /* 5eme */
1.201 brouard 7907: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7908: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7909: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7910: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7911: continue;
1.238 brouard 7912: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7913: strcpy(gplotlabel,"(");
1.238 brouard 7914: 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);
7915: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7916: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7917: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7918: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7919: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7920: vlv= nbcode[Tvaraff[k]][lv];
7921: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7922: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7923: }
7924: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7925: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7926: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7927: }
1.264 brouard 7928: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7929: fprintf(ficgp,"\n#\n");
7930: if(invalidvarcomb[k1]){
7931: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7932: continue;
7933: }
1.227 brouard 7934:
1.241 brouard 7935: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7936: 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 7937: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7938: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7939: k=3;
7940: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7941: if(j==1)
7942: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7943: else
7944: fprintf(ficgp,", '' ");
7945: l=(nlstate+ndeath)*(cpt-1) +j;
7946: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7947: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7948: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7949: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7950: } /* nlstate */
7951: fprintf(ficgp,", '' ");
7952: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7953: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7954: l=(nlstate+ndeath)*(cpt-1) +j;
7955: if(j < nlstate)
7956: fprintf(ficgp,"$%d +",k+l);
7957: else
7958: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7959: }
1.264 brouard 7960: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7961: } /* end cpt state*/
7962: } /* end covariate */
7963: } /* end nres */
1.227 brouard 7964:
1.220 brouard 7965: /* 6eme */
1.202 brouard 7966: /* CV preval stable (period) for each covariate */
1.237 brouard 7967: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7968: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7969: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7970: continue;
1.255 brouard 7971: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7972: strcpy(gplotlabel,"(");
1.288 brouard 7973: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7974: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7975: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7976: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7977: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7978: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7979: vlv= nbcode[Tvaraff[k]][lv];
7980: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7981: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7982: }
1.237 brouard 7983: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7984: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7985: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7986: }
1.264 brouard 7987: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7988: fprintf(ficgp,"\n#\n");
1.223 brouard 7989: if(invalidvarcomb[k1]){
1.227 brouard 7990: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7991: continue;
1.223 brouard 7992: }
1.227 brouard 7993:
1.241 brouard 7994: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7995: 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 7996: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7997: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7998: k=3; /* Offset */
1.255 brouard 7999: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8000: if(i==1)
8001: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8002: else
8003: fprintf(ficgp,", '' ");
1.255 brouard 8004: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8005: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8006: for (j=2; j<= nlstate ; j ++)
8007: fprintf(ficgp,"+$%d",k+l+j-1);
8008: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8009: } /* nlstate */
1.264 brouard 8010: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8011: } /* end cpt state*/
8012: } /* end covariate */
1.227 brouard 8013:
8014:
1.220 brouard 8015: /* 7eme */
1.296 brouard 8016: if(prevbcast == 1){
1.288 brouard 8017: /* CV backward prevalence for each covariate */
1.237 brouard 8018: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8019: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8020: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8021: continue;
1.268 brouard 8022: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8023: strcpy(gplotlabel,"(");
1.288 brouard 8024: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8025: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
8026: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
8027: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8028: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 8029: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 8030: vlv= nbcode[Tvaraff[k]][lv];
8031: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8032: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8033: }
1.237 brouard 8034: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8035: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8036: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8037: }
1.264 brouard 8038: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8039: fprintf(ficgp,"\n#\n");
8040: if(invalidvarcomb[k1]){
8041: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8042: continue;
8043: }
8044:
1.241 brouard 8045: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8046: 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 8047: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8048: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8049: k=3; /* Offset */
1.268 brouard 8050: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8051: if(i==1)
8052: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8053: else
8054: fprintf(ficgp,", '' ");
8055: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8056: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8057: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8058: /* 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 8059: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8060: /* for (j=2; j<= nlstate ; j ++) */
8061: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8062: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8063: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8064: } /* nlstate */
1.264 brouard 8065: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8066: } /* end cpt state*/
8067: } /* end covariate */
1.296 brouard 8068: } /* End if prevbcast */
1.218 brouard 8069:
1.223 brouard 8070: /* 8eme */
1.218 brouard 8071: if(prevfcast==1){
1.288 brouard 8072: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8073:
1.237 brouard 8074: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8075: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8076: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8077: continue;
1.211 brouard 8078: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8079: strcpy(gplotlabel,"(");
1.288 brouard 8080: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8081: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8082: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8083: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8084: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8085: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8086: vlv= nbcode[Tvaraff[k]][lv];
8087: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8088: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8089: }
1.237 brouard 8090: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8091: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8092: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8093: }
1.264 brouard 8094: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8095: fprintf(ficgp,"\n#\n");
8096: if(invalidvarcomb[k1]){
8097: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8098: continue;
8099: }
8100:
8101: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8102: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8103: 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 8104: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8105: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8106:
8107: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8108: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8109: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8110: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8111: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8112: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8113: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8114: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8115: if(i==istart){
1.227 brouard 8116: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8117: }else{
8118: fprintf(ficgp,",\\\n '' ");
8119: }
8120: if(cptcoveff ==0){ /* No covariate */
8121: ioffset=2; /* Age is in 2 */
8122: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8123: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8124: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8125: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8126: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8127: if(i==nlstate+1){
1.270 brouard 8128: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8129: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8130: fprintf(ficgp,",\\\n '' ");
8131: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8132: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8133: offyear, \
1.268 brouard 8134: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8135: }else
1.227 brouard 8136: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8137: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8138: }else{ /* more than 2 covariates */
1.270 brouard 8139: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8140: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8141: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8142: iyearc=ioffset-1;
8143: iagec=ioffset;
1.227 brouard 8144: fprintf(ficgp," u %d:(",ioffset);
8145: kl=0;
8146: strcpy(gplotcondition,"(");
8147: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
8148: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
8149: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8150: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8151: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8152: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
8153: kl++;
8154: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8155: kl++;
8156: if(k <cptcoveff && cptcoveff>1)
8157: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8158: }
8159: strcpy(gplotcondition+strlen(gplotcondition),")");
8160: /* 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 *\/ */
8161: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8162: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8163: /* '' 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*/
8164: if(i==nlstate+1){
1.270 brouard 8165: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8166: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8167: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8168: fprintf(ficgp," u %d:(",iagec);
8169: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8170: iyearc, iagec, offyear, \
8171: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8172: /* '' 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 8173: }else{
8174: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8175: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8176: }
8177: } /* end if covariate */
8178: } /* nlstate */
1.264 brouard 8179: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8180: } /* end cpt state*/
8181: } /* end covariate */
8182: } /* End if prevfcast */
1.227 brouard 8183:
1.296 brouard 8184: if(prevbcast==1){
1.268 brouard 8185: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8186:
8187: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8188: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8189: if(m != 1 && TKresult[nres]!= k1)
8190: continue;
8191: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8192: strcpy(gplotlabel,"(");
8193: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
8194: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8195: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8196: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8197: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8198: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8199: vlv= nbcode[Tvaraff[k]][lv];
8200: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8201: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8202: }
8203: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8204: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8205: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8206: }
8207: strcpy(gplotlabel+strlen(gplotlabel),")");
8208: fprintf(ficgp,"\n#\n");
8209: if(invalidvarcomb[k1]){
8210: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8211: continue;
8212: }
8213:
8214: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8215: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8216: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8217: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8218: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8219:
8220: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8221: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8222: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8223: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8224: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8225: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8226: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8227: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8228: if(i==istart){
8229: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8230: }else{
8231: fprintf(ficgp,",\\\n '' ");
8232: }
8233: if(cptcoveff ==0){ /* No covariate */
8234: ioffset=2; /* Age is in 2 */
8235: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8236: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8237: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8238: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8239: fprintf(ficgp," u %d:(", ioffset);
8240: if(i==nlstate+1){
1.270 brouard 8241: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8242: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8243: fprintf(ficgp,",\\\n '' ");
8244: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8245: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8246: offbyear, \
8247: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8248: }else
8249: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8250: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8251: }else{ /* more than 2 covariates */
1.270 brouard 8252: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8253: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8254: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8255: iyearc=ioffset-1;
8256: iagec=ioffset;
1.268 brouard 8257: fprintf(ficgp," u %d:(",ioffset);
8258: kl=0;
8259: strcpy(gplotcondition,"(");
8260: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
8261: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
8262: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8263: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8264: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8265: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
8266: kl++;
8267: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8268: kl++;
8269: if(k <cptcoveff && cptcoveff>1)
8270: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8271: }
8272: strcpy(gplotcondition+strlen(gplotcondition),")");
8273: /* 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 *\/ */
8274: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8275: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8276: /* '' 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*/
8277: if(i==nlstate+1){
1.270 brouard 8278: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
8279: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 8280: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8281: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 8282: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 8283: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
8284: iyearc,iagec,offbyear, \
8285: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 8286: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
8287: }else{
8288: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
8289: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
8290: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
8291: }
8292: } /* end if covariate */
8293: } /* nlstate */
8294: fprintf(ficgp,"\nset out; unset label;\n");
8295: } /* end cpt state*/
8296: } /* end covariate */
1.296 brouard 8297: } /* End if prevbcast */
1.268 brouard 8298:
1.227 brouard 8299:
1.238 brouard 8300: /* 9eme writing MLE parameters */
8301: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 8302: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 8303: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 8304: for(k=1; k <=(nlstate+ndeath); k++){
8305: if (k != i) {
1.227 brouard 8306: fprintf(ficgp,"# current state %d\n",k);
8307: for(j=1; j <=ncovmodel; j++){
8308: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
8309: jk++;
8310: }
8311: fprintf(ficgp,"\n");
1.126 brouard 8312: }
8313: }
1.223 brouard 8314: }
1.187 brouard 8315: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8316:
1.145 brouard 8317: /*goto avoid;*/
1.238 brouard 8318: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8319: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8320: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8321: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8322: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8323: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8324: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8325: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8326: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8327: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8328: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8329: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8330: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8331: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8332: fprintf(ficgp,"#\n");
1.223 brouard 8333: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8334: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8335: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8336: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8337: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
8338: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 8339: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 8340: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8341: continue;
1.264 brouard 8342: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
8343: strcpy(gplotlabel,"(");
1.276 brouard 8344: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 8345: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8346: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8347: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8348: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8349: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8350: vlv= nbcode[Tvaraff[k]][lv];
8351: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8352: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8353: }
1.237 brouard 8354: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8355: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8356: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8357: }
1.264 brouard 8358: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8359: fprintf(ficgp,"\n#\n");
1.264 brouard 8360: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8361: fprintf(ficgp,"\nset key outside ");
8362: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8363: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8364: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8365: if (ng==1){
8366: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8367: fprintf(ficgp,"\nunset log y");
8368: }else if (ng==2){
8369: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8370: fprintf(ficgp,"\nset log y");
8371: }else if (ng==3){
8372: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8373: fprintf(ficgp,"\nset log y");
8374: }else
8375: fprintf(ficgp,"\nunset title ");
8376: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8377: i=1;
8378: for(k2=1; k2<=nlstate; k2++) {
8379: k3=i;
8380: for(k=1; k<=(nlstate+ndeath); k++) {
8381: if (k != k2){
8382: switch( ng) {
8383: case 1:
8384: if(nagesqr==0)
8385: fprintf(ficgp," p%d+p%d*x",i,i+1);
8386: else /* nagesqr =1 */
8387: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8388: break;
8389: case 2: /* ng=2 */
8390: if(nagesqr==0)
8391: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8392: else /* nagesqr =1 */
8393: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8394: break;
8395: case 3:
8396: if(nagesqr==0)
8397: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8398: else /* nagesqr =1 */
8399: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8400: break;
8401: }
8402: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8403: ijp=1; /* product no age */
8404: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8405: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8406: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8407: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.325 brouard 8408: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
1.268 brouard 8409: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.325 brouard 8410: if(DummyV[j]==0){/* Bug valgrind */
1.268 brouard 8411: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8412: }else{ /* quantitative */
8413: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8414: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8415: }
8416: ij++;
1.237 brouard 8417: }
1.268 brouard 8418: }
8419: }else if(cptcovprod >0){
8420: if(j==Tprod[ijp]) { /* */
8421: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8422: if(ijp <=cptcovprod) { /* Product */
8423: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8424: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8425: /* 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)]); */
8426: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8427: }else{ /* Vn is dummy and Vm is quanti */
8428: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8429: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8430: }
8431: }else{ /* Vn*Vm Vn is quanti */
8432: if(DummyV[Tvard[ijp][2]]==0){
8433: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8434: }else{ /* Both quanti */
8435: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8436: }
1.237 brouard 8437: }
1.268 brouard 8438: ijp++;
1.237 brouard 8439: }
1.268 brouard 8440: } /* end Tprod */
1.237 brouard 8441: } else{ /* simple covariate */
1.264 brouard 8442: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8443: if(Dummy[j]==0){
8444: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8445: }else{ /* quantitative */
8446: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8447: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8448: }
1.237 brouard 8449: } /* end simple */
8450: } /* end j */
1.223 brouard 8451: }else{
8452: i=i-ncovmodel;
8453: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8454: fprintf(ficgp," (1.");
8455: }
1.227 brouard 8456:
1.223 brouard 8457: if(ng != 1){
8458: fprintf(ficgp,")/(1");
1.227 brouard 8459:
1.264 brouard 8460: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8461: if(nagesqr==0)
1.264 brouard 8462: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8463: else /* nagesqr =1 */
1.264 brouard 8464: 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 8465:
1.223 brouard 8466: ij=1;
8467: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8468: if(cptcovage >0){
8469: if((j-2)==Tage[ij]) { /* Bug valgrind */
8470: if(ij <=cptcovage) { /* Bug valgrind */
8471: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8472: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8473: ij++;
8474: }
8475: }
8476: }else
8477: 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 8478: }
8479: fprintf(ficgp,")");
8480: }
8481: fprintf(ficgp,")");
8482: if(ng ==2)
1.276 brouard 8483: 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 8484: else /* ng= 3 */
1.276 brouard 8485: 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 8486: }else{ /* end ng <> 1 */
8487: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8488: 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 8489: }
8490: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8491: fprintf(ficgp,",");
8492: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8493: fprintf(ficgp,",");
8494: i=i+ncovmodel;
8495: } /* end k */
8496: } /* end k2 */
1.276 brouard 8497: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8498: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8499: } /* end k1 */
1.223 brouard 8500: } /* end ng */
8501: /* avoid: */
8502: fflush(ficgp);
1.126 brouard 8503: } /* end gnuplot */
8504:
8505:
8506: /*************** Moving average **************/
1.219 brouard 8507: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8508: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8509:
1.222 brouard 8510: int i, cpt, cptcod;
8511: int modcovmax =1;
8512: int mobilavrange, mob;
8513: int iage=0;
1.288 brouard 8514: int firstA1=0, firstA2=0;
1.222 brouard 8515:
1.266 brouard 8516: double sum=0., sumr=0.;
1.222 brouard 8517: double age;
1.266 brouard 8518: double *sumnewp, *sumnewm, *sumnewmr;
8519: double *agemingood, *agemaxgood;
8520: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8521:
8522:
1.278 brouard 8523: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8524: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8525:
8526: sumnewp = vector(1,ncovcombmax);
8527: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8528: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8529: agemingood = vector(1,ncovcombmax);
1.266 brouard 8530: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8531: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8532: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8533:
8534: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8535: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8536: sumnewp[cptcod]=0.;
1.266 brouard 8537: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8538: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8539: }
8540: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8541:
1.266 brouard 8542: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8543: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8544: else mobilavrange=mobilav;
8545: for (age=bage; age<=fage; age++)
8546: for (i=1; i<=nlstate;i++)
8547: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8548: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8549: /* We keep the original values on the extreme ages bage, fage and for
8550: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8551: we use a 5 terms etc. until the borders are no more concerned.
8552: */
8553: for (mob=3;mob <=mobilavrange;mob=mob+2){
8554: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8555: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8556: sumnewm[cptcod]=0.;
8557: for (i=1; i<=nlstate;i++){
1.222 brouard 8558: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8559: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8560: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8561: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8562: }
8563: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8564: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8565: } /* end i */
8566: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8567: } /* end cptcod */
1.222 brouard 8568: }/* end age */
8569: }/* end mob */
1.266 brouard 8570: }else{
8571: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8572: return -1;
1.266 brouard 8573: }
8574:
8575: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8576: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8577: if(invalidvarcomb[cptcod]){
8578: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8579: continue;
8580: }
1.219 brouard 8581:
1.266 brouard 8582: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8583: sumnewm[cptcod]=0.;
8584: sumnewmr[cptcod]=0.;
8585: for (i=1; i<=nlstate;i++){
8586: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8587: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8588: }
8589: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8590: agemingoodr[cptcod]=age;
8591: }
8592: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8593: agemingood[cptcod]=age;
8594: }
8595: } /* age */
8596: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
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(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8604: agemaxgoodr[cptcod]=age;
1.222 brouard 8605: }
8606: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8607: agemaxgood[cptcod]=age;
8608: }
8609: } /* age */
8610: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8611: /* but they will change */
1.288 brouard 8612: firstA1=0;firstA2=0;
1.266 brouard 8613: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8614: sumnewm[cptcod]=0.;
8615: sumnewmr[cptcod]=0.;
8616: for (i=1; i<=nlstate;i++){
8617: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8618: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8619: }
8620: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8621: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8622: agemaxgoodr[cptcod]=age; /* age min */
8623: for (i=1; i<=nlstate;i++)
8624: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8625: }else{ /* bad we change the value with the values of good ages */
8626: for (i=1; i<=nlstate;i++){
8627: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8628: } /* i */
8629: } /* end bad */
8630: }else{
8631: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8632: agemaxgood[cptcod]=age;
8633: }else{ /* bad we change the value with the values of good ages */
8634: for (i=1; i<=nlstate;i++){
8635: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8636: } /* i */
8637: } /* end bad */
8638: }/* end else */
8639: sum=0.;sumr=0.;
8640: for (i=1; i<=nlstate;i++){
8641: sum+=mobaverage[(int)age][i][cptcod];
8642: sumr+=probs[(int)age][i][cptcod];
8643: }
8644: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8645: if(!firstA1){
8646: firstA1=1;
8647: 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);
8648: }
8649: 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 8650: } /* end bad */
8651: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8652: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8653: if(!firstA2){
8654: firstA2=1;
8655: 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);
8656: }
8657: 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 8658: } /* end bad */
8659: }/* age */
1.266 brouard 8660:
8661: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8662: sumnewm[cptcod]=0.;
1.266 brouard 8663: sumnewmr[cptcod]=0.;
1.222 brouard 8664: for (i=1; i<=nlstate;i++){
8665: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8666: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8667: }
8668: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8669: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8670: agemingoodr[cptcod]=age;
8671: for (i=1; i<=nlstate;i++)
8672: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8673: }else{ /* bad we change the value with the values of good ages */
8674: for (i=1; i<=nlstate;i++){
8675: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8676: } /* i */
8677: } /* end bad */
8678: }else{
8679: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8680: agemingood[cptcod]=age;
8681: }else{ /* bad */
8682: for (i=1; i<=nlstate;i++){
8683: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8684: } /* i */
8685: } /* end bad */
8686: }/* end else */
8687: sum=0.;sumr=0.;
8688: for (i=1; i<=nlstate;i++){
8689: sum+=mobaverage[(int)age][i][cptcod];
8690: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8691: }
1.266 brouard 8692: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8693: 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 8694: } /* end bad */
8695: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8696: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8697: 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 8698: } /* end bad */
8699: }/* age */
1.266 brouard 8700:
1.222 brouard 8701:
8702: for (age=bage; age<=fage; age++){
1.235 brouard 8703: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8704: sumnewp[cptcod]=0.;
8705: sumnewm[cptcod]=0.;
8706: for (i=1; i<=nlstate;i++){
8707: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8708: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8709: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8710: }
8711: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8712: }
8713: /* printf("\n"); */
8714: /* } */
1.266 brouard 8715:
1.222 brouard 8716: /* brutal averaging */
1.266 brouard 8717: /* for (i=1; i<=nlstate;i++){ */
8718: /* for (age=1; age<=bage; age++){ */
8719: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8720: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8721: /* } */
8722: /* for (age=fage; age<=AGESUP; age++){ */
8723: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8724: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8725: /* } */
8726: /* } /\* end i status *\/ */
8727: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8728: /* for (age=1; age<=AGESUP; age++){ */
8729: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8730: /* mobaverage[(int)age][i][cptcod]=0.; */
8731: /* } */
8732: /* } */
1.222 brouard 8733: }/* end cptcod */
1.266 brouard 8734: free_vector(agemaxgoodr,1, ncovcombmax);
8735: free_vector(agemaxgood,1, ncovcombmax);
8736: free_vector(agemingood,1, ncovcombmax);
8737: free_vector(agemingoodr,1, ncovcombmax);
8738: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8739: free_vector(sumnewm,1, ncovcombmax);
8740: free_vector(sumnewp,1, ncovcombmax);
8741: return 0;
8742: }/* End movingaverage */
1.218 brouard 8743:
1.126 brouard 8744:
1.296 brouard 8745:
1.126 brouard 8746: /************** Forecasting ******************/
1.296 brouard 8747: /* 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)*/
8748: 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){
8749: /* dateintemean, mean date of interviews
8750: dateprojd, year, month, day of starting projection
8751: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8752: agemin, agemax range of age
8753: dateprev1 dateprev2 range of dates during which prevalence is computed
8754: */
1.296 brouard 8755: /* double anprojd, mprojd, jprojd; */
8756: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8757: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8758: double agec; /* generic age */
1.296 brouard 8759: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8760: double *popeffectif,*popcount;
8761: double ***p3mat;
1.218 brouard 8762: /* double ***mobaverage; */
1.126 brouard 8763: char fileresf[FILENAMELENGTH];
8764:
8765: agelim=AGESUP;
1.211 brouard 8766: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8767: in each health status at the date of interview (if between dateprev1 and dateprev2).
8768: We still use firstpass and lastpass as another selection.
8769: */
1.214 brouard 8770: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8771: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8772:
1.201 brouard 8773: strcpy(fileresf,"F_");
8774: strcat(fileresf,fileresu);
1.126 brouard 8775: if((ficresf=fopen(fileresf,"w"))==NULL) {
8776: printf("Problem with forecast resultfile: %s\n", fileresf);
8777: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8778: }
1.235 brouard 8779: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8780: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8781:
1.225 brouard 8782: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8783:
8784:
8785: stepsize=(int) (stepm+YEARM-1)/YEARM;
8786: if (stepm<=12) stepsize=1;
8787: if(estepm < stepm){
8788: printf ("Problem %d lower than %d\n",estepm, stepm);
8789: }
1.270 brouard 8790: else{
8791: hstepm=estepm;
8792: }
8793: if(estepm > stepm){ /* Yes every two year */
8794: stepsize=2;
8795: }
1.296 brouard 8796: hstepm=hstepm/stepm;
1.126 brouard 8797:
1.296 brouard 8798:
8799: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8800: /* fractional in yp1 *\/ */
8801: /* aintmean=yp; */
8802: /* yp2=modf((yp1*12),&yp); */
8803: /* mintmean=yp; */
8804: /* yp1=modf((yp2*30.5),&yp); */
8805: /* jintmean=yp; */
8806: /* if(jintmean==0) jintmean=1; */
8807: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8808:
1.296 brouard 8809:
8810: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8811: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8812: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8813: i1=pow(2,cptcoveff);
1.126 brouard 8814: if (cptcovn < 1){i1=1;}
8815:
1.296 brouard 8816: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8817:
8818: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8819:
1.126 brouard 8820: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8821: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8822: for(k=1; k<=i1;k++){
1.253 brouard 8823: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8824: continue;
1.227 brouard 8825: if(invalidvarcomb[k]){
8826: printf("\nCombination (%d) projection ignored because no cases \n",k);
8827: continue;
8828: }
8829: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8830: for(j=1;j<=cptcoveff;j++) {
8831: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8832: }
1.235 brouard 8833: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8834: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8835: }
1.227 brouard 8836: fprintf(ficresf," yearproj age");
8837: for(j=1; j<=nlstate+ndeath;j++){
8838: for(i=1; i<=nlstate;i++)
8839: fprintf(ficresf," p%d%d",i,j);
8840: fprintf(ficresf," wp.%d",j);
8841: }
1.296 brouard 8842: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8843: fprintf(ficresf,"\n");
1.296 brouard 8844: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8845: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8846: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8847: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8848: nhstepm = nhstepm/hstepm;
8849: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8850: oldm=oldms;savm=savms;
1.268 brouard 8851: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8852: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8853: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8854: for (h=0; h<=nhstepm; h++){
8855: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8856: break;
8857: }
8858: }
8859: fprintf(ficresf,"\n");
8860: for(j=1;j<=cptcoveff;j++)
8861: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8862: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8863:
8864: for(j=1; j<=nlstate+ndeath;j++) {
8865: ppij=0.;
8866: for(i=1; i<=nlstate;i++) {
1.278 brouard 8867: if (mobilav>=1)
8868: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8869: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8870: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8871: }
1.268 brouard 8872: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8873: } /* end i */
8874: fprintf(ficresf," %.3f", ppij);
8875: }/* end j */
1.227 brouard 8876: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8877: } /* end agec */
1.266 brouard 8878: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8879: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8880: } /* end yearp */
8881: } /* end k */
1.219 brouard 8882:
1.126 brouard 8883: fclose(ficresf);
1.215 brouard 8884: printf("End of Computing forecasting \n");
8885: fprintf(ficlog,"End of Computing forecasting\n");
8886:
1.126 brouard 8887: }
8888:
1.269 brouard 8889: /************** Back Forecasting ******************/
1.296 brouard 8890: /* 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){ */
8891: 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){
8892: /* back1, year, month, day of starting backprojection
1.267 brouard 8893: agemin, agemax range of age
8894: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8895: anback2 year of end of backprojection (same day and month as back1).
8896: prevacurrent and prev are prevalences.
1.267 brouard 8897: */
8898: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8899: double agec; /* generic age */
1.302 brouard 8900: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 8901: double *popeffectif,*popcount;
8902: double ***p3mat;
8903: /* double ***mobaverage; */
8904: char fileresfb[FILENAMELENGTH];
8905:
1.268 brouard 8906: agelim=AGEINF;
1.267 brouard 8907: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8908: in each health status at the date of interview (if between dateprev1 and dateprev2).
8909: We still use firstpass and lastpass as another selection.
8910: */
8911: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8912: /* firstpass, lastpass, stepm, weightopt, model); */
8913:
8914: /*Do we need to compute prevalence again?*/
8915:
8916: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8917:
8918: strcpy(fileresfb,"FB_");
8919: strcat(fileresfb,fileresu);
8920: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8921: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8922: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8923: }
8924: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8925: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8926:
8927: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8928:
8929:
8930: stepsize=(int) (stepm+YEARM-1)/YEARM;
8931: if (stepm<=12) stepsize=1;
8932: if(estepm < stepm){
8933: printf ("Problem %d lower than %d\n",estepm, stepm);
8934: }
1.270 brouard 8935: else{
8936: hstepm=estepm;
8937: }
8938: if(estepm >= stepm){ /* Yes every two year */
8939: stepsize=2;
8940: }
1.267 brouard 8941:
8942: hstepm=hstepm/stepm;
1.296 brouard 8943: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8944: /* fractional in yp1 *\/ */
8945: /* aintmean=yp; */
8946: /* yp2=modf((yp1*12),&yp); */
8947: /* mintmean=yp; */
8948: /* yp1=modf((yp2*30.5),&yp); */
8949: /* jintmean=yp; */
8950: /* if(jintmean==0) jintmean=1; */
8951: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8952:
8953: i1=pow(2,cptcoveff);
8954: if (cptcovn < 1){i1=1;}
8955:
1.296 brouard 8956: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8957: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8958:
8959: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8960:
8961: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8962: for(k=1; k<=i1;k++){
8963: if(i1 != 1 && TKresult[nres]!= k)
8964: continue;
8965: if(invalidvarcomb[k]){
8966: printf("\nCombination (%d) projection ignored because no cases \n",k);
8967: continue;
8968: }
1.268 brouard 8969: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8970: for(j=1;j<=cptcoveff;j++) {
8971: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8972: }
8973: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8974: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8975: }
8976: fprintf(ficresfb," yearbproj age");
8977: for(j=1; j<=nlstate+ndeath;j++){
8978: for(i=1; i<=nlstate;i++)
1.268 brouard 8979: fprintf(ficresfb," b%d%d",i,j);
8980: fprintf(ficresfb," b.%d",j);
1.267 brouard 8981: }
1.296 brouard 8982: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8983: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8984: fprintf(ficresfb,"\n");
1.296 brouard 8985: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8986: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8987: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8988: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8989: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8990: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8991: nhstepm = nhstepm/hstepm;
8992: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8993: oldm=oldms;savm=savms;
1.268 brouard 8994: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8995: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8996: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8997: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8998: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8999: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9000: for (h=0; h<=nhstepm; h++){
1.268 brouard 9001: if (h*hstepm/YEARM*stepm ==-yearp) {
9002: break;
9003: }
9004: }
9005: fprintf(ficresfb,"\n");
9006: for(j=1;j<=cptcoveff;j++)
9007: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 9008: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9009: for(i=1; i<=nlstate+ndeath;i++) {
9010: ppij=0.;ppi=0.;
9011: for(j=1; j<=nlstate;j++) {
9012: /* if (mobilav==1) */
1.269 brouard 9013: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9014: ppi=ppi+prevacurrent[(int)agec][j][k];
9015: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9016: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9017: /* else { */
9018: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9019: /* } */
1.268 brouard 9020: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9021: } /* end j */
9022: if(ppi <0.99){
9023: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9024: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9025: }
9026: fprintf(ficresfb," %.3f", ppij);
9027: }/* end j */
1.267 brouard 9028: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9029: } /* end agec */
9030: } /* end yearp */
9031: } /* end k */
1.217 brouard 9032:
1.267 brouard 9033: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9034:
1.267 brouard 9035: fclose(ficresfb);
9036: printf("End of Computing Back forecasting \n");
9037: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9038:
1.267 brouard 9039: }
1.217 brouard 9040:
1.269 brouard 9041: /* Variance of prevalence limit: varprlim */
9042: 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 9043: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9044:
9045: char fileresvpl[FILENAMELENGTH];
9046: FILE *ficresvpl;
9047: double **oldm, **savm;
9048: double **varpl; /* Variances of prevalence limits by age */
9049: int i1, k, nres, j ;
9050:
9051: strcpy(fileresvpl,"VPL_");
9052: strcat(fileresvpl,fileresu);
9053: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9054: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9055: exit(0);
9056: }
1.288 brouard 9057: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9058: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9059:
9060: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9061: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9062:
9063: i1=pow(2,cptcoveff);
9064: if (cptcovn < 1){i1=1;}
9065:
9066: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9067: for(k=1; k<=i1;k++){
9068: if(i1 != 1 && TKresult[nres]!= k)
9069: continue;
9070: fprintf(ficresvpl,"\n#****** ");
9071: printf("\n#****** ");
9072: fprintf(ficlog,"\n#****** ");
9073: for(j=1;j<=cptcoveff;j++) {
9074: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9075: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9076: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9077: }
9078: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9079: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9080: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9081: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9082: }
9083: fprintf(ficresvpl,"******\n");
9084: printf("******\n");
9085: fprintf(ficlog,"******\n");
9086:
9087: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9088: oldm=oldms;savm=savms;
9089: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9090: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9091: /*}*/
9092: }
9093:
9094: fclose(ficresvpl);
1.288 brouard 9095: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9096: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9097:
9098: }
9099: /* Variance of back prevalence: varbprlim */
9100: 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){
9101: /*------- Variance of back (stable) prevalence------*/
9102:
9103: char fileresvbl[FILENAMELENGTH];
9104: FILE *ficresvbl;
9105:
9106: double **oldm, **savm;
9107: double **varbpl; /* Variances of back prevalence limits by age */
9108: int i1, k, nres, j ;
9109:
9110: strcpy(fileresvbl,"VBL_");
9111: strcat(fileresvbl,fileresu);
9112: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9113: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9114: exit(0);
9115: }
9116: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9117: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9118:
9119:
9120: i1=pow(2,cptcoveff);
9121: if (cptcovn < 1){i1=1;}
9122:
9123: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9124: for(k=1; k<=i1;k++){
9125: if(i1 != 1 && TKresult[nres]!= k)
9126: continue;
9127: fprintf(ficresvbl,"\n#****** ");
9128: printf("\n#****** ");
9129: fprintf(ficlog,"\n#****** ");
9130: for(j=1;j<=cptcoveff;j++) {
9131: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9132: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9133: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9134: }
9135: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9136: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9137: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9138: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9139: }
9140: fprintf(ficresvbl,"******\n");
9141: printf("******\n");
9142: fprintf(ficlog,"******\n");
9143:
9144: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9145: oldm=oldms;savm=savms;
9146:
9147: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9148: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9149: /*}*/
9150: }
9151:
9152: fclose(ficresvbl);
9153: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9154: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9155:
9156: } /* End of varbprlim */
9157:
1.126 brouard 9158: /************** Forecasting *****not tested NB*************/
1.227 brouard 9159: /* 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 9160:
1.227 brouard 9161: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
9162: /* int *popage; */
9163: /* double calagedatem, agelim, kk1, kk2; */
9164: /* double *popeffectif,*popcount; */
9165: /* double ***p3mat,***tabpop,***tabpopprev; */
9166: /* /\* double ***mobaverage; *\/ */
9167: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 9168:
1.227 brouard 9169: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9170: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9171: /* agelim=AGESUP; */
9172: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 9173:
1.227 brouard 9174: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 9175:
9176:
1.227 brouard 9177: /* strcpy(filerespop,"POP_"); */
9178: /* strcat(filerespop,fileresu); */
9179: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
9180: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
9181: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
9182: /* } */
9183: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
9184: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 9185:
1.227 brouard 9186: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 9187:
1.227 brouard 9188: /* /\* if (mobilav!=0) { *\/ */
9189: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
9190: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
9191: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9192: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9193: /* /\* } *\/ */
9194: /* /\* } *\/ */
1.126 brouard 9195:
1.227 brouard 9196: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
9197: /* if (stepm<=12) stepsize=1; */
1.126 brouard 9198:
1.227 brouard 9199: /* agelim=AGESUP; */
1.126 brouard 9200:
1.227 brouard 9201: /* hstepm=1; */
9202: /* hstepm=hstepm/stepm; */
1.218 brouard 9203:
1.227 brouard 9204: /* if (popforecast==1) { */
9205: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
9206: /* printf("Problem with population file : %s\n",popfile);exit(0); */
9207: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
9208: /* } */
9209: /* popage=ivector(0,AGESUP); */
9210: /* popeffectif=vector(0,AGESUP); */
9211: /* popcount=vector(0,AGESUP); */
1.126 brouard 9212:
1.227 brouard 9213: /* i=1; */
9214: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 9215:
1.227 brouard 9216: /* imx=i; */
9217: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
9218: /* } */
1.218 brouard 9219:
1.227 brouard 9220: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
9221: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
9222: /* k=k+1; */
9223: /* fprintf(ficrespop,"\n#******"); */
9224: /* for(j=1;j<=cptcoveff;j++) { */
9225: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
9226: /* } */
9227: /* fprintf(ficrespop,"******\n"); */
9228: /* fprintf(ficrespop,"# Age"); */
9229: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
9230: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 9231:
1.227 brouard 9232: /* for (cpt=0; cpt<=0;cpt++) { */
9233: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 9234:
1.227 brouard 9235: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9236: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9237: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9238:
1.227 brouard 9239: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9240: /* oldm=oldms;savm=savms; */
9241: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 9242:
1.227 brouard 9243: /* for (h=0; h<=nhstepm; h++){ */
9244: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9245: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9246: /* } */
9247: /* for(j=1; j<=nlstate+ndeath;j++) { */
9248: /* kk1=0.;kk2=0; */
9249: /* for(i=1; i<=nlstate;i++) { */
9250: /* if (mobilav==1) */
9251: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
9252: /* else { */
9253: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
9254: /* } */
9255: /* } */
9256: /* if (h==(int)(calagedatem+12*cpt)){ */
9257: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
9258: /* /\*fprintf(ficrespop," %.3f", kk1); */
9259: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
9260: /* } */
9261: /* } */
9262: /* for(i=1; i<=nlstate;i++){ */
9263: /* kk1=0.; */
9264: /* for(j=1; j<=nlstate;j++){ */
9265: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
9266: /* } */
9267: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
9268: /* } */
1.218 brouard 9269:
1.227 brouard 9270: /* if (h==(int)(calagedatem+12*cpt)) */
9271: /* for(j=1; j<=nlstate;j++) */
9272: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
9273: /* } */
9274: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9275: /* } */
9276: /* } */
1.218 brouard 9277:
1.227 brouard 9278: /* /\******\/ */
1.218 brouard 9279:
1.227 brouard 9280: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
9281: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
9282: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9283: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9284: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9285:
1.227 brouard 9286: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9287: /* oldm=oldms;savm=savms; */
9288: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9289: /* for (h=0; h<=nhstepm; h++){ */
9290: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9291: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9292: /* } */
9293: /* for(j=1; j<=nlstate+ndeath;j++) { */
9294: /* kk1=0.;kk2=0; */
9295: /* for(i=1; i<=nlstate;i++) { */
9296: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
9297: /* } */
9298: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
9299: /* } */
9300: /* } */
9301: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9302: /* } */
9303: /* } */
9304: /* } */
9305: /* } */
1.218 brouard 9306:
1.227 brouard 9307: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 9308:
1.227 brouard 9309: /* if (popforecast==1) { */
9310: /* free_ivector(popage,0,AGESUP); */
9311: /* free_vector(popeffectif,0,AGESUP); */
9312: /* free_vector(popcount,0,AGESUP); */
9313: /* } */
9314: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9315: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9316: /* fclose(ficrespop); */
9317: /* } /\* End of popforecast *\/ */
1.218 brouard 9318:
1.126 brouard 9319: int fileappend(FILE *fichier, char *optionfich)
9320: {
9321: if((fichier=fopen(optionfich,"a"))==NULL) {
9322: printf("Problem with file: %s\n", optionfich);
9323: fprintf(ficlog,"Problem with file: %s\n", optionfich);
9324: return (0);
9325: }
9326: fflush(fichier);
9327: return (1);
9328: }
9329:
9330:
9331: /**************** function prwizard **********************/
9332: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9333: {
9334:
9335: /* Wizard to print covariance matrix template */
9336:
1.164 brouard 9337: char ca[32], cb[32];
9338: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9339: int numlinepar;
9340:
9341: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9342: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9343: for(i=1; i <=nlstate; i++){
9344: jj=0;
9345: for(j=1; j <=nlstate+ndeath; j++){
9346: if(j==i) continue;
9347: jj++;
9348: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9349: printf("%1d%1d",i,j);
9350: fprintf(ficparo,"%1d%1d",i,j);
9351: for(k=1; k<=ncovmodel;k++){
9352: /* printf(" %lf",param[i][j][k]); */
9353: /* fprintf(ficparo," %lf",param[i][j][k]); */
9354: printf(" 0.");
9355: fprintf(ficparo," 0.");
9356: }
9357: printf("\n");
9358: fprintf(ficparo,"\n");
9359: }
9360: }
9361: printf("# Scales (for hessian or gradient estimation)\n");
9362: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9363: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9364: for(i=1; i <=nlstate; i++){
9365: jj=0;
9366: for(j=1; j <=nlstate+ndeath; j++){
9367: if(j==i) continue;
9368: jj++;
9369: fprintf(ficparo,"%1d%1d",i,j);
9370: printf("%1d%1d",i,j);
9371: fflush(stdout);
9372: for(k=1; k<=ncovmodel;k++){
9373: /* printf(" %le",delti3[i][j][k]); */
9374: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9375: printf(" 0.");
9376: fprintf(ficparo," 0.");
9377: }
9378: numlinepar++;
9379: printf("\n");
9380: fprintf(ficparo,"\n");
9381: }
9382: }
9383: printf("# Covariance matrix\n");
9384: /* # 121 Var(a12)\n\ */
9385: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9386: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9387: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9388: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9389: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9390: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9391: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9392: fflush(stdout);
9393: fprintf(ficparo,"# Covariance matrix\n");
9394: /* # 121 Var(a12)\n\ */
9395: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9396: /* # ...\n\ */
9397: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9398:
9399: for(itimes=1;itimes<=2;itimes++){
9400: jj=0;
9401: for(i=1; i <=nlstate; i++){
9402: for(j=1; j <=nlstate+ndeath; j++){
9403: if(j==i) continue;
9404: for(k=1; k<=ncovmodel;k++){
9405: jj++;
9406: ca[0]= k+'a'-1;ca[1]='\0';
9407: if(itimes==1){
9408: printf("#%1d%1d%d",i,j,k);
9409: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9410: }else{
9411: printf("%1d%1d%d",i,j,k);
9412: fprintf(ficparo,"%1d%1d%d",i,j,k);
9413: /* printf(" %.5le",matcov[i][j]); */
9414: }
9415: ll=0;
9416: for(li=1;li <=nlstate; li++){
9417: for(lj=1;lj <=nlstate+ndeath; lj++){
9418: if(lj==li) continue;
9419: for(lk=1;lk<=ncovmodel;lk++){
9420: ll++;
9421: if(ll<=jj){
9422: cb[0]= lk +'a'-1;cb[1]='\0';
9423: if(ll<jj){
9424: if(itimes==1){
9425: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9426: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9427: }else{
9428: printf(" 0.");
9429: fprintf(ficparo," 0.");
9430: }
9431: }else{
9432: if(itimes==1){
9433: printf(" Var(%s%1d%1d)",ca,i,j);
9434: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9435: }else{
9436: printf(" 0.");
9437: fprintf(ficparo," 0.");
9438: }
9439: }
9440: }
9441: } /* end lk */
9442: } /* end lj */
9443: } /* end li */
9444: printf("\n");
9445: fprintf(ficparo,"\n");
9446: numlinepar++;
9447: } /* end k*/
9448: } /*end j */
9449: } /* end i */
9450: } /* end itimes */
9451:
9452: } /* end of prwizard */
9453: /******************* Gompertz Likelihood ******************************/
9454: double gompertz(double x[])
9455: {
1.302 brouard 9456: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9457: int i,n=0; /* n is the size of the sample */
9458:
1.220 brouard 9459: for (i=1;i<=imx ; i++) {
1.126 brouard 9460: sump=sump+weight[i];
9461: /* sump=sump+1;*/
9462: num=num+1;
9463: }
1.302 brouard 9464: L=0.0;
9465: /* agegomp=AGEGOMP; */
1.126 brouard 9466: /* for (i=0; i<=imx; i++)
9467: 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]);*/
9468:
1.302 brouard 9469: for (i=1;i<=imx ; i++) {
9470: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9471: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9472: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9473: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9474: * +
9475: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9476: */
9477: if (wav[i] > 1 || agedc[i] < AGESUP) {
9478: if (cens[i] == 1){
9479: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9480: } else if (cens[i] == 0){
1.126 brouard 9481: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9482: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9483: } else
9484: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9485: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9486: L=L+A*weight[i];
1.126 brouard 9487: /* 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 9488: }
9489: }
1.126 brouard 9490:
1.302 brouard 9491: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9492:
9493: return -2*L*num/sump;
9494: }
9495:
1.136 brouard 9496: #ifdef GSL
9497: /******************* Gompertz_f Likelihood ******************************/
9498: double gompertz_f(const gsl_vector *v, void *params)
9499: {
1.302 brouard 9500: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9501: double *x= (double *) v->data;
9502: int i,n=0; /* n is the size of the sample */
9503:
9504: for (i=0;i<=imx-1 ; i++) {
9505: sump=sump+weight[i];
9506: /* sump=sump+1;*/
9507: num=num+1;
9508: }
9509:
9510:
9511: /* for (i=0; i<=imx; i++)
9512: 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]);*/
9513: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9514: for (i=1;i<=imx ; i++)
9515: {
9516: if (cens[i] == 1 && wav[i]>1)
9517: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9518:
9519: if (cens[i] == 0 && wav[i]>1)
9520: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9521: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9522:
9523: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9524: if (wav[i] > 1 ) { /* ??? */
9525: LL=LL+A*weight[i];
9526: /* 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]);*/
9527: }
9528: }
9529:
9530: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9531: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9532:
9533: return -2*LL*num/sump;
9534: }
9535: #endif
9536:
1.126 brouard 9537: /******************* Printing html file ***********/
1.201 brouard 9538: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9539: int lastpass, int stepm, int weightopt, char model[],\
9540: int imx, double p[],double **matcov,double agemortsup){
9541: int i,k;
9542:
9543: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9544: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9545: for (i=1;i<=2;i++)
9546: 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 9547: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9548: fprintf(fichtm,"</ul>");
9549:
9550: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9551:
9552: 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>");
9553:
9554: for (k=agegomp;k<(agemortsup-2);k++)
9555: 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]);
9556:
9557:
9558: fflush(fichtm);
9559: }
9560:
9561: /******************* Gnuplot file **************/
1.201 brouard 9562: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9563:
9564: char dirfileres[132],optfileres[132];
1.164 brouard 9565:
1.126 brouard 9566: int ng;
9567:
9568:
9569: /*#ifdef windows */
9570: fprintf(ficgp,"cd \"%s\" \n",pathc);
9571: /*#endif */
9572:
9573:
9574: strcpy(dirfileres,optionfilefiname);
9575: strcpy(optfileres,"vpl");
1.199 brouard 9576: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9577: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9578: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9579: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9580: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9581:
9582: }
9583:
1.136 brouard 9584: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9585: {
1.126 brouard 9586:
1.136 brouard 9587: /*-------- data file ----------*/
9588: FILE *fic;
9589: char dummy[]=" ";
1.240 brouard 9590: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9591: int lstra;
1.136 brouard 9592: int linei, month, year,iout;
1.302 brouard 9593: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9594: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9595: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9596: char *stratrunc;
1.223 brouard 9597:
1.240 brouard 9598: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9599: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9600:
1.240 brouard 9601: for(v=1; v <=ncovcol;v++){
9602: DummyV[v]=0;
9603: FixedV[v]=0;
9604: }
9605: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9606: DummyV[v]=1;
9607: FixedV[v]=0;
9608: }
9609: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9610: DummyV[v]=0;
9611: FixedV[v]=1;
9612: }
9613: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9614: DummyV[v]=1;
9615: FixedV[v]=1;
9616: }
9617: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9618: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9619: 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]);
9620: }
1.126 brouard 9621:
1.136 brouard 9622: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9623: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9624: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9625: }
1.126 brouard 9626:
1.302 brouard 9627: /* Is it a BOM UTF-8 Windows file? */
9628: /* First data line */
9629: linei=0;
9630: while(fgets(line, MAXLINE, fic)) {
9631: noffset=0;
9632: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
9633: {
9634: noffset=noffset+3;
9635: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
9636: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
9637: fflush(ficlog); return 1;
9638: }
9639: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
9640: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
9641: {
9642: noffset=noffset+2;
1.304 brouard 9643: 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);
9644: 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 9645: fflush(ficlog); return 1;
9646: }
9647: else if( line[0] == 0 && line[1] == 0)
9648: {
9649: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
9650: noffset=noffset+4;
1.304 brouard 9651: 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);
9652: 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 9653: fflush(ficlog); return 1;
9654: }
9655: } else{
9656: ;/*printf(" Not a BOM file\n");*/
9657: }
9658: /* If line starts with a # it is a comment */
9659: if (line[noffset] == '#') {
9660: linei=linei+1;
9661: break;
9662: }else{
9663: break;
9664: }
9665: }
9666: fclose(fic);
9667: if((fic=fopen(datafile,"r"))==NULL) {
9668: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9669: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
9670: }
9671: /* Not a Bom file */
9672:
1.136 brouard 9673: i=1;
9674: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9675: linei=linei+1;
9676: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9677: if(line[j] == '\t')
9678: line[j] = ' ';
9679: }
9680: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9681: ;
9682: };
9683: line[j+1]=0; /* Trims blanks at end of line */
9684: if(line[0]=='#'){
9685: fprintf(ficlog,"Comment line\n%s\n",line);
9686: printf("Comment line\n%s\n",line);
9687: continue;
9688: }
9689: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9690: strcpy(line, linetmp);
1.223 brouard 9691:
9692: /* Loops on waves */
9693: for (j=maxwav;j>=1;j--){
9694: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9695: cutv(stra, strb, line, ' ');
9696: if(strb[0]=='.') { /* Missing value */
9697: lval=-1;
9698: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9699: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9700: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9701: 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);
9702: 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);
9703: return 1;
9704: }
9705: }else{
9706: errno=0;
9707: /* what_kind_of_number(strb); */
9708: dval=strtod(strb,&endptr);
9709: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9710: /* if(strb != endptr && *endptr == '\0') */
9711: /* dval=dlval; */
9712: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9713: if( strb[0]=='\0' || (*endptr != '\0')){
9714: 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);
9715: 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);
9716: return 1;
9717: }
9718: cotqvar[j][iv][i]=dval;
9719: cotvar[j][ntv+iv][i]=dval;
9720: }
9721: strcpy(line,stra);
1.223 brouard 9722: }/* end loop ntqv */
1.225 brouard 9723:
1.223 brouard 9724: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9725: cutv(stra, strb, line, ' ');
9726: if(strb[0]=='.') { /* Missing value */
9727: lval=-1;
9728: }else{
9729: errno=0;
9730: lval=strtol(strb,&endptr,10);
9731: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9732: if( strb[0]=='\0' || (*endptr != '\0')){
9733: 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);
9734: 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);
9735: return 1;
9736: }
9737: }
9738: if(lval <-1 || lval >1){
9739: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 9740: 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 9741: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9742: For example, for multinomial values like 1, 2 and 3,\n \
9743: build V1=0 V2=0 for the reference value (1),\n \
9744: V1=1 V2=0 for (2) \n \
1.223 brouard 9745: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9746: output of IMaCh is often meaningless.\n \
1.319 brouard 9747: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 9748: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 9749: 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 9750: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9751: For example, for multinomial values like 1, 2 and 3,\n \
9752: build V1=0 V2=0 for the reference value (1),\n \
9753: V1=1 V2=0 for (2) \n \
1.223 brouard 9754: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9755: output of IMaCh is often meaningless.\n \
1.319 brouard 9756: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 9757: return 1;
9758: }
9759: cotvar[j][iv][i]=(double)(lval);
9760: strcpy(line,stra);
1.223 brouard 9761: }/* end loop ntv */
1.225 brouard 9762:
1.223 brouard 9763: /* Statuses at wave */
1.137 brouard 9764: cutv(stra, strb, line, ' ');
1.223 brouard 9765: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9766: lval=-1;
1.136 brouard 9767: }else{
1.238 brouard 9768: errno=0;
9769: lval=strtol(strb,&endptr,10);
9770: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9771: if( strb[0]=='\0' || (*endptr != '\0')){
9772: 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);
9773: 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);
9774: return 1;
9775: }
1.136 brouard 9776: }
1.225 brouard 9777:
1.136 brouard 9778: s[j][i]=lval;
1.225 brouard 9779:
1.223 brouard 9780: /* Date of Interview */
1.136 brouard 9781: strcpy(line,stra);
9782: cutv(stra, strb,line,' ');
1.169 brouard 9783: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9784: }
1.169 brouard 9785: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9786: month=99;
9787: year=9999;
1.136 brouard 9788: }else{
1.225 brouard 9789: 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);
9790: 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);
9791: return 1;
1.136 brouard 9792: }
9793: anint[j][i]= (double) year;
1.302 brouard 9794: mint[j][i]= (double)month;
9795: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
9796: /* 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]); */
9797: /* 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]); */
9798: /* } */
1.136 brouard 9799: strcpy(line,stra);
1.223 brouard 9800: } /* End loop on waves */
1.225 brouard 9801:
1.223 brouard 9802: /* Date of death */
1.136 brouard 9803: cutv(stra, strb,line,' ');
1.169 brouard 9804: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9805: }
1.169 brouard 9806: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9807: month=99;
9808: year=9999;
9809: }else{
1.141 brouard 9810: 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 9811: 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);
9812: return 1;
1.136 brouard 9813: }
9814: andc[i]=(double) year;
9815: moisdc[i]=(double) month;
9816: strcpy(line,stra);
9817:
1.223 brouard 9818: /* Date of birth */
1.136 brouard 9819: cutv(stra, strb,line,' ');
1.169 brouard 9820: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9821: }
1.169 brouard 9822: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9823: month=99;
9824: year=9999;
9825: }else{
1.141 brouard 9826: 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);
9827: 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 9828: return 1;
1.136 brouard 9829: }
9830: if (year==9999) {
1.141 brouard 9831: 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);
9832: 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 9833: return 1;
9834:
1.136 brouard 9835: }
9836: annais[i]=(double)(year);
1.302 brouard 9837: moisnais[i]=(double)(month);
9838: for (j=1;j<=maxwav;j++){
9839: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
9840: 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]);
9841: 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]);
9842: }
9843: }
9844:
1.136 brouard 9845: strcpy(line,stra);
1.225 brouard 9846:
1.223 brouard 9847: /* Sample weight */
1.136 brouard 9848: cutv(stra, strb,line,' ');
9849: errno=0;
9850: dval=strtod(strb,&endptr);
9851: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9852: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9853: 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 9854: fflush(ficlog);
9855: return 1;
9856: }
9857: weight[i]=dval;
9858: strcpy(line,stra);
1.225 brouard 9859:
1.223 brouard 9860: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9861: cutv(stra, strb, line, ' ');
9862: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9863: lval=-1;
1.311 brouard 9864: coqvar[iv][i]=NAN;
9865: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 9866: }else{
1.225 brouard 9867: errno=0;
9868: /* what_kind_of_number(strb); */
9869: dval=strtod(strb,&endptr);
9870: /* if(strb != endptr && *endptr == '\0') */
9871: /* dval=dlval; */
9872: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9873: if( strb[0]=='\0' || (*endptr != '\0')){
9874: 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);
9875: 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);
9876: return 1;
9877: }
9878: coqvar[iv][i]=dval;
1.226 brouard 9879: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9880: }
9881: strcpy(line,stra);
9882: }/* end loop nqv */
1.136 brouard 9883:
1.223 brouard 9884: /* Covariate values */
1.136 brouard 9885: for (j=ncovcol;j>=1;j--){
9886: cutv(stra, strb,line,' ');
1.223 brouard 9887: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9888: lval=-1;
1.136 brouard 9889: }else{
1.225 brouard 9890: errno=0;
9891: lval=strtol(strb,&endptr,10);
9892: if( strb[0]=='\0' || (*endptr != '\0')){
9893: 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);
9894: 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);
9895: return 1;
9896: }
1.136 brouard 9897: }
9898: if(lval <-1 || lval >1){
1.225 brouard 9899: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9900: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9901: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9902: For example, for multinomial values like 1, 2 and 3,\n \
9903: build V1=0 V2=0 for the reference value (1),\n \
9904: V1=1 V2=0 for (2) \n \
1.136 brouard 9905: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9906: output of IMaCh is often meaningless.\n \
1.136 brouard 9907: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9908: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9909: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9910: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9911: For example, for multinomial values like 1, 2 and 3,\n \
9912: build V1=0 V2=0 for the reference value (1),\n \
9913: V1=1 V2=0 for (2) \n \
1.136 brouard 9914: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9915: output of IMaCh is often meaningless.\n \
1.136 brouard 9916: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9917: return 1;
1.136 brouard 9918: }
9919: covar[j][i]=(double)(lval);
9920: strcpy(line,stra);
9921: }
9922: lstra=strlen(stra);
1.225 brouard 9923:
1.136 brouard 9924: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9925: stratrunc = &(stra[lstra-9]);
9926: num[i]=atol(stratrunc);
9927: }
9928: else
9929: num[i]=atol(stra);
9930: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9931: 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;}*/
9932:
9933: i=i+1;
9934: } /* End loop reading data */
1.225 brouard 9935:
1.136 brouard 9936: *imax=i-1; /* Number of individuals */
9937: fclose(fic);
1.225 brouard 9938:
1.136 brouard 9939: return (0);
1.164 brouard 9940: /* endread: */
1.225 brouard 9941: printf("Exiting readdata: ");
9942: fclose(fic);
9943: return (1);
1.223 brouard 9944: }
1.126 brouard 9945:
1.234 brouard 9946: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9947: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9948: while (*p2 == ' ')
1.234 brouard 9949: p2++;
9950: /* while ((*p1++ = *p2++) !=0) */
9951: /* ; */
9952: /* do */
9953: /* while (*p2 == ' ') */
9954: /* p2++; */
9955: /* while (*p1++ == *p2++); */
9956: *stri=p2;
1.145 brouard 9957: }
9958:
1.235 brouard 9959: int decoderesult ( char resultline[], int nres)
1.230 brouard 9960: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9961: {
1.235 brouard 9962: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9963: char resultsav[MAXLINE];
1.234 brouard 9964: int resultmodel[MAXLINE];
9965: int modelresult[MAXLINE];
1.230 brouard 9966: char stra[80], strb[80], strc[80], strd[80],stre[80];
9967:
1.234 brouard 9968: removefirstspace(&resultline);
1.230 brouard 9969:
9970: if (strstr(resultline,"v") !=0){
9971: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9972: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9973: return 1;
9974: }
9975: trimbb(resultsav, resultline);
9976: if (strlen(resultsav) >1){
9977: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9978: }
1.253 brouard 9979: if(j == 0){ /* Resultline but no = */
9980: TKresult[nres]=0; /* Combination for the nresult and the model */
9981: return (0);
9982: }
1.234 brouard 9983: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.318 brouard 9984: 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 9985: 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 9986: }
9987: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9988: if(nbocc(resultsav,'=') >1){
1.318 brouard 9989: 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" */
9990: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.234 brouard 9991: }else
9992: cutl(strc,strd,resultsav,'=');
1.318 brouard 9993: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 9994:
1.230 brouard 9995: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 9996: 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 9997: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9998: /* cptcovsel++; */
9999: if (nbocc(stra,'=') >0)
10000: strcpy(resultsav,stra); /* and analyzes it */
10001: }
1.235 brouard 10002: /* Checking for missing or useless values in comparison of current model needs */
1.318 brouard 10003: for(k1=1; k1<= cptcovt ;k1++){ /* Loop on model. model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10004: 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 10005: match=0;
1.318 brouard 10006: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10007: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 10008: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10009: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10010: break;
10011: }
10012: }
10013: if(match == 0){
1.310 brouard 10014: printf("Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
10015: fprintf(ficlog,"Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
10016: return 1;
1.234 brouard 10017: }
10018: }
10019: }
1.235 brouard 10020: /* Checking for missing or useless values in comparison of current model needs */
1.318 brouard 10021: 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 10022: match=0;
1.318 brouard 10023: 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 10024: if(Typevar[k1]==0){ /* Single */
1.237 brouard 10025: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.318 brouard 10026: 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 10027: ++match;
10028: }
10029: }
10030: }
10031: if(match == 0){
10032: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
1.310 brouard 10033: fprintf(ficlog,"Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
10034: return 1;
1.234 brouard 10035: }else if(match > 1){
10036: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 10037: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
10038: return 1;
1.234 brouard 10039: }
10040: }
1.235 brouard 10041:
1.234 brouard 10042: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10043: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10044: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
10045: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
10046: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10047: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
10048: /* 1 0 0 0 */
10049: /* 2 1 0 0 */
10050: /* 3 0 1 0 */
10051: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
10052: /* 5 0 0 1 */
10053: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
10054: /* 7 0 1 1 */
10055: /* 8 1 1 1 */
1.237 brouard 10056: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
10057: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
10058: /* V5*age V5 known which value for nres? */
10059: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.318 brouard 10060: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* loop on model line */
1.235 brouard 10061: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 10062: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 10063: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
10064: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 10065: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
10066: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
10067: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 10068: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
10069: k4++;;
10070: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
1.318 brouard 10071: k3q= resultmodel[k1]; /* resultmodel[1(V5)] = 25.1=k3q */
10072: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.237 brouard 10073: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
10074: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
10075: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 10076: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
10077: k4q++;;
10078: }
10079: }
1.234 brouard 10080:
1.235 brouard 10081: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 10082: return (0);
10083: }
1.235 brouard 10084:
1.230 brouard 10085: int decodemodel( char model[], int lastobs)
10086: /**< This routine decodes the model and returns:
1.224 brouard 10087: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
10088: * - nagesqr = 1 if age*age in the model, otherwise 0.
10089: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
10090: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
10091: * - cptcovage number of covariates with age*products =2
10092: * - cptcovs number of simple covariates
10093: * - 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
10094: * which is a new column after the 9 (ncovcol) variables.
1.319 brouard 10095: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 10096: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
10097: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
10098: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
10099: */
1.319 brouard 10100: /* 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 10101: {
1.238 brouard 10102: int i, j, k, ks, v;
1.227 brouard 10103: int j1, k1, k2, k3, k4;
1.136 brouard 10104: char modelsav[80];
1.145 brouard 10105: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 10106: char *strpt;
1.136 brouard 10107:
1.145 brouard 10108: /*removespace(model);*/
1.136 brouard 10109: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 10110: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 10111: if (strstr(model,"AGE") !=0){
1.192 brouard 10112: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
10113: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 10114: return 1;
10115: }
1.141 brouard 10116: if (strstr(model,"v") !=0){
10117: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
10118: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
10119: return 1;
10120: }
1.187 brouard 10121: strcpy(modelsav,model);
10122: if ((strpt=strstr(model,"age*age")) !=0){
10123: printf(" strpt=%s, model=%s\n",strpt, model);
10124: if(strpt != model){
1.234 brouard 10125: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10126: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10127: corresponding column of parameters.\n",model);
1.234 brouard 10128: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10129: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10130: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 10131: return 1;
1.225 brouard 10132: }
1.187 brouard 10133: nagesqr=1;
10134: if (strstr(model,"+age*age") !=0)
1.234 brouard 10135: substrchaine(modelsav, model, "+age*age");
1.187 brouard 10136: else if (strstr(model,"age*age+") !=0)
1.234 brouard 10137: substrchaine(modelsav, model, "age*age+");
1.187 brouard 10138: else
1.234 brouard 10139: substrchaine(modelsav, model, "age*age");
1.187 brouard 10140: }else
10141: nagesqr=0;
10142: if (strlen(modelsav) >1){
10143: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
10144: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 10145: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 10146: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 10147: * cst, age and age*age
10148: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
10149: /* including age products which are counted in cptcovage.
10150: * but the covariates which are products must be treated
10151: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 10152: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
10153: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 10154:
10155:
1.187 brouard 10156: /* Design
10157: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
10158: * < ncovcol=8 >
10159: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
10160: * k= 1 2 3 4 5 6 7 8
10161: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
10162: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 10163: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
10164: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 10165: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
10166: * Tage[++cptcovage]=k
10167: * if products, new covar are created after ncovcol with k1
10168: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
10169: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
10170: * 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
10171: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
10172: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
10173: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
10174: * < ncovcol=8 >
10175: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
10176: * k= 1 2 3 4 5 6 7 8 9 10 11 12
10177: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 10178: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 10179: * p Tprod[1]@2={ 6, 5}
10180: *p Tvard[1][1]@4= {7, 8, 5, 6}
10181: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
10182: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 10183: *How to reorganize? Tvars(orted)
1.187 brouard 10184: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
10185: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
10186: * {2, 1, 4, 8, 5, 6, 3, 7}
10187: * Struct []
10188: */
1.225 brouard 10189:
1.187 brouard 10190: /* This loop fills the array Tvar from the string 'model'.*/
10191: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
10192: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
10193: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
10194: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
10195: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
10196: /* k=1 Tvar[1]=2 (from V2) */
10197: /* k=5 Tvar[5] */
10198: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 10199: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 10200: /* } */
1.198 brouard 10201: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 10202: /*
10203: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 10204: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
10205: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
10206: }
1.187 brouard 10207: cptcovage=0;
1.319 brouard 10208: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
10209: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
10210: 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" */
10211: if (nbocc(modelsav,'+')==0)
10212: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 10213: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
10214: /*scanf("%d",i);*/
1.319 brouard 10215: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
10216: 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 10217: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
10218: /* covar is not filled and then is empty */
10219: cptcovprod--;
10220: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 10221: 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 10222: Typevar[k]=1; /* 1 for age product */
1.319 brouard 10223: cptcovage++; /* Counts the number of covariates which include age as a product */
10224: 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 10225: /*printf("stre=%s ", stre);*/
10226: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
10227: cptcovprod--;
10228: cutl(stre,strb,strc,'V');
10229: Tvar[k]=atoi(stre);
10230: Typevar[k]=1; /* 1 for age product */
10231: cptcovage++;
10232: Tage[cptcovage]=k;
10233: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
10234: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
10235: cptcovn++;
10236: cptcovprodnoage++;k1++;
10237: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
10238: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
10239: because this model-covariate is a construction we invent a new column
10240: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.319 brouard 10241: If already ncovcol=4 and model=V2 + V1 +V1*V4 +age*V3 +V3*V2
10242: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
10243: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=4 etc */
1.234 brouard 10244: Typevar[k]=2; /* 2 for double fixed dummy covariates */
10245: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
10246: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 10247: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 10248: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
10249: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
10250: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
10251: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
10252: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 10253: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 10254: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
10255: for (i=1; i<=lastobs;i++){
10256: /* Computes the new covariate which is a product of
10257: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
10258: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
10259: }
10260: } /* End age is not in the model */
10261: } /* End if model includes a product */
1.319 brouard 10262: else { /* not a product */
1.234 brouard 10263: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
10264: /* scanf("%d",i);*/
10265: cutl(strd,strc,strb,'V');
10266: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
10267: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
10268: Tvar[k]=atoi(strd);
10269: Typevar[k]=0; /* 0 for simple covariates */
10270: }
10271: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 10272: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 10273: scanf("%d",i);*/
1.187 brouard 10274: } /* end of loop + on total covariates */
10275: } /* end if strlen(modelsave == 0) age*age might exist */
10276: } /* end if strlen(model == 0) */
1.136 brouard 10277:
10278: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
10279: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 10280:
1.136 brouard 10281: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 10282: printf("cptcovprod=%d ", cptcovprod);
10283: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
10284: scanf("%d ",i);*/
10285:
10286:
1.230 brouard 10287: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
10288: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 10289: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
10290: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
10291: k = 1 2 3 4 5 6 7 8 9
10292: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 10293: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 10294: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
10295: Dummy[k] 1 0 0 0 3 1 1 2 3
10296: Tmodelind[combination of covar]=k;
1.225 brouard 10297: */
10298: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 10299: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 10300: /* 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 10301: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 10302: printf("Model=1+age+%s\n\
1.227 brouard 10303: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10304: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10305: 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 10306: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 10307: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10308: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10309: 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 10310: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 10311: 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 */
10312: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 10313: Fixed[k]= 0;
10314: Dummy[k]= 0;
1.225 brouard 10315: ncoveff++;
1.232 brouard 10316: ncovf++;
1.234 brouard 10317: nsd++;
10318: modell[k].maintype= FTYPE;
10319: TvarsD[nsd]=Tvar[k];
10320: TvarsDind[nsd]=k;
10321: TvarF[ncovf]=Tvar[k];
10322: TvarFind[ncovf]=k;
10323: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10324: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10325: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
10326: Fixed[k]= 0;
10327: Dummy[k]= 0;
10328: ncoveff++;
10329: ncovf++;
10330: modell[k].maintype= FTYPE;
10331: TvarF[ncovf]=Tvar[k];
10332: TvarFind[ncovf]=k;
1.230 brouard 10333: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10334: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10335: }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 10336: Fixed[k]= 0;
10337: Dummy[k]= 1;
1.230 brouard 10338: nqfveff++;
1.234 brouard 10339: modell[k].maintype= FTYPE;
10340: modell[k].subtype= FQ;
10341: nsq++;
10342: TvarsQ[nsq]=Tvar[k];
10343: TvarsQind[nsq]=k;
1.232 brouard 10344: ncovf++;
1.234 brouard 10345: TvarF[ncovf]=Tvar[k];
10346: TvarFind[ncovf]=k;
1.231 brouard 10347: 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 10348: 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 10349: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10350: Fixed[k]= 1;
10351: Dummy[k]= 0;
1.225 brouard 10352: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10353: modell[k].maintype= VTYPE;
10354: modell[k].subtype= VD;
10355: nsd++;
10356: TvarsD[nsd]=Tvar[k];
10357: TvarsDind[nsd]=k;
10358: ncovv++; /* Only simple time varying variables */
10359: TvarV[ncovv]=Tvar[k];
1.242 brouard 10360: 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 10361: 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 */
10362: 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 10363: 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);
10364: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10365: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10366: Fixed[k]= 1;
10367: Dummy[k]= 1;
10368: nqtveff++;
10369: modell[k].maintype= VTYPE;
10370: modell[k].subtype= VQ;
10371: ncovv++; /* Only simple time varying variables */
10372: nsq++;
1.319 brouard 10373: TvarsQ[nsq]=Tvar[k]; /* k=1 Tvar=5 nsq=1 TvarsQ[1]=5 */
1.234 brouard 10374: TvarsQind[nsq]=k;
10375: TvarV[ncovv]=Tvar[k];
1.242 brouard 10376: 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 10377: 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 */
10378: 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 10379: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10380: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10381: 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 10382: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 10383: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 10384: ncova++;
10385: TvarA[ncova]=Tvar[k];
10386: TvarAind[ncova]=k;
1.231 brouard 10387: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 10388: Fixed[k]= 2;
10389: Dummy[k]= 2;
10390: modell[k].maintype= ATYPE;
10391: modell[k].subtype= APFD;
10392: /* ncoveff++; */
1.227 brouard 10393: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 10394: Fixed[k]= 2;
10395: Dummy[k]= 3;
10396: modell[k].maintype= ATYPE;
10397: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10398: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10399: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10400: Fixed[k]= 3;
10401: Dummy[k]= 2;
10402: modell[k].maintype= ATYPE;
10403: modell[k].subtype= APVD; /* Product age * varying dummy */
10404: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10405: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10406: Fixed[k]= 3;
10407: Dummy[k]= 3;
10408: modell[k].maintype= ATYPE;
10409: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10410: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10411: }
10412: }else if (Typevar[k] == 2) { /* product without age */
10413: k1=Tposprod[k];
10414: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10415: if(Tvard[k1][2] <=ncovcol){
10416: Fixed[k]= 1;
10417: Dummy[k]= 0;
10418: modell[k].maintype= FTYPE;
10419: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10420: ncovf++; /* Fixed variables without age */
10421: TvarF[ncovf]=Tvar[k];
10422: TvarFind[ncovf]=k;
10423: }else if(Tvard[k1][2] <=ncovcol+nqv){
10424: Fixed[k]= 0; /* or 2 ?*/
10425: Dummy[k]= 1;
10426: modell[k].maintype= FTYPE;
10427: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10428: ncovf++; /* Varying variables without age */
10429: TvarF[ncovf]=Tvar[k];
10430: TvarFind[ncovf]=k;
10431: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10432: Fixed[k]= 1;
10433: Dummy[k]= 0;
10434: modell[k].maintype= VTYPE;
10435: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10436: ncovv++; /* Varying variables without age */
10437: TvarV[ncovv]=Tvar[k];
10438: TvarVind[ncovv]=k;
10439: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10440: Fixed[k]= 1;
10441: Dummy[k]= 1;
10442: modell[k].maintype= VTYPE;
10443: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10444: ncovv++; /* Varying variables without age */
10445: TvarV[ncovv]=Tvar[k];
10446: TvarVind[ncovv]=k;
10447: }
1.227 brouard 10448: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10449: if(Tvard[k1][2] <=ncovcol){
10450: Fixed[k]= 0; /* or 2 ?*/
10451: Dummy[k]= 1;
10452: modell[k].maintype= FTYPE;
10453: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10454: ncovf++; /* Fixed variables without age */
10455: TvarF[ncovf]=Tvar[k];
10456: TvarFind[ncovf]=k;
10457: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10458: Fixed[k]= 1;
10459: Dummy[k]= 1;
10460: modell[k].maintype= VTYPE;
10461: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10462: ncovv++; /* Varying variables without age */
10463: TvarV[ncovv]=Tvar[k];
10464: TvarVind[ncovv]=k;
10465: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10466: Fixed[k]= 1;
10467: Dummy[k]= 1;
10468: modell[k].maintype= VTYPE;
10469: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10470: ncovv++; /* Varying variables without age */
10471: TvarV[ncovv]=Tvar[k];
10472: TvarVind[ncovv]=k;
10473: ncovv++; /* Varying variables without age */
10474: TvarV[ncovv]=Tvar[k];
10475: TvarVind[ncovv]=k;
10476: }
1.227 brouard 10477: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10478: if(Tvard[k1][2] <=ncovcol){
10479: Fixed[k]= 1;
10480: Dummy[k]= 1;
10481: modell[k].maintype= VTYPE;
10482: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10483: ncovv++; /* Varying variables without age */
10484: TvarV[ncovv]=Tvar[k];
10485: TvarVind[ncovv]=k;
10486: }else if(Tvard[k1][2] <=ncovcol+nqv){
10487: Fixed[k]= 1;
10488: Dummy[k]= 1;
10489: modell[k].maintype= VTYPE;
10490: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10491: ncovv++; /* Varying variables without age */
10492: TvarV[ncovv]=Tvar[k];
10493: TvarVind[ncovv]=k;
10494: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10495: Fixed[k]= 1;
10496: Dummy[k]= 0;
10497: modell[k].maintype= VTYPE;
10498: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10499: ncovv++; /* Varying variables without age */
10500: TvarV[ncovv]=Tvar[k];
10501: TvarVind[ncovv]=k;
10502: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10503: Fixed[k]= 1;
10504: Dummy[k]= 1;
10505: modell[k].maintype= VTYPE;
10506: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10507: ncovv++; /* Varying variables without age */
10508: TvarV[ncovv]=Tvar[k];
10509: TvarVind[ncovv]=k;
10510: }
1.227 brouard 10511: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10512: if(Tvard[k1][2] <=ncovcol){
10513: Fixed[k]= 1;
10514: Dummy[k]= 1;
10515: modell[k].maintype= VTYPE;
10516: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10517: ncovv++; /* Varying variables without age */
10518: TvarV[ncovv]=Tvar[k];
10519: TvarVind[ncovv]=k;
10520: }else if(Tvard[k1][2] <=ncovcol+nqv){
10521: Fixed[k]= 1;
10522: Dummy[k]= 1;
10523: modell[k].maintype= VTYPE;
10524: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10525: ncovv++; /* Varying variables without age */
10526: TvarV[ncovv]=Tvar[k];
10527: TvarVind[ncovv]=k;
10528: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10529: Fixed[k]= 1;
10530: Dummy[k]= 1;
10531: modell[k].maintype= VTYPE;
10532: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10533: ncovv++; /* Varying variables without age */
10534: TvarV[ncovv]=Tvar[k];
10535: TvarVind[ncovv]=k;
10536: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10537: Fixed[k]= 1;
10538: Dummy[k]= 1;
10539: modell[k].maintype= VTYPE;
10540: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10541: ncovv++; /* Varying variables without age */
10542: TvarV[ncovv]=Tvar[k];
10543: TvarVind[ncovv]=k;
10544: }
1.227 brouard 10545: }else{
1.240 brouard 10546: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10547: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10548: } /*end k1*/
1.225 brouard 10549: }else{
1.226 brouard 10550: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10551: 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 10552: }
1.227 brouard 10553: 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 10554: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10555: 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]);
10556: }
10557: /* Searching for doublons in the model */
10558: for(k1=1; k1<= cptcovt;k1++){
10559: for(k2=1; k2 <k1;k2++){
1.285 brouard 10560: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10561: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10562: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10563: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10564: 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]);
10565: 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 10566: return(1);
10567: }
10568: }else if (Typevar[k1] ==2){
10569: k3=Tposprod[k1];
10570: k4=Tposprod[k2];
10571: 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])) ){
10572: 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]]);
10573: 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);
10574: return(1);
10575: }
10576: }
1.227 brouard 10577: }
10578: }
1.225 brouard 10579: }
10580: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10581: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10582: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10583: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10584: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10585: /*endread:*/
1.225 brouard 10586: printf("Exiting decodemodel: ");
10587: return (1);
1.136 brouard 10588: }
10589:
1.169 brouard 10590: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10591: {/* Check ages at death */
1.136 brouard 10592: int i, m;
1.218 brouard 10593: int firstone=0;
10594:
1.136 brouard 10595: for (i=1; i<=imx; i++) {
10596: for(m=2; (m<= maxwav); m++) {
10597: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10598: anint[m][i]=9999;
1.216 brouard 10599: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10600: s[m][i]=-1;
1.136 brouard 10601: }
10602: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10603: *nberr = *nberr + 1;
1.218 brouard 10604: if(firstone == 0){
10605: firstone=1;
1.260 brouard 10606: 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 10607: }
1.262 brouard 10608: 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 10609: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10610: }
10611: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10612: (*nberr)++;
1.259 brouard 10613: 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 10614: 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 10615: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10616: }
10617: }
10618: }
10619:
10620: for (i=1; i<=imx; i++) {
10621: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10622: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10623: 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 10624: if (s[m][i] >= nlstate+1) {
1.169 brouard 10625: if(agedc[i]>0){
10626: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10627: agev[m][i]=agedc[i];
1.214 brouard 10628: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10629: }else {
1.136 brouard 10630: if ((int)andc[i]!=9999){
10631: nbwarn++;
10632: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10633: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10634: agev[m][i]=-1;
10635: }
10636: }
1.169 brouard 10637: } /* agedc > 0 */
1.214 brouard 10638: } /* end if */
1.136 brouard 10639: else if(s[m][i] !=9){ /* Standard case, age in fractional
10640: years but with the precision of a month */
10641: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10642: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10643: agev[m][i]=1;
10644: else if(agev[m][i] < *agemin){
10645: *agemin=agev[m][i];
10646: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10647: }
10648: else if(agev[m][i] >*agemax){
10649: *agemax=agev[m][i];
1.156 brouard 10650: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10651: }
10652: /*agev[m][i]=anint[m][i]-annais[i];*/
10653: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10654: } /* en if 9*/
1.136 brouard 10655: else { /* =9 */
1.214 brouard 10656: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10657: agev[m][i]=1;
10658: s[m][i]=-1;
10659: }
10660: }
1.214 brouard 10661: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10662: agev[m][i]=1;
1.214 brouard 10663: else{
10664: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10665: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10666: agev[m][i]=0;
10667: }
10668: } /* End for lastpass */
10669: }
1.136 brouard 10670:
10671: for (i=1; i<=imx; i++) {
10672: for(m=firstpass; (m<=lastpass); m++){
10673: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10674: (*nberr)++;
1.136 brouard 10675: 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);
10676: 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);
10677: return 1;
10678: }
10679: }
10680: }
10681:
10682: /*for (i=1; i<=imx; i++){
10683: for (m=firstpass; (m<lastpass); m++){
10684: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10685: }
10686:
10687: }*/
10688:
10689:
1.139 brouard 10690: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10691: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10692:
10693: return (0);
1.164 brouard 10694: /* endread:*/
1.136 brouard 10695: printf("Exiting calandcheckages: ");
10696: return (1);
10697: }
10698:
1.172 brouard 10699: #if defined(_MSC_VER)
10700: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10701: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10702: //#include "stdafx.h"
10703: //#include <stdio.h>
10704: //#include <tchar.h>
10705: //#include <windows.h>
10706: //#include <iostream>
10707: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10708:
10709: LPFN_ISWOW64PROCESS fnIsWow64Process;
10710:
10711: BOOL IsWow64()
10712: {
10713: BOOL bIsWow64 = FALSE;
10714:
10715: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10716: // (HANDLE, PBOOL);
10717:
10718: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10719:
10720: HMODULE module = GetModuleHandle(_T("kernel32"));
10721: const char funcName[] = "IsWow64Process";
10722: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10723: GetProcAddress(module, funcName);
10724:
10725: if (NULL != fnIsWow64Process)
10726: {
10727: if (!fnIsWow64Process(GetCurrentProcess(),
10728: &bIsWow64))
10729: //throw std::exception("Unknown error");
10730: printf("Unknown error\n");
10731: }
10732: return bIsWow64 != FALSE;
10733: }
10734: #endif
1.177 brouard 10735:
1.191 brouard 10736: void syscompilerinfo(int logged)
1.292 brouard 10737: {
10738: #include <stdint.h>
10739:
10740: /* #include "syscompilerinfo.h"*/
1.185 brouard 10741: /* command line Intel compiler 32bit windows, XP compatible:*/
10742: /* /GS /W3 /Gy
10743: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10744: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10745: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10746: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10747: */
10748: /* 64 bits */
1.185 brouard 10749: /*
10750: /GS /W3 /Gy
10751: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10752: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10753: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10754: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10755: /* Optimization are useless and O3 is slower than O2 */
10756: /*
10757: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10758: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10759: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10760: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10761: */
1.186 brouard 10762: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10763: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10764: /PDB:"visual studio
10765: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10766: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10767: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10768: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10769: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10770: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10771: uiAccess='false'"
10772: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10773: /NOLOGO /TLBID:1
10774: */
1.292 brouard 10775:
10776:
1.177 brouard 10777: #if defined __INTEL_COMPILER
1.178 brouard 10778: #if defined(__GNUC__)
10779: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10780: #endif
1.177 brouard 10781: #elif defined(__GNUC__)
1.179 brouard 10782: #ifndef __APPLE__
1.174 brouard 10783: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10784: #endif
1.177 brouard 10785: struct utsname sysInfo;
1.178 brouard 10786: int cross = CROSS;
10787: if (cross){
10788: printf("Cross-");
1.191 brouard 10789: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10790: }
1.174 brouard 10791: #endif
10792:
1.191 brouard 10793: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10794: #if defined(__clang__)
1.191 brouard 10795: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10796: #endif
10797: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10798: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10799: #endif
10800: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10801: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10802: #endif
10803: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10804: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10805: #endif
10806: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10807: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10808: #endif
10809: #if defined(_MSC_VER)
1.191 brouard 10810: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10811: #endif
10812: #if defined(__PGI)
1.191 brouard 10813: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10814: #endif
10815: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10816: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10817: #endif
1.191 brouard 10818: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10819:
1.167 brouard 10820: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10821: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10822: // Windows (x64 and x86)
1.191 brouard 10823: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10824: #elif __unix__ // all unices, not all compilers
10825: // Unix
1.191 brouard 10826: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10827: #elif __linux__
10828: // linux
1.191 brouard 10829: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10830: #elif __APPLE__
1.174 brouard 10831: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10832: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10833: #endif
10834:
10835: /* __MINGW32__ */
10836: /* __CYGWIN__ */
10837: /* __MINGW64__ */
10838: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10839: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10840: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10841: /* _WIN64 // Defined for applications for Win64. */
10842: /* _M_X64 // Defined for compilations that target x64 processors. */
10843: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10844:
1.167 brouard 10845: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10846: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10847: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10848: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10849: #else
1.191 brouard 10850: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10851: #endif
10852:
1.169 brouard 10853: #if defined(__GNUC__)
10854: # if defined(__GNUC_PATCHLEVEL__)
10855: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10856: + __GNUC_MINOR__ * 100 \
10857: + __GNUC_PATCHLEVEL__)
10858: # else
10859: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10860: + __GNUC_MINOR__ * 100)
10861: # endif
1.174 brouard 10862: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10863: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10864:
10865: if (uname(&sysInfo) != -1) {
10866: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10867: 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 10868: }
10869: else
10870: perror("uname() error");
1.179 brouard 10871: //#ifndef __INTEL_COMPILER
10872: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10873: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10874: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10875: #endif
1.169 brouard 10876: #endif
1.172 brouard 10877:
1.286 brouard 10878: // void main ()
1.172 brouard 10879: // {
1.169 brouard 10880: #if defined(_MSC_VER)
1.174 brouard 10881: if (IsWow64()){
1.191 brouard 10882: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10883: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10884: }
10885: else{
1.191 brouard 10886: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10887: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10888: }
1.172 brouard 10889: // printf("\nPress Enter to continue...");
10890: // getchar();
10891: // }
10892:
1.169 brouard 10893: #endif
10894:
1.167 brouard 10895:
1.219 brouard 10896: }
1.136 brouard 10897:
1.219 brouard 10898: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10899: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10900: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10901: /* double ftolpl = 1.e-10; */
1.180 brouard 10902: double age, agebase, agelim;
1.203 brouard 10903: double tot;
1.180 brouard 10904:
1.202 brouard 10905: strcpy(filerespl,"PL_");
10906: strcat(filerespl,fileresu);
10907: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10908: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10909: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10910: }
1.288 brouard 10911: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10912: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10913: pstamp(ficrespl);
1.288 brouard 10914: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10915: fprintf(ficrespl,"#Age ");
10916: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10917: fprintf(ficrespl,"\n");
1.180 brouard 10918:
1.219 brouard 10919: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10920:
1.219 brouard 10921: agebase=ageminpar;
10922: agelim=agemaxpar;
1.180 brouard 10923:
1.227 brouard 10924: /* i1=pow(2,ncoveff); */
1.234 brouard 10925: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10926: if (cptcovn < 1){i1=1;}
1.180 brouard 10927:
1.238 brouard 10928: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10929: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10930: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10931: continue;
1.235 brouard 10932:
1.238 brouard 10933: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10934: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10935: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10936: /* k=k+1; */
10937: /* to clean */
10938: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10939: fprintf(ficrespl,"#******");
10940: printf("#******");
10941: fprintf(ficlog,"#******");
10942: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10943: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10944: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10945: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10946: }
10947: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10948: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10949: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10950: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10951: }
10952: fprintf(ficrespl,"******\n");
10953: printf("******\n");
10954: fprintf(ficlog,"******\n");
10955: if(invalidvarcomb[k]){
10956: printf("\nCombination (%d) ignored because no case \n",k);
10957: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10958: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10959: continue;
10960: }
1.219 brouard 10961:
1.238 brouard 10962: fprintf(ficrespl,"#Age ");
10963: for(j=1;j<=cptcoveff;j++) {
10964: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10965: }
10966: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10967: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10968:
1.238 brouard 10969: for (age=agebase; age<=agelim; age++){
10970: /* for (age=agebase; age<=agebase; age++){ */
10971: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10972: fprintf(ficrespl,"%.0f ",age );
10973: for(j=1;j<=cptcoveff;j++)
10974: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10975: tot=0.;
10976: for(i=1; i<=nlstate;i++){
10977: tot += prlim[i][i];
10978: fprintf(ficrespl," %.5f", prlim[i][i]);
10979: }
10980: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10981: } /* Age */
10982: /* was end of cptcod */
10983: } /* cptcov */
10984: } /* nres */
1.219 brouard 10985: return 0;
1.180 brouard 10986: }
10987:
1.218 brouard 10988: 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 10989: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10990:
10991: /* Computes the back prevalence limit for any combination of covariate values
10992: * at any age between ageminpar and agemaxpar
10993: */
1.235 brouard 10994: int i, j, k, i1, nres=0 ;
1.217 brouard 10995: /* double ftolpl = 1.e-10; */
10996: double age, agebase, agelim;
10997: double tot;
1.218 brouard 10998: /* double ***mobaverage; */
10999: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 11000:
11001: strcpy(fileresplb,"PLB_");
11002: strcat(fileresplb,fileresu);
11003: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 11004: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
11005: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 11006: }
1.288 brouard 11007: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
11008: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 11009: pstamp(ficresplb);
1.288 brouard 11010: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 11011: fprintf(ficresplb,"#Age ");
11012: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
11013: fprintf(ficresplb,"\n");
11014:
1.218 brouard 11015:
11016: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
11017:
11018: agebase=ageminpar;
11019: agelim=agemaxpar;
11020:
11021:
1.227 brouard 11022: i1=pow(2,cptcoveff);
1.218 brouard 11023: if (cptcovn < 1){i1=1;}
1.227 brouard 11024:
1.238 brouard 11025: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11026: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11027: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11028: continue;
11029: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
11030: fprintf(ficresplb,"#******");
11031: printf("#******");
11032: fprintf(ficlog,"#******");
11033: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
11034: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11035: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11036: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11037: }
11038: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11039: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11040: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11041: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11042: }
11043: fprintf(ficresplb,"******\n");
11044: printf("******\n");
11045: fprintf(ficlog,"******\n");
11046: if(invalidvarcomb[k]){
11047: printf("\nCombination (%d) ignored because no cases \n",k);
11048: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
11049: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
11050: continue;
11051: }
1.218 brouard 11052:
1.238 brouard 11053: fprintf(ficresplb,"#Age ");
11054: for(j=1;j<=cptcoveff;j++) {
11055: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11056: }
11057: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
11058: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 11059:
11060:
1.238 brouard 11061: for (age=agebase; age<=agelim; age++){
11062: /* for (age=agebase; age<=agebase; age++){ */
11063: if(mobilavproj > 0){
11064: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
11065: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11066: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 11067: }else if (mobilavproj == 0){
11068: 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);
11069: 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);
11070: exit(1);
11071: }else{
11072: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11073: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 11074: /* printf("TOTOT\n"); */
11075: /* exit(1); */
1.238 brouard 11076: }
11077: fprintf(ficresplb,"%.0f ",age );
11078: for(j=1;j<=cptcoveff;j++)
11079: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11080: tot=0.;
11081: for(i=1; i<=nlstate;i++){
11082: tot += bprlim[i][i];
11083: fprintf(ficresplb," %.5f", bprlim[i][i]);
11084: }
11085: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
11086: } /* Age */
11087: /* was end of cptcod */
1.255 brouard 11088: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 11089: } /* end of any combination */
11090: } /* end of nres */
1.218 brouard 11091: /* hBijx(p, bage, fage); */
11092: /* fclose(ficrespijb); */
11093:
11094: return 0;
1.217 brouard 11095: }
1.218 brouard 11096:
1.180 brouard 11097: int hPijx(double *p, int bage, int fage){
11098: /*------------- h Pij x at various ages ------------*/
11099:
11100: int stepsize;
11101: int agelim;
11102: int hstepm;
11103: int nhstepm;
1.235 brouard 11104: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 11105:
11106: double agedeb;
11107: double ***p3mat;
11108:
1.201 brouard 11109: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 11110: if((ficrespij=fopen(filerespij,"w"))==NULL) {
11111: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
11112: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
11113: }
11114: printf("Computing pij: result on file '%s' \n", filerespij);
11115: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
11116:
11117: stepsize=(int) (stepm+YEARM-1)/YEARM;
11118: /*if (stepm<=24) stepsize=2;*/
11119:
11120: agelim=AGESUP;
11121: hstepm=stepsize*YEARM; /* Every year of age */
11122: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 11123:
1.180 brouard 11124: /* hstepm=1; aff par mois*/
11125: pstamp(ficrespij);
11126: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 11127: i1= pow(2,cptcoveff);
1.218 brouard 11128: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11129: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11130: /* k=k+1; */
1.235 brouard 11131: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11132: for(k=1; k<=i1;k++){
1.253 brouard 11133: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11134: continue;
1.183 brouard 11135: fprintf(ficrespij,"\n#****** ");
1.227 brouard 11136: for(j=1;j<=cptcoveff;j++)
1.198 brouard 11137: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11138: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11139: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11140: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11141: }
1.183 brouard 11142: fprintf(ficrespij,"******\n");
11143:
11144: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
11145: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
11146: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
11147:
11148: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 11149:
1.183 brouard 11150: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11151: oldm=oldms;savm=savms;
1.235 brouard 11152: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 11153: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
11154: for(i=1; i<=nlstate;i++)
11155: for(j=1; j<=nlstate+ndeath;j++)
11156: fprintf(ficrespij," %1d-%1d",i,j);
11157: fprintf(ficrespij,"\n");
11158: for (h=0; h<=nhstepm; h++){
11159: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11160: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 11161: for(i=1; i<=nlstate;i++)
11162: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 11163: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 11164: fprintf(ficrespij,"\n");
11165: }
1.183 brouard 11166: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11167: fprintf(ficrespij,"\n");
11168: }
1.180 brouard 11169: /*}*/
11170: }
1.218 brouard 11171: return 0;
1.180 brouard 11172: }
1.218 brouard 11173:
11174: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 11175: /*------------- h Bij x at various ages ------------*/
11176:
11177: int stepsize;
1.218 brouard 11178: /* int agelim; */
11179: int ageminl;
1.217 brouard 11180: int hstepm;
11181: int nhstepm;
1.238 brouard 11182: int h, i, i1, j, k, nres;
1.218 brouard 11183:
1.217 brouard 11184: double agedeb;
11185: double ***p3mat;
1.218 brouard 11186:
11187: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
11188: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
11189: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11190: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11191: }
11192: printf("Computing pij back: result on file '%s' \n", filerespijb);
11193: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
11194:
11195: stepsize=(int) (stepm+YEARM-1)/YEARM;
11196: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 11197:
1.218 brouard 11198: /* agelim=AGESUP; */
1.289 brouard 11199: ageminl=AGEINF; /* was 30 */
1.218 brouard 11200: hstepm=stepsize*YEARM; /* Every year of age */
11201: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
11202:
11203: /* hstepm=1; aff par mois*/
11204: pstamp(ficrespijb);
1.255 brouard 11205: 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 11206: i1= pow(2,cptcoveff);
1.218 brouard 11207: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11208: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11209: /* k=k+1; */
1.238 brouard 11210: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11211: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11212: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11213: continue;
11214: fprintf(ficrespijb,"\n#****** ");
11215: for(j=1;j<=cptcoveff;j++)
11216: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11217: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11218: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11219: }
11220: fprintf(ficrespijb,"******\n");
1.264 brouard 11221: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 11222: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
11223: continue;
11224: }
11225:
11226: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
11227: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
11228: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 11229: 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 */
11230: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 11231:
11232: /* nhstepm=nhstepm*YEARM; aff par mois*/
11233:
1.266 brouard 11234: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
11235: /* and memory limitations if stepm is small */
11236:
1.238 brouard 11237: /* oldm=oldms;savm=savms; */
11238: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.325 brouard 11239: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
1.238 brouard 11240: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 11241: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 11242: for(i=1; i<=nlstate;i++)
11243: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 11244: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 11245: fprintf(ficrespijb,"\n");
1.238 brouard 11246: for (h=0; h<=nhstepm; h++){
11247: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11248: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
11249: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
11250: for(i=1; i<=nlstate;i++)
11251: for(j=1; j<=nlstate+ndeath;j++)
1.325 brouard 11252: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.238 brouard 11253: fprintf(ficrespijb,"\n");
11254: }
11255: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11256: fprintf(ficrespijb,"\n");
11257: } /* end age deb */
11258: } /* end combination */
11259: } /* end nres */
1.218 brouard 11260: return 0;
11261: } /* hBijx */
1.217 brouard 11262:
1.180 brouard 11263:
1.136 brouard 11264: /***********************************************/
11265: /**************** Main Program *****************/
11266: /***********************************************/
11267:
11268: int main(int argc, char *argv[])
11269: {
11270: #ifdef GSL
11271: const gsl_multimin_fminimizer_type *T;
11272: size_t iteri = 0, it;
11273: int rval = GSL_CONTINUE;
11274: int status = GSL_SUCCESS;
11275: double ssval;
11276: #endif
11277: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 11278: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
11279: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 11280: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 11281: int jj, ll, li, lj, lk;
1.136 brouard 11282: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 11283: int num_filled;
1.136 brouard 11284: int itimes;
11285: int NDIM=2;
11286: int vpopbased=0;
1.235 brouard 11287: int nres=0;
1.258 brouard 11288: int endishere=0;
1.277 brouard 11289: int noffset=0;
1.274 brouard 11290: int ncurrv=0; /* Temporary variable */
11291:
1.164 brouard 11292: char ca[32], cb[32];
1.136 brouard 11293: /* FILE *fichtm; *//* Html File */
11294: /* FILE *ficgp;*/ /*Gnuplot File */
11295: struct stat info;
1.191 brouard 11296: double agedeb=0.;
1.194 brouard 11297:
11298: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 11299: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 11300:
1.165 brouard 11301: double fret;
1.191 brouard 11302: double dum=0.; /* Dummy variable */
1.136 brouard 11303: double ***p3mat;
1.218 brouard 11304: /* double ***mobaverage; */
1.319 brouard 11305: double wald;
1.164 brouard 11306:
11307: char line[MAXLINE];
1.197 brouard 11308: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
11309:
1.234 brouard 11310: char modeltemp[MAXLINE];
1.230 brouard 11311: char resultline[MAXLINE];
11312:
1.136 brouard 11313: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 11314: char *tok, *val; /* pathtot */
1.290 brouard 11315: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 11316: int c, h , cpt, c2;
1.191 brouard 11317: int jl=0;
11318: int i1, j1, jk, stepsize=0;
1.194 brouard 11319: int count=0;
11320:
1.164 brouard 11321: int *tab;
1.136 brouard 11322: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 11323: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
11324: /* double anprojf, mprojf, jprojf; */
11325: /* double jintmean,mintmean,aintmean; */
11326: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11327: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11328: double yrfproj= 10.0; /* Number of years of forward projections */
11329: double yrbproj= 10.0; /* Number of years of backward projections */
11330: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11331: int mobilav=0,popforecast=0;
1.191 brouard 11332: int hstepm=0, nhstepm=0;
1.136 brouard 11333: int agemortsup;
11334: float sumlpop=0.;
11335: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11336: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11337:
1.191 brouard 11338: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11339: double ftolpl=FTOL;
11340: double **prlim;
1.217 brouard 11341: double **bprlim;
1.317 brouard 11342: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
11343: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 11344: double ***paramstart; /* Matrix of starting parameter values */
11345: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11346: double **matcov; /* Matrix of covariance */
1.203 brouard 11347: double **hess; /* Hessian matrix */
1.136 brouard 11348: double ***delti3; /* Scale */
11349: double *delti; /* Scale */
11350: double ***eij, ***vareij;
11351: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11352:
1.136 brouard 11353: double *epj, vepp;
1.164 brouard 11354:
1.273 brouard 11355: double dateprev1, dateprev2;
1.296 brouard 11356: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11357: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11358:
1.217 brouard 11359:
1.136 brouard 11360: double **ximort;
1.145 brouard 11361: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11362: int *dcwave;
11363:
1.164 brouard 11364: char z[1]="c";
1.136 brouard 11365:
11366: /*char *strt;*/
11367: char strtend[80];
1.126 brouard 11368:
1.164 brouard 11369:
1.126 brouard 11370: /* setlocale (LC_ALL, ""); */
11371: /* bindtextdomain (PACKAGE, LOCALEDIR); */
11372: /* textdomain (PACKAGE); */
11373: /* setlocale (LC_CTYPE, ""); */
11374: /* setlocale (LC_MESSAGES, ""); */
11375:
11376: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 11377: rstart_time = time(NULL);
11378: /* (void) gettimeofday(&start_time,&tzp);*/
11379: start_time = *localtime(&rstart_time);
1.126 brouard 11380: curr_time=start_time;
1.157 brouard 11381: /*tml = *localtime(&start_time.tm_sec);*/
11382: /* strcpy(strstart,asctime(&tml)); */
11383: strcpy(strstart,asctime(&start_time));
1.126 brouard 11384:
11385: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 11386: /* tp.tm_sec = tp.tm_sec +86400; */
11387: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 11388: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
11389: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
11390: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 11391: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 11392: /* strt=asctime(&tmg); */
11393: /* printf("Time(after) =%s",strstart); */
11394: /* (void) time (&time_value);
11395: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
11396: * tm = *localtime(&time_value);
11397: * strstart=asctime(&tm);
11398: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11399: */
11400:
11401: nberr=0; /* Number of errors and warnings */
11402: nbwarn=0;
1.184 brouard 11403: #ifdef WIN32
11404: _getcwd(pathcd, size);
11405: #else
1.126 brouard 11406: getcwd(pathcd, size);
1.184 brouard 11407: #endif
1.191 brouard 11408: syscompilerinfo(0);
1.196 brouard 11409: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11410: if(argc <=1){
11411: printf("\nEnter the parameter file name: ");
1.205 brouard 11412: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11413: printf("ERROR Empty parameter file name\n");
11414: goto end;
11415: }
1.126 brouard 11416: i=strlen(pathr);
11417: if(pathr[i-1]=='\n')
11418: pathr[i-1]='\0';
1.156 brouard 11419: i=strlen(pathr);
1.205 brouard 11420: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11421: pathr[i-1]='\0';
1.205 brouard 11422: }
11423: i=strlen(pathr);
11424: if( i==0 ){
11425: printf("ERROR Empty parameter file name\n");
11426: goto end;
11427: }
11428: for (tok = pathr; tok != NULL; ){
1.126 brouard 11429: printf("Pathr |%s|\n",pathr);
11430: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11431: printf("val= |%s| pathr=%s\n",val,pathr);
11432: strcpy (pathtot, val);
11433: if(pathr[0] == '\0') break; /* Dirty */
11434: }
11435: }
1.281 brouard 11436: else if (argc<=2){
11437: strcpy(pathtot,argv[1]);
11438: }
1.126 brouard 11439: else{
11440: strcpy(pathtot,argv[1]);
1.281 brouard 11441: strcpy(z,argv[2]);
11442: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11443: }
11444: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11445: /*cygwin_split_path(pathtot,path,optionfile);
11446: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11447: /* cutv(path,optionfile,pathtot,'\\');*/
11448:
11449: /* Split argv[0], imach program to get pathimach */
11450: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11451: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11452: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11453: /* strcpy(pathimach,argv[0]); */
11454: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11455: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11456: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11457: #ifdef WIN32
11458: _chdir(path); /* Can be a relative path */
11459: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11460: #else
1.126 brouard 11461: chdir(path); /* Can be a relative path */
1.184 brouard 11462: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11463: #endif
11464: printf("Current directory %s!\n",pathcd);
1.126 brouard 11465: strcpy(command,"mkdir ");
11466: strcat(command,optionfilefiname);
11467: if((outcmd=system(command)) != 0){
1.169 brouard 11468: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11469: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11470: /* fclose(ficlog); */
11471: /* exit(1); */
11472: }
11473: /* if((imk=mkdir(optionfilefiname))<0){ */
11474: /* perror("mkdir"); */
11475: /* } */
11476:
11477: /*-------- arguments in the command line --------*/
11478:
1.186 brouard 11479: /* Main Log file */
1.126 brouard 11480: strcat(filelog, optionfilefiname);
11481: strcat(filelog,".log"); /* */
11482: if((ficlog=fopen(filelog,"w"))==NULL) {
11483: printf("Problem with logfile %s\n",filelog);
11484: goto end;
11485: }
11486: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11487: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11488: fprintf(ficlog,"\nEnter the parameter file name: \n");
11489: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11490: path=%s \n\
11491: optionfile=%s\n\
11492: optionfilext=%s\n\
1.156 brouard 11493: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11494:
1.197 brouard 11495: syscompilerinfo(1);
1.167 brouard 11496:
1.126 brouard 11497: printf("Local time (at start):%s",strstart);
11498: fprintf(ficlog,"Local time (at start): %s",strstart);
11499: fflush(ficlog);
11500: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11501: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11502:
11503: /* */
11504: strcpy(fileres,"r");
11505: strcat(fileres, optionfilefiname);
1.201 brouard 11506: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11507: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11508: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11509:
1.186 brouard 11510: /* Main ---------arguments file --------*/
1.126 brouard 11511:
11512: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11513: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11514: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11515: fflush(ficlog);
1.149 brouard 11516: /* goto end; */
11517: exit(70);
1.126 brouard 11518: }
11519:
11520: strcpy(filereso,"o");
1.201 brouard 11521: strcat(filereso,fileresu);
1.126 brouard 11522: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11523: printf("Problem with Output resultfile: %s\n", filereso);
11524: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11525: fflush(ficlog);
11526: goto end;
11527: }
1.278 brouard 11528: /*-------- Rewriting parameter file ----------*/
11529: strcpy(rfileres,"r"); /* "Rparameterfile */
11530: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11531: strcat(rfileres,"."); /* */
11532: strcat(rfileres,optionfilext); /* Other files have txt extension */
11533: if((ficres =fopen(rfileres,"w"))==NULL) {
11534: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11535: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11536: fflush(ficlog);
11537: goto end;
11538: }
11539: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11540:
1.278 brouard 11541:
1.126 brouard 11542: /* Reads comments: lines beginning with '#' */
11543: numlinepar=0;
1.277 brouard 11544: /* Is it a BOM UTF-8 Windows file? */
11545: /* First parameter line */
1.197 brouard 11546: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11547: noffset=0;
11548: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11549: {
11550: noffset=noffset+3;
11551: printf("# File is an UTF8 Bom.\n"); // 0xBF
11552: }
1.302 brouard 11553: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
11554: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 11555: {
11556: noffset=noffset+2;
11557: printf("# File is an UTF16BE BOM file\n");
11558: }
11559: else if( line[0] == 0 && line[1] == 0)
11560: {
11561: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11562: noffset=noffset+4;
11563: printf("# File is an UTF16BE BOM file\n");
11564: }
11565: } else{
11566: ;/*printf(" Not a BOM file\n");*/
11567: }
11568:
1.197 brouard 11569: /* If line starts with a # it is a comment */
1.277 brouard 11570: if (line[noffset] == '#') {
1.197 brouard 11571: numlinepar++;
11572: fputs(line,stdout);
11573: fputs(line,ficparo);
1.278 brouard 11574: fputs(line,ficres);
1.197 brouard 11575: fputs(line,ficlog);
11576: continue;
11577: }else
11578: break;
11579: }
11580: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11581: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11582: if (num_filled != 5) {
11583: printf("Should be 5 parameters\n");
1.283 brouard 11584: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11585: }
1.126 brouard 11586: numlinepar++;
1.197 brouard 11587: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11588: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11589: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11590: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11591: }
11592: /* Second parameter line */
11593: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11594: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11595: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11596: if (line[0] == '#') {
11597: numlinepar++;
1.283 brouard 11598: printf("%s",line);
11599: fprintf(ficres,"%s",line);
11600: fprintf(ficparo,"%s",line);
11601: fprintf(ficlog,"%s",line);
1.197 brouard 11602: continue;
11603: }else
11604: break;
11605: }
1.223 brouard 11606: 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", \
11607: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11608: if (num_filled != 11) {
11609: 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 11610: printf("but line=%s\n",line);
1.283 brouard 11611: 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");
11612: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11613: }
1.286 brouard 11614: if( lastpass > maxwav){
11615: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11616: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11617: fflush(ficlog);
11618: goto end;
11619: }
11620: 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 11621: 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 11622: 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 11623: 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 11624: }
1.203 brouard 11625: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11626: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11627: /* Third parameter line */
11628: while(fgets(line, MAXLINE, ficpar)) {
11629: /* If line starts with a # it is a comment */
11630: if (line[0] == '#') {
11631: numlinepar++;
1.283 brouard 11632: printf("%s",line);
11633: fprintf(ficres,"%s",line);
11634: fprintf(ficparo,"%s",line);
11635: fprintf(ficlog,"%s",line);
1.197 brouard 11636: continue;
11637: }else
11638: break;
11639: }
1.201 brouard 11640: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11641: if (num_filled != 1){
1.302 brouard 11642: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
11643: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 11644: model[0]='\0';
11645: goto end;
11646: }
11647: else{
11648: if (model[0]=='+'){
11649: for(i=1; i<=strlen(model);i++)
11650: modeltemp[i-1]=model[i];
1.201 brouard 11651: strcpy(model,modeltemp);
1.197 brouard 11652: }
11653: }
1.199 brouard 11654: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11655: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11656: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11657: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11658: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11659: }
11660: /* 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); */
11661: /* numlinepar=numlinepar+3; /\* In general *\/ */
11662: /* 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 11663: /* 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); */
11664: /* 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 11665: fflush(ficlog);
1.190 brouard 11666: /* if(model[0]=='#'|| model[0]== '\0'){ */
11667: if(model[0]=='#'){
1.279 brouard 11668: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11669: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11670: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11671: if(mle != -1){
1.279 brouard 11672: 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 11673: exit(1);
11674: }
11675: }
1.126 brouard 11676: while((c=getc(ficpar))=='#' && c!= EOF){
11677: ungetc(c,ficpar);
11678: fgets(line, MAXLINE, ficpar);
11679: numlinepar++;
1.195 brouard 11680: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11681: z[0]=line[1];
11682: }
11683: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11684: fputs(line, stdout);
11685: //puts(line);
1.126 brouard 11686: fputs(line,ficparo);
11687: fputs(line,ficlog);
11688: }
11689: ungetc(c,ficpar);
11690:
11691:
1.290 brouard 11692: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11693: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11694: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11695: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11696: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11697: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11698: v1+v2*age+v2*v3 makes cptcovn = 3
11699: */
11700: if (strlen(model)>1)
1.187 brouard 11701: 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 11702: else
1.187 brouard 11703: ncovmodel=2; /* Constant and age */
1.133 brouard 11704: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11705: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11706: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11707: 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);
11708: 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);
11709: fflush(stdout);
11710: fclose (ficlog);
11711: goto end;
11712: }
1.126 brouard 11713: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11714: delti=delti3[1][1];
11715: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11716: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11717: /* We could also provide initial parameters values giving by simple logistic regression
11718: * only one way, that is without matrix product. We will have nlstate maximizations */
11719: /* for(i=1;i<nlstate;i++){ */
11720: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11721: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11722: /* } */
1.126 brouard 11723: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11724: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11725: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11726: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11727: fclose (ficparo);
11728: fclose (ficlog);
11729: goto end;
11730: exit(0);
1.220 brouard 11731: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11732: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11733: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11734: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11735: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11736: matcov=matrix(1,npar,1,npar);
1.203 brouard 11737: hess=matrix(1,npar,1,npar);
1.220 brouard 11738: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11739: /* Read guessed parameters */
1.126 brouard 11740: /* Reads comments: lines beginning with '#' */
11741: while((c=getc(ficpar))=='#' && c!= EOF){
11742: ungetc(c,ficpar);
11743: fgets(line, MAXLINE, ficpar);
11744: numlinepar++;
1.141 brouard 11745: fputs(line,stdout);
1.126 brouard 11746: fputs(line,ficparo);
11747: fputs(line,ficlog);
11748: }
11749: ungetc(c,ficpar);
11750:
11751: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11752: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11753: for(i=1; i <=nlstate; i++){
1.234 brouard 11754: j=0;
1.126 brouard 11755: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11756: if(jj==i) continue;
11757: j++;
1.292 brouard 11758: while((c=getc(ficpar))=='#' && c!= EOF){
11759: ungetc(c,ficpar);
11760: fgets(line, MAXLINE, ficpar);
11761: numlinepar++;
11762: fputs(line,stdout);
11763: fputs(line,ficparo);
11764: fputs(line,ficlog);
11765: }
11766: ungetc(c,ficpar);
1.234 brouard 11767: fscanf(ficpar,"%1d%1d",&i1,&j1);
11768: if ((i1 != i) || (j1 != jj)){
11769: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11770: It might be a problem of design; if ncovcol and the model are correct\n \
11771: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11772: exit(1);
11773: }
11774: fprintf(ficparo,"%1d%1d",i1,j1);
11775: if(mle==1)
11776: printf("%1d%1d",i,jj);
11777: fprintf(ficlog,"%1d%1d",i,jj);
11778: for(k=1; k<=ncovmodel;k++){
11779: fscanf(ficpar," %lf",¶m[i][j][k]);
11780: if(mle==1){
11781: printf(" %lf",param[i][j][k]);
11782: fprintf(ficlog," %lf",param[i][j][k]);
11783: }
11784: else
11785: fprintf(ficlog," %lf",param[i][j][k]);
11786: fprintf(ficparo," %lf",param[i][j][k]);
11787: }
11788: fscanf(ficpar,"\n");
11789: numlinepar++;
11790: if(mle==1)
11791: printf("\n");
11792: fprintf(ficlog,"\n");
11793: fprintf(ficparo,"\n");
1.126 brouard 11794: }
11795: }
11796: fflush(ficlog);
1.234 brouard 11797:
1.251 brouard 11798: /* Reads parameters values */
1.126 brouard 11799: p=param[1][1];
1.251 brouard 11800: pstart=paramstart[1][1];
1.126 brouard 11801:
11802: /* Reads comments: lines beginning with '#' */
11803: while((c=getc(ficpar))=='#' && c!= EOF){
11804: ungetc(c,ficpar);
11805: fgets(line, MAXLINE, ficpar);
11806: numlinepar++;
1.141 brouard 11807: fputs(line,stdout);
1.126 brouard 11808: fputs(line,ficparo);
11809: fputs(line,ficlog);
11810: }
11811: ungetc(c,ficpar);
11812:
11813: for(i=1; i <=nlstate; i++){
11814: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11815: fscanf(ficpar,"%1d%1d",&i1,&j1);
11816: if ( (i1-i) * (j1-j) != 0){
11817: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11818: exit(1);
11819: }
11820: printf("%1d%1d",i,j);
11821: fprintf(ficparo,"%1d%1d",i1,j1);
11822: fprintf(ficlog,"%1d%1d",i1,j1);
11823: for(k=1; k<=ncovmodel;k++){
11824: fscanf(ficpar,"%le",&delti3[i][j][k]);
11825: printf(" %le",delti3[i][j][k]);
11826: fprintf(ficparo," %le",delti3[i][j][k]);
11827: fprintf(ficlog," %le",delti3[i][j][k]);
11828: }
11829: fscanf(ficpar,"\n");
11830: numlinepar++;
11831: printf("\n");
11832: fprintf(ficparo,"\n");
11833: fprintf(ficlog,"\n");
1.126 brouard 11834: }
11835: }
11836: fflush(ficlog);
1.234 brouard 11837:
1.145 brouard 11838: /* Reads covariance matrix */
1.126 brouard 11839: delti=delti3[1][1];
1.220 brouard 11840:
11841:
1.126 brouard 11842: /* 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 11843:
1.126 brouard 11844: /* Reads comments: lines beginning with '#' */
11845: while((c=getc(ficpar))=='#' && c!= EOF){
11846: ungetc(c,ficpar);
11847: fgets(line, MAXLINE, ficpar);
11848: numlinepar++;
1.141 brouard 11849: fputs(line,stdout);
1.126 brouard 11850: fputs(line,ficparo);
11851: fputs(line,ficlog);
11852: }
11853: ungetc(c,ficpar);
1.220 brouard 11854:
1.126 brouard 11855: matcov=matrix(1,npar,1,npar);
1.203 brouard 11856: hess=matrix(1,npar,1,npar);
1.131 brouard 11857: for(i=1; i <=npar; i++)
11858: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11859:
1.194 brouard 11860: /* Scans npar lines */
1.126 brouard 11861: for(i=1; i <=npar; i++){
1.226 brouard 11862: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11863: if(count != 3){
1.226 brouard 11864: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11865: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11866: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11867: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11868: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11869: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11870: exit(1);
1.220 brouard 11871: }else{
1.226 brouard 11872: if(mle==1)
11873: printf("%1d%1d%d",i1,j1,jk);
11874: }
11875: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11876: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11877: for(j=1; j <=i; j++){
1.226 brouard 11878: fscanf(ficpar," %le",&matcov[i][j]);
11879: if(mle==1){
11880: printf(" %.5le",matcov[i][j]);
11881: }
11882: fprintf(ficlog," %.5le",matcov[i][j]);
11883: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11884: }
11885: fscanf(ficpar,"\n");
11886: numlinepar++;
11887: if(mle==1)
1.220 brouard 11888: printf("\n");
1.126 brouard 11889: fprintf(ficlog,"\n");
11890: fprintf(ficparo,"\n");
11891: }
1.194 brouard 11892: /* End of read covariance matrix npar lines */
1.126 brouard 11893: for(i=1; i <=npar; i++)
11894: for(j=i+1;j<=npar;j++)
1.226 brouard 11895: matcov[i][j]=matcov[j][i];
1.126 brouard 11896:
11897: if(mle==1)
11898: printf("\n");
11899: fprintf(ficlog,"\n");
11900:
11901: fflush(ficlog);
11902:
11903: } /* End of mle != -3 */
1.218 brouard 11904:
1.186 brouard 11905: /* Main data
11906: */
1.290 brouard 11907: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11908: /* num=lvector(1,n); */
11909: /* moisnais=vector(1,n); */
11910: /* annais=vector(1,n); */
11911: /* moisdc=vector(1,n); */
11912: /* andc=vector(1,n); */
11913: /* weight=vector(1,n); */
11914: /* agedc=vector(1,n); */
11915: /* cod=ivector(1,n); */
11916: /* for(i=1;i<=n;i++){ */
11917: num=lvector(firstobs,lastobs);
11918: moisnais=vector(firstobs,lastobs);
11919: annais=vector(firstobs,lastobs);
11920: moisdc=vector(firstobs,lastobs);
11921: andc=vector(firstobs,lastobs);
11922: weight=vector(firstobs,lastobs);
11923: agedc=vector(firstobs,lastobs);
11924: cod=ivector(firstobs,lastobs);
11925: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11926: num[i]=0;
11927: moisnais[i]=0;
11928: annais[i]=0;
11929: moisdc[i]=0;
11930: andc[i]=0;
11931: agedc[i]=0;
11932: cod[i]=0;
11933: weight[i]=1.0; /* Equal weights, 1 by default */
11934: }
1.290 brouard 11935: mint=matrix(1,maxwav,firstobs,lastobs);
11936: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 11937: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
11938: printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel));
1.126 brouard 11939: tab=ivector(1,NCOVMAX);
1.144 brouard 11940: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11941: 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 11942:
1.136 brouard 11943: /* Reads data from file datafile */
11944: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11945: goto end;
11946:
11947: /* Calculation of the number of parameters from char model */
1.234 brouard 11948: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11949: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11950: k=3 V4 Tvar[k=3]= 4 (from V4)
11951: k=2 V1 Tvar[k=2]= 1 (from V1)
11952: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11953: */
11954:
11955: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11956: TvarsDind=ivector(1,NCOVMAX); /* */
11957: TvarsD=ivector(1,NCOVMAX); /* */
11958: TvarsQind=ivector(1,NCOVMAX); /* */
11959: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11960: TvarF=ivector(1,NCOVMAX); /* */
11961: TvarFind=ivector(1,NCOVMAX); /* */
11962: TvarV=ivector(1,NCOVMAX); /* */
11963: TvarVind=ivector(1,NCOVMAX); /* */
11964: TvarA=ivector(1,NCOVMAX); /* */
11965: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11966: TvarFD=ivector(1,NCOVMAX); /* */
11967: TvarFDind=ivector(1,NCOVMAX); /* */
11968: TvarFQ=ivector(1,NCOVMAX); /* */
11969: TvarFQind=ivector(1,NCOVMAX); /* */
11970: TvarVD=ivector(1,NCOVMAX); /* */
11971: TvarVDind=ivector(1,NCOVMAX); /* */
11972: TvarVQ=ivector(1,NCOVMAX); /* */
11973: TvarVQind=ivector(1,NCOVMAX); /* */
11974:
1.230 brouard 11975: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11976: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11977: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11978: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11979: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11980: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11981: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11982: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11983: */
11984: /* For model-covariate k tells which data-covariate to use but
11985: because this model-covariate is a construction we invent a new column
11986: ncovcol + k1
11987: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11988: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11989: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11990: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11991: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11992: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11993: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11994: */
1.145 brouard 11995: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11996: 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 11997: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11998: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11999: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 12000: 4 covariates (3 plus signs)
12001: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
12002: */
1.230 brouard 12003: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 12004: * individual dummy, fixed or varying:
12005: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
12006: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 12007: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
12008: * V1 df, V2 qf, V3 & V4 dv, V5 qv
12009: * Tmodelind[1]@9={9,0,3,2,}*/
12010: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
12011: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 12012: * individual quantitative, fixed or varying:
12013: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
12014: * 3, 1, 0, 0, 0, 0, 0, 0},
12015: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 12016: /* Main decodemodel */
12017:
1.187 brouard 12018:
1.223 brouard 12019: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 12020: goto end;
12021:
1.137 brouard 12022: if((double)(lastobs-imx)/(double)imx > 1.10){
12023: nbwarn++;
12024: 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);
12025: 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);
12026: }
1.136 brouard 12027: /* if(mle==1){*/
1.137 brouard 12028: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
12029: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 12030: }
12031:
12032: /*-calculation of age at interview from date of interview and age at death -*/
12033: agev=matrix(1,maxwav,1,imx);
12034:
12035: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
12036: goto end;
12037:
1.126 brouard 12038:
1.136 brouard 12039: agegomp=(int)agemin;
1.290 brouard 12040: free_vector(moisnais,firstobs,lastobs);
12041: free_vector(annais,firstobs,lastobs);
1.126 brouard 12042: /* free_matrix(mint,1,maxwav,1,n);
12043: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 12044: /* free_vector(moisdc,1,n); */
12045: /* free_vector(andc,1,n); */
1.145 brouard 12046: /* */
12047:
1.126 brouard 12048: wav=ivector(1,imx);
1.214 brouard 12049: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
12050: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
12051: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
12052: 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.*/
12053: bh=imatrix(1,lastpass-firstpass+2,1,imx);
12054: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 12055:
12056: /* Concatenates waves */
1.214 brouard 12057: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
12058: Death is a valid wave (if date is known).
12059: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
12060: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
12061: and mw[mi+1][i]. dh depends on stepm.
12062: */
12063:
1.126 brouard 12064: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 12065: /* Concatenates waves */
1.145 brouard 12066:
1.290 brouard 12067: free_vector(moisdc,firstobs,lastobs);
12068: free_vector(andc,firstobs,lastobs);
1.215 brouard 12069:
1.126 brouard 12070: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
12071: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
12072: ncodemax[1]=1;
1.145 brouard 12073: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 12074: cptcoveff=0;
1.220 brouard 12075: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
12076: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 12077: }
12078:
12079: ncovcombmax=pow(2,cptcoveff);
12080: invalidvarcomb=ivector(1, ncovcombmax);
12081: for(i=1;i<ncovcombmax;i++)
12082: invalidvarcomb[i]=0;
12083:
1.211 brouard 12084: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 12085: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 12086: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 12087:
1.200 brouard 12088: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 12089: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 12090: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 12091: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
12092: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
12093: * (currently 0 or 1) in the data.
12094: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
12095: * corresponding modality (h,j).
12096: */
12097:
1.145 brouard 12098: h=0;
12099: /*if (cptcovn > 0) */
1.126 brouard 12100: m=pow(2,cptcoveff);
12101:
1.144 brouard 12102: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 12103: * For k=4 covariates, h goes from 1 to m=2**k
12104: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
12105: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 12106: * h\k 1 2 3 4
1.143 brouard 12107: *______________________________
12108: * 1 i=1 1 i=1 1 i=1 1 i=1 1
12109: * 2 2 1 1 1
12110: * 3 i=2 1 2 1 1
12111: * 4 2 2 1 1
12112: * 5 i=3 1 i=2 1 2 1
12113: * 6 2 1 2 1
12114: * 7 i=4 1 2 2 1
12115: * 8 2 2 2 1
1.197 brouard 12116: * 9 i=5 1 i=3 1 i=2 1 2
12117: * 10 2 1 1 2
12118: * 11 i=6 1 2 1 2
12119: * 12 2 2 1 2
12120: * 13 i=7 1 i=4 1 2 2
12121: * 14 2 1 2 2
12122: * 15 i=8 1 2 2 2
12123: * 16 2 2 2 2
1.143 brouard 12124: */
1.212 brouard 12125: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 12126: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
12127: * and the value of each covariate?
12128: * V1=1, V2=1, V3=2, V4=1 ?
12129: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
12130: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
12131: * In order to get the real value in the data, we use nbcode
12132: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
12133: * We are keeping this crazy system in order to be able (in the future?)
12134: * to have more than 2 values (0 or 1) for a covariate.
12135: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
12136: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
12137: * bbbbbbbb
12138: * 76543210
12139: * h-1 00000101 (6-1=5)
1.219 brouard 12140: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 12141: * &
12142: * 1 00000001 (1)
1.219 brouard 12143: * 00000000 = 1 & ((h-1) >> (k-1))
12144: * +1= 00000001 =1
1.211 brouard 12145: *
12146: * h=14, k=3 => h'=h-1=13, k'=k-1=2
12147: * h' 1101 =2^3+2^2+0x2^1+2^0
12148: * >>k' 11
12149: * & 00000001
12150: * = 00000001
12151: * +1 = 00000010=2 = codtabm(14,3)
12152: * Reverse h=6 and m=16?
12153: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
12154: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
12155: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
12156: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
12157: * V3=decodtabm(14,3,2**4)=2
12158: * h'=13 1101 =2^3+2^2+0x2^1+2^0
12159: *(h-1) >> (j-1) 0011 =13 >> 2
12160: * &1 000000001
12161: * = 000000001
12162: * +1= 000000010 =2
12163: * 2211
12164: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
12165: * V3=2
1.220 brouard 12166: * codtabm and decodtabm are identical
1.211 brouard 12167: */
12168:
1.145 brouard 12169:
12170: free_ivector(Ndum,-1,NCOVMAX);
12171:
12172:
1.126 brouard 12173:
1.186 brouard 12174: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 12175: strcpy(optionfilegnuplot,optionfilefiname);
12176: if(mle==-3)
1.201 brouard 12177: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 12178: strcat(optionfilegnuplot,".gp");
12179:
12180: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
12181: printf("Problem with file %s",optionfilegnuplot);
12182: }
12183: else{
1.204 brouard 12184: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 12185: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 12186: //fprintf(ficgp,"set missing 'NaNq'\n");
12187: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 12188: }
12189: /* fclose(ficgp);*/
1.186 brouard 12190:
12191:
12192: /* Initialisation of --------- index.htm --------*/
1.126 brouard 12193:
12194: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
12195: if(mle==-3)
1.201 brouard 12196: strcat(optionfilehtm,"-MORT_");
1.126 brouard 12197: strcat(optionfilehtm,".htm");
12198: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 12199: printf("Problem with %s \n",optionfilehtm);
12200: exit(0);
1.126 brouard 12201: }
12202:
12203: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
12204: strcat(optionfilehtmcov,"-cov.htm");
12205: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
12206: printf("Problem with %s \n",optionfilehtmcov), exit(0);
12207: }
12208: else{
12209: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
12210: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12211: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 12212: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
12213: }
12214:
1.324 brouard 12215: 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 日本学術振興会 (<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 12216: <hr size=\"2\" color=\"#EC5E5E\"> \n\
12217: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 12218: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12219: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 12220: \n\
12221: <hr size=\"2\" color=\"#EC5E5E\">\
12222: <ul><li><h4>Parameter files</h4>\n\
12223: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
12224: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
12225: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
12226: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
12227: - Date and time at start: %s</ul>\n",\
12228: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
12229: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
12230: fileres,fileres,\
12231: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
12232: fflush(fichtm);
12233:
12234: strcpy(pathr,path);
12235: strcat(pathr,optionfilefiname);
1.184 brouard 12236: #ifdef WIN32
12237: _chdir(optionfilefiname); /* Move to directory named optionfile */
12238: #else
1.126 brouard 12239: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 12240: #endif
12241:
1.126 brouard 12242:
1.220 brouard 12243: /* Calculates basic frequencies. Computes observed prevalence at single age
12244: and for any valid combination of covariates
1.126 brouard 12245: and prints on file fileres'p'. */
1.251 brouard 12246: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 12247: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 12248:
12249: fprintf(fichtm,"\n");
1.286 brouard 12250: 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 12251: ftol, stepm);
12252: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
12253: ncurrv=1;
12254: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
12255: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
12256: ncurrv=i;
12257: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12258: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 12259: ncurrv=i;
12260: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12261: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 12262: ncurrv=i;
12263: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
12264: 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", \
12265: nlstate, ndeath, maxwav, mle, weightopt);
12266:
12267: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
12268: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
12269:
12270:
1.317 brouard 12271: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 12272: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
12273: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 12274: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 12275: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 12276: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12277: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12278: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12279: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 12280:
1.126 brouard 12281: /* For Powell, parameters are in a vector p[] starting at p[1]
12282: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
12283: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
12284:
12285: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 12286: /* For mortality only */
1.126 brouard 12287: if (mle==-3){
1.136 brouard 12288: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 12289: for(i=1;i<=NDIM;i++)
12290: for(j=1;j<=NDIM;j++)
12291: ximort[i][j]=0.;
1.186 brouard 12292: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 12293: cens=ivector(firstobs,lastobs);
12294: ageexmed=vector(firstobs,lastobs);
12295: agecens=vector(firstobs,lastobs);
12296: dcwave=ivector(firstobs,lastobs);
1.223 brouard 12297:
1.126 brouard 12298: for (i=1; i<=imx; i++){
12299: dcwave[i]=-1;
12300: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 12301: if (s[m][i]>nlstate) {
12302: dcwave[i]=m;
12303: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
12304: break;
12305: }
1.126 brouard 12306: }
1.226 brouard 12307:
1.126 brouard 12308: for (i=1; i<=imx; i++) {
12309: if (wav[i]>0){
1.226 brouard 12310: ageexmed[i]=agev[mw[1][i]][i];
12311: j=wav[i];
12312: agecens[i]=1.;
12313:
12314: if (ageexmed[i]> 1 && wav[i] > 0){
12315: agecens[i]=agev[mw[j][i]][i];
12316: cens[i]= 1;
12317: }else if (ageexmed[i]< 1)
12318: cens[i]= -1;
12319: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
12320: cens[i]=0 ;
1.126 brouard 12321: }
12322: else cens[i]=-1;
12323: }
12324:
12325: for (i=1;i<=NDIM;i++) {
12326: for (j=1;j<=NDIM;j++)
1.226 brouard 12327: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 12328: }
12329:
1.302 brouard 12330: p[1]=0.0268; p[NDIM]=0.083;
12331: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12332:
12333:
1.136 brouard 12334: #ifdef GSL
12335: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12336: #else
1.126 brouard 12337: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12338: #endif
1.201 brouard 12339: strcpy(filerespow,"POW-MORT_");
12340: strcat(filerespow,fileresu);
1.126 brouard 12341: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12342: printf("Problem with resultfile: %s\n", filerespow);
12343: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12344: }
1.136 brouard 12345: #ifdef GSL
12346: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12347: #else
1.126 brouard 12348: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12349: #endif
1.126 brouard 12350: /* for (i=1;i<=nlstate;i++)
12351: for(j=1;j<=nlstate+ndeath;j++)
12352: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12353: */
12354: fprintf(ficrespow,"\n");
1.136 brouard 12355: #ifdef GSL
12356: /* gsl starts here */
12357: T = gsl_multimin_fminimizer_nmsimplex;
12358: gsl_multimin_fminimizer *sfm = NULL;
12359: gsl_vector *ss, *x;
12360: gsl_multimin_function minex_func;
12361:
12362: /* Initial vertex size vector */
12363: ss = gsl_vector_alloc (NDIM);
12364:
12365: if (ss == NULL){
12366: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
12367: }
12368: /* Set all step sizes to 1 */
12369: gsl_vector_set_all (ss, 0.001);
12370:
12371: /* Starting point */
1.126 brouard 12372:
1.136 brouard 12373: x = gsl_vector_alloc (NDIM);
12374:
12375: if (x == NULL){
12376: gsl_vector_free(ss);
12377: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
12378: }
12379:
12380: /* Initialize method and iterate */
12381: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 12382: /* gsl_vector_set(x, 0, 0.0268); */
12383: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 12384: gsl_vector_set(x, 0, p[1]);
12385: gsl_vector_set(x, 1, p[2]);
12386:
12387: minex_func.f = &gompertz_f;
12388: minex_func.n = NDIM;
12389: minex_func.params = (void *)&p; /* ??? */
12390:
12391: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
12392: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
12393:
12394: printf("Iterations beginning .....\n\n");
12395: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
12396:
12397: iteri=0;
12398: while (rval == GSL_CONTINUE){
12399: iteri++;
12400: status = gsl_multimin_fminimizer_iterate(sfm);
12401:
12402: if (status) printf("error: %s\n", gsl_strerror (status));
12403: fflush(0);
12404:
12405: if (status)
12406: break;
12407:
12408: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12409: ssval = gsl_multimin_fminimizer_size (sfm);
12410:
12411: if (rval == GSL_SUCCESS)
12412: printf ("converged to a local maximum at\n");
12413:
12414: printf("%5d ", iteri);
12415: for (it = 0; it < NDIM; it++){
12416: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12417: }
12418: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12419: }
12420:
12421: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12422:
12423: gsl_vector_free(x); /* initial values */
12424: gsl_vector_free(ss); /* inital step size */
12425: for (it=0; it<NDIM; it++){
12426: p[it+1]=gsl_vector_get(sfm->x,it);
12427: fprintf(ficrespow," %.12lf", p[it]);
12428: }
12429: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12430: #endif
12431: #ifdef POWELL
12432: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12433: #endif
1.126 brouard 12434: fclose(ficrespow);
12435:
1.203 brouard 12436: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12437:
12438: for(i=1; i <=NDIM; i++)
12439: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12440: matcov[i][j]=matcov[j][i];
1.126 brouard 12441:
12442: printf("\nCovariance matrix\n ");
1.203 brouard 12443: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12444: for(i=1; i <=NDIM; i++) {
12445: for(j=1;j<=NDIM;j++){
1.220 brouard 12446: printf("%f ",matcov[i][j]);
12447: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12448: }
1.203 brouard 12449: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12450: }
12451:
12452: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12453: for (i=1;i<=NDIM;i++) {
1.126 brouard 12454: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12455: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12456: }
1.302 brouard 12457: lsurv=vector(agegomp,AGESUP);
12458: lpop=vector(agegomp,AGESUP);
12459: tpop=vector(agegomp,AGESUP);
1.126 brouard 12460: lsurv[agegomp]=100000;
12461:
12462: for (k=agegomp;k<=AGESUP;k++) {
12463: agemortsup=k;
12464: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12465: }
12466:
12467: for (k=agegomp;k<agemortsup;k++)
12468: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12469:
12470: for (k=agegomp;k<agemortsup;k++){
12471: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12472: sumlpop=sumlpop+lpop[k];
12473: }
12474:
12475: tpop[agegomp]=sumlpop;
12476: for (k=agegomp;k<(agemortsup-3);k++){
12477: /* tpop[k+1]=2;*/
12478: tpop[k+1]=tpop[k]-lpop[k];
12479: }
12480:
12481:
12482: printf("\nAge lx qx dx Lx Tx e(x)\n");
12483: for (k=agegomp;k<(agemortsup-2);k++)
12484: 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]);
12485:
12486:
12487: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12488: ageminpar=50;
12489: agemaxpar=100;
1.194 brouard 12490: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12491: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12492: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12493: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12494: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12495: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12496: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12497: }else{
12498: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12499: 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 12500: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12501: }
1.201 brouard 12502: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12503: stepm, weightopt,\
12504: model,imx,p,matcov,agemortsup);
12505:
1.302 brouard 12506: free_vector(lsurv,agegomp,AGESUP);
12507: free_vector(lpop,agegomp,AGESUP);
12508: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 12509: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12510: free_ivector(dcwave,firstobs,lastobs);
12511: free_vector(agecens,firstobs,lastobs);
12512: free_vector(ageexmed,firstobs,lastobs);
12513: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12514: #ifdef GSL
1.136 brouard 12515: #endif
1.186 brouard 12516: } /* Endof if mle==-3 mortality only */
1.205 brouard 12517: /* Standard */
12518: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12519: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12520: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12521: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12522: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12523: for (k=1; k<=npar;k++)
12524: printf(" %d %8.5f",k,p[k]);
12525: printf("\n");
1.205 brouard 12526: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12527: /* mlikeli uses func not funcone */
1.247 brouard 12528: /* for(i=1;i<nlstate;i++){ */
12529: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12530: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12531: /* } */
1.205 brouard 12532: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12533: }
12534: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12535: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12536: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12537: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12538: }
12539: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12540: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12541: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12542: for (k=1; k<=npar;k++)
12543: printf(" %d %8.5f",k,p[k]);
12544: printf("\n");
12545:
12546: /*--------- results files --------------*/
1.283 brouard 12547: /* 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 12548:
12549:
12550: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12551: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 12552: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12553:
12554: printf("#model= 1 + age ");
12555: fprintf(ficres,"#model= 1 + age ");
12556: fprintf(ficlog,"#model= 1 + age ");
12557: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
12558: </ul>", model);
12559:
12560: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
12561: fprintf(fichtm, "<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(ficres," + V%d ",Tvar[j]);
12572: fprintf(ficlog," + V%d ",Tvar[j]);
12573: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
12574: }else if(Typevar[j]==1) {
12575: printf(" + V%d*age ",Tvar[j]);
12576: fprintf(ficres," + V%d*age ",Tvar[j]);
12577: fprintf(ficlog," + V%d*age ",Tvar[j]);
12578: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
12579: }else if(Typevar[j]==2) {
12580: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12581: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12582: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12583: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12584: }
12585: }
12586: printf("\n");
12587: fprintf(ficres,"\n");
12588: fprintf(ficlog,"\n");
12589: fprintf(fichtm, "</tr>");
12590: fprintf(fichtm, "\n");
12591:
12592:
1.126 brouard 12593: for(i=1,jk=1; i <=nlstate; i++){
12594: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12595: if (k != i) {
1.319 brouard 12596: fprintf(fichtm, "<tr>");
1.225 brouard 12597: printf("%d%d ",i,k);
12598: fprintf(ficlog,"%d%d ",i,k);
12599: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 12600: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 12601: for(j=1; j <=ncovmodel; j++){
12602: printf("%12.7f ",p[jk]);
12603: fprintf(ficlog,"%12.7f ",p[jk]);
12604: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 12605: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 12606: jk++;
12607: }
12608: printf("\n");
12609: fprintf(ficlog,"\n");
12610: fprintf(ficres,"\n");
1.319 brouard 12611: fprintf(fichtm, "</tr>\n");
1.225 brouard 12612: }
1.126 brouard 12613: }
12614: }
1.319 brouard 12615: /* fprintf(fichtm,"</tr>\n"); */
12616: fprintf(fichtm,"</table>\n");
12617: fprintf(fichtm, "\n");
12618:
1.203 brouard 12619: if(mle != 0){
12620: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12621: ftolhess=ftol; /* Usually correct */
1.203 brouard 12622: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12623: 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");
12624: 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 12625: 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 12626: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
12627: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
12628: if(nagesqr==1){
12629: printf(" + age*age ");
12630: fprintf(ficres," + age*age ");
12631: fprintf(ficlog," + age*age ");
12632: fprintf(fichtm, "<th>+ age*age</th>");
12633: }
12634: for(j=1;j <=ncovmodel-2;j++){
12635: if(Typevar[j]==0) {
12636: printf(" + V%d ",Tvar[j]);
12637: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
12638: }else if(Typevar[j]==1) {
12639: printf(" + V%d*age ",Tvar[j]);
12640: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
12641: }else if(Typevar[j]==2) {
12642: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12643: }
12644: }
12645: fprintf(fichtm, "</tr>\n");
12646:
1.203 brouard 12647: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12648: for(k=1; k <=(nlstate+ndeath); k++){
12649: if (k != i) {
1.319 brouard 12650: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 12651: printf("%d%d ",i,k);
12652: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 12653: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 12654: for(j=1; j <=ncovmodel; j++){
1.319 brouard 12655: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 12656: printf("%12.7f(%12.7f) W=%8.3f CI=[%12.7f ; %12.7f] ",p[jk],sqrt(matcov[jk][jk]), p[jk]/sqrt(matcov[jk][jk]), p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
12657: fprintf(ficlog,"%12.7f(%12.7f) W=%8.3f CI=[%12.7f ; %12.7f] ",p[jk],sqrt(matcov[jk][jk]), p[jk]/sqrt(matcov[jk][jk]), p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
1.319 brouard 12658: if(fabs(wald) > 1.96){
1.321 brouard 12659: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 12660: }else{
12661: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
12662: }
1.324 brouard 12663: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 12664: 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 12665: jk++;
12666: }
12667: printf("\n");
12668: fprintf(ficlog,"\n");
1.319 brouard 12669: fprintf(fichtm, "</tr>\n");
1.225 brouard 12670: }
12671: }
1.193 brouard 12672: }
1.203 brouard 12673: } /* end of hesscov and Wald tests */
1.319 brouard 12674: fprintf(fichtm,"</table>\n");
1.225 brouard 12675:
1.203 brouard 12676: /* */
1.126 brouard 12677: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12678: printf("# Scales (for hessian or gradient estimation)\n");
12679: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12680: for(i=1,jk=1; i <=nlstate; i++){
12681: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12682: if (j!=i) {
12683: fprintf(ficres,"%1d%1d",i,j);
12684: printf("%1d%1d",i,j);
12685: fprintf(ficlog,"%1d%1d",i,j);
12686: for(k=1; k<=ncovmodel;k++){
12687: printf(" %.5e",delti[jk]);
12688: fprintf(ficlog," %.5e",delti[jk]);
12689: fprintf(ficres," %.5e",delti[jk]);
12690: jk++;
12691: }
12692: printf("\n");
12693: fprintf(ficlog,"\n");
12694: fprintf(ficres,"\n");
12695: }
1.126 brouard 12696: }
12697: }
12698:
12699: 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 12700: if(mle >= 1) /* To big for the screen */
1.126 brouard 12701: 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");
12702: 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");
12703: /* # 121 Var(a12)\n\ */
12704: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12705: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12706: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12707: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12708: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12709: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12710: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12711:
12712:
12713: /* Just to have a covariance matrix which will be more understandable
12714: even is we still don't want to manage dictionary of variables
12715: */
12716: for(itimes=1;itimes<=2;itimes++){
12717: jj=0;
12718: for(i=1; i <=nlstate; i++){
1.225 brouard 12719: for(j=1; j <=nlstate+ndeath; j++){
12720: if(j==i) continue;
12721: for(k=1; k<=ncovmodel;k++){
12722: jj++;
12723: ca[0]= k+'a'-1;ca[1]='\0';
12724: if(itimes==1){
12725: if(mle>=1)
12726: printf("#%1d%1d%d",i,j,k);
12727: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12728: fprintf(ficres,"#%1d%1d%d",i,j,k);
12729: }else{
12730: if(mle>=1)
12731: printf("%1d%1d%d",i,j,k);
12732: fprintf(ficlog,"%1d%1d%d",i,j,k);
12733: fprintf(ficres,"%1d%1d%d",i,j,k);
12734: }
12735: ll=0;
12736: for(li=1;li <=nlstate; li++){
12737: for(lj=1;lj <=nlstate+ndeath; lj++){
12738: if(lj==li) continue;
12739: for(lk=1;lk<=ncovmodel;lk++){
12740: ll++;
12741: if(ll<=jj){
12742: cb[0]= lk +'a'-1;cb[1]='\0';
12743: if(ll<jj){
12744: if(itimes==1){
12745: if(mle>=1)
12746: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12747: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12748: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12749: }else{
12750: if(mle>=1)
12751: printf(" %.5e",matcov[jj][ll]);
12752: fprintf(ficlog," %.5e",matcov[jj][ll]);
12753: fprintf(ficres," %.5e",matcov[jj][ll]);
12754: }
12755: }else{
12756: if(itimes==1){
12757: if(mle>=1)
12758: printf(" Var(%s%1d%1d)",ca,i,j);
12759: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12760: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12761: }else{
12762: if(mle>=1)
12763: printf(" %.7e",matcov[jj][ll]);
12764: fprintf(ficlog," %.7e",matcov[jj][ll]);
12765: fprintf(ficres," %.7e",matcov[jj][ll]);
12766: }
12767: }
12768: }
12769: } /* end lk */
12770: } /* end lj */
12771: } /* end li */
12772: if(mle>=1)
12773: printf("\n");
12774: fprintf(ficlog,"\n");
12775: fprintf(ficres,"\n");
12776: numlinepar++;
12777: } /* end k*/
12778: } /*end j */
1.126 brouard 12779: } /* end i */
12780: } /* end itimes */
12781:
12782: fflush(ficlog);
12783: fflush(ficres);
1.225 brouard 12784: while(fgets(line, MAXLINE, ficpar)) {
12785: /* If line starts with a # it is a comment */
12786: if (line[0] == '#') {
12787: numlinepar++;
12788: fputs(line,stdout);
12789: fputs(line,ficparo);
12790: fputs(line,ficlog);
1.299 brouard 12791: fputs(line,ficres);
1.225 brouard 12792: continue;
12793: }else
12794: break;
12795: }
12796:
1.209 brouard 12797: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12798: /* ungetc(c,ficpar); */
12799: /* fgets(line, MAXLINE, ficpar); */
12800: /* fputs(line,stdout); */
12801: /* fputs(line,ficparo); */
12802: /* } */
12803: /* ungetc(c,ficpar); */
1.126 brouard 12804:
12805: estepm=0;
1.209 brouard 12806: 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 12807:
12808: if (num_filled != 6) {
12809: 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);
12810: 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);
12811: goto end;
12812: }
12813: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12814: }
12815: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12816: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12817:
1.209 brouard 12818: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12819: if (estepm==0 || estepm < stepm) estepm=stepm;
12820: if (fage <= 2) {
12821: bage = ageminpar;
12822: fage = agemaxpar;
12823: }
12824:
12825: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12826: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12827: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12828:
1.186 brouard 12829: /* Other stuffs, more or less useful */
1.254 brouard 12830: while(fgets(line, MAXLINE, ficpar)) {
12831: /* If line starts with a # it is a comment */
12832: if (line[0] == '#') {
12833: numlinepar++;
12834: fputs(line,stdout);
12835: fputs(line,ficparo);
12836: fputs(line,ficlog);
1.299 brouard 12837: fputs(line,ficres);
1.254 brouard 12838: continue;
12839: }else
12840: break;
12841: }
12842:
12843: 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){
12844:
12845: if (num_filled != 7) {
12846: 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);
12847: 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);
12848: goto end;
12849: }
12850: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12851: 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);
12852: 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);
12853: 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 12854: }
1.254 brouard 12855:
12856: while(fgets(line, MAXLINE, ficpar)) {
12857: /* If line starts with a # it is a comment */
12858: if (line[0] == '#') {
12859: numlinepar++;
12860: fputs(line,stdout);
12861: fputs(line,ficparo);
12862: fputs(line,ficlog);
1.299 brouard 12863: fputs(line,ficres);
1.254 brouard 12864: continue;
12865: }else
12866: break;
1.126 brouard 12867: }
12868:
12869:
12870: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12871: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12872:
1.254 brouard 12873: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12874: if (num_filled != 1) {
12875: 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);
12876: 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);
12877: goto end;
12878: }
12879: printf("pop_based=%d\n",popbased);
12880: fprintf(ficlog,"pop_based=%d\n",popbased);
12881: fprintf(ficparo,"pop_based=%d\n",popbased);
12882: fprintf(ficres,"pop_based=%d\n",popbased);
12883: }
12884:
1.258 brouard 12885: /* Results */
1.307 brouard 12886: endishere=0;
1.258 brouard 12887: nresult=0;
1.308 brouard 12888: parameterline=0;
1.258 brouard 12889: do{
12890: if(!fgets(line, MAXLINE, ficpar)){
12891: endishere=1;
1.308 brouard 12892: parameterline=15;
1.258 brouard 12893: }else if (line[0] == '#') {
12894: /* If line starts with a # it is a comment */
1.254 brouard 12895: numlinepar++;
12896: fputs(line,stdout);
12897: fputs(line,ficparo);
12898: fputs(line,ficlog);
1.299 brouard 12899: fputs(line,ficres);
1.254 brouard 12900: continue;
1.258 brouard 12901: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12902: parameterline=11;
1.296 brouard 12903: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12904: parameterline=12;
1.307 brouard 12905: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 12906: parameterline=13;
1.307 brouard 12907: }
1.258 brouard 12908: else{
12909: parameterline=14;
1.254 brouard 12910: }
1.308 brouard 12911: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 12912: case 11:
1.296 brouard 12913: 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)){
12914: 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 12915: 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);
12916: 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);
12917: 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);
12918: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12919: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12920: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12921: prvforecast = 1;
12922: }
12923: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 12924: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12925: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12926: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 12927: prvforecast = 2;
12928: }
12929: else {
12930: 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);
12931: 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);
12932: goto end;
1.258 brouard 12933: }
1.254 brouard 12934: break;
1.258 brouard 12935: case 12:
1.296 brouard 12936: 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)){
12937: 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);
12938: 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);
12939: 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);
12940: 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);
12941: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12942: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12943: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12944: prvbackcast = 1;
12945: }
12946: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 12947: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12948: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12949: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 12950: prvbackcast = 2;
12951: }
12952: else {
12953: 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);
12954: 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);
12955: goto end;
1.258 brouard 12956: }
1.230 brouard 12957: break;
1.258 brouard 12958: case 13:
1.307 brouard 12959: num_filled=sscanf(line,"result:%[^\n]\n",resultline);
12960: nresult++; /* Sum of resultlines */
12961: printf("Result %d: result:%s\n",nresult, resultline);
1.318 brouard 12962: if(nresult > MAXRESULTLINESPONE-1){
12963: 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);
12964: 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 12965: goto end;
12966: }
1.310 brouard 12967: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 12968: fprintf(ficparo,"result: %s\n",resultline);
12969: fprintf(ficres,"result: %s\n",resultline);
12970: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 12971: } else
12972: goto end;
1.307 brouard 12973: break;
12974: case 14:
12975: printf("Error: Unknown command '%s'\n",line);
12976: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 12977: if(line[0] == ' ' || line[0] == '\n'){
12978: printf("It should not be an empty line '%s'\n",line);
12979: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
12980: }
1.307 brouard 12981: if(ncovmodel >=2 && nresult==0 ){
12982: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
12983: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12984: }
1.307 brouard 12985: /* goto end; */
12986: break;
1.308 brouard 12987: case 15:
12988: printf("End of resultlines.\n");
12989: fprintf(ficlog,"End of resultlines.\n");
12990: break;
12991: default: /* parameterline =0 */
1.307 brouard 12992: nresult=1;
12993: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 12994: } /* End switch parameterline */
12995: }while(endishere==0); /* End do */
1.126 brouard 12996:
1.230 brouard 12997: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12998: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12999:
13000: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 13001: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 13002: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13003: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13004: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 13005: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13006: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13007: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13008: }else{
1.270 brouard 13009: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 13010: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
13011: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
13012: if(prvforecast==1){
13013: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
13014: jprojd=jproj1;
13015: mprojd=mproj1;
13016: anprojd=anproj1;
13017: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
13018: jprojf=jproj2;
13019: mprojf=mproj2;
13020: anprojf=anproj2;
13021: } else if(prvforecast == 2){
13022: dateprojd=dateintmean;
13023: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
13024: dateprojf=dateintmean+yrfproj;
13025: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
13026: }
13027: if(prvbackcast==1){
13028: datebackd=(jback1+12*mback1+365*anback1)/365;
13029: jbackd=jback1;
13030: mbackd=mback1;
13031: anbackd=anback1;
13032: datebackf=(jback2+12*mback2+365*anback2)/365;
13033: jbackf=jback2;
13034: mbackf=mback2;
13035: anbackf=anback2;
13036: } else if(prvbackcast == 2){
13037: datebackd=dateintmean;
13038: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
13039: datebackf=dateintmean-yrbproj;
13040: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
13041: }
13042:
13043: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 13044: }
13045: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 13046: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
13047: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 13048:
1.225 brouard 13049: /*------------ free_vector -------------*/
13050: /* chdir(path); */
1.220 brouard 13051:
1.215 brouard 13052: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
13053: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
13054: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
13055: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 13056: free_lvector(num,firstobs,lastobs);
13057: free_vector(agedc,firstobs,lastobs);
1.126 brouard 13058: /*free_matrix(covar,0,NCOVMAX,1,n);*/
13059: /*free_matrix(covar,1,NCOVMAX,1,n);*/
13060: fclose(ficparo);
13061: fclose(ficres);
1.220 brouard 13062:
13063:
1.186 brouard 13064: /* Other results (useful)*/
1.220 brouard 13065:
13066:
1.126 brouard 13067: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 13068: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
13069: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 13070: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 13071: fclose(ficrespl);
13072:
13073: /*------------- h Pij x at various ages ------------*/
1.180 brouard 13074: /*#include "hpijx.h"*/
13075: hPijx(p, bage, fage);
1.145 brouard 13076: fclose(ficrespij);
1.227 brouard 13077:
1.220 brouard 13078: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 13079: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 13080: k=1;
1.126 brouard 13081: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 13082:
1.269 brouard 13083: /* Prevalence for each covariate combination in probs[age][status][cov] */
13084: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13085: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 13086: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 13087: for(k=1;k<=ncovcombmax;k++)
13088: probs[i][j][k]=0.;
1.269 brouard 13089: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
13090: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 13091: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 13092: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13093: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 13094: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 13095: for(k=1;k<=ncovcombmax;k++)
13096: mobaverages[i][j][k]=0.;
1.219 brouard 13097: mobaverage=mobaverages;
13098: if (mobilav!=0) {
1.235 brouard 13099: printf("Movingaveraging observed prevalence\n");
1.258 brouard 13100: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 13101: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
13102: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
13103: printf(" Error in movingaverage mobilav=%d\n",mobilav);
13104: }
1.269 brouard 13105: } else if (mobilavproj !=0) {
1.235 brouard 13106: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 13107: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 13108: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
13109: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
13110: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
13111: }
1.269 brouard 13112: }else{
13113: printf("Internal error moving average\n");
13114: fflush(stdout);
13115: exit(1);
1.219 brouard 13116: }
13117: }/* end if moving average */
1.227 brouard 13118:
1.126 brouard 13119: /*---------- Forecasting ------------------*/
1.296 brouard 13120: if(prevfcast==1){
13121: /* /\* if(stepm ==1){*\/ */
13122: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13123: /*This done previously after freqsummary.*/
13124: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
13125: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
13126:
13127: /* } else if (prvforecast==2){ */
13128: /* /\* if(stepm ==1){*\/ */
13129: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13130: /* } */
13131: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
13132: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 13133: }
1.269 brouard 13134:
1.296 brouard 13135: /* Prevbcasting */
13136: if(prevbcast==1){
1.219 brouard 13137: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13138: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13139: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13140:
13141: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
13142:
13143: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 13144:
1.219 brouard 13145: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
13146: fclose(ficresplb);
13147:
1.222 brouard 13148: hBijx(p, bage, fage, mobaverage);
13149: fclose(ficrespijb);
1.219 brouard 13150:
1.296 brouard 13151: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
13152: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
13153: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
13154: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
13155: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
13156: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
13157:
13158:
1.269 brouard 13159: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13160:
13161:
1.269 brouard 13162: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 13163: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13164: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13165: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 13166: } /* end Prevbcasting */
1.268 brouard 13167:
1.186 brouard 13168:
13169: /* ------ Other prevalence ratios------------ */
1.126 brouard 13170:
1.215 brouard 13171: free_ivector(wav,1,imx);
13172: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
13173: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
13174: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 13175:
13176:
1.127 brouard 13177: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 13178:
1.201 brouard 13179: strcpy(filerese,"E_");
13180: strcat(filerese,fileresu);
1.126 brouard 13181: if((ficreseij=fopen(filerese,"w"))==NULL) {
13182: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13183: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13184: }
1.208 brouard 13185: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
13186: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 13187:
13188: pstamp(ficreseij);
1.219 brouard 13189:
1.235 brouard 13190: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13191: if (cptcovn < 1){i1=1;}
13192:
13193: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13194: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13195: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13196: continue;
1.219 brouard 13197: fprintf(ficreseij,"\n#****** ");
1.235 brouard 13198: printf("\n#****** ");
1.225 brouard 13199: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 13200: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 13201: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13202: }
13203: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13204: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13205: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 13206: }
13207: fprintf(ficreseij,"******\n");
1.235 brouard 13208: printf("******\n");
1.219 brouard 13209:
13210: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13211: oldm=oldms;savm=savms;
1.235 brouard 13212: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 13213:
1.219 brouard 13214: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 13215: }
13216: fclose(ficreseij);
1.208 brouard 13217: printf("done evsij\n");fflush(stdout);
13218: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 13219:
1.218 brouard 13220:
1.227 brouard 13221: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 13222:
1.201 brouard 13223: strcpy(filerest,"T_");
13224: strcat(filerest,fileresu);
1.127 brouard 13225: if((ficrest=fopen(filerest,"w"))==NULL) {
13226: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
13227: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
13228: }
1.208 brouard 13229: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
13230: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 13231: strcpy(fileresstde,"STDE_");
13232: strcat(fileresstde,fileresu);
1.126 brouard 13233: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 13234: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
13235: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 13236: }
1.227 brouard 13237: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
13238: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 13239:
1.201 brouard 13240: strcpy(filerescve,"CVE_");
13241: strcat(filerescve,fileresu);
1.126 brouard 13242: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 13243: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
13244: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 13245: }
1.227 brouard 13246: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
13247: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 13248:
1.201 brouard 13249: strcpy(fileresv,"V_");
13250: strcat(fileresv,fileresu);
1.126 brouard 13251: if((ficresvij=fopen(fileresv,"w"))==NULL) {
13252: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
13253: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
13254: }
1.227 brouard 13255: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
13256: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 13257:
1.235 brouard 13258: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13259: if (cptcovn < 1){i1=1;}
13260:
13261: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13262: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13263: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13264: continue;
1.321 brouard 13265: printf("\n# model %s \n#****** Result for:", model);
13266: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
13267: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.227 brouard 13268: for(j=1;j<=cptcoveff;j++){
13269: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13270: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13271: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13272: }
1.235 brouard 13273: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13274: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13275: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13276: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13277: }
1.208 brouard 13278: fprintf(ficrest,"******\n");
1.227 brouard 13279: fprintf(ficlog,"******\n");
13280: printf("******\n");
1.208 brouard 13281:
13282: fprintf(ficresstdeij,"\n#****** ");
13283: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 13284: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 13285: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13286: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 13287: }
1.235 brouard 13288: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13289: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13290: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13291: }
1.208 brouard 13292: fprintf(ficresstdeij,"******\n");
13293: fprintf(ficrescveij,"******\n");
13294:
13295: fprintf(ficresvij,"\n#****** ");
1.238 brouard 13296: /* pstamp(ficresvij); */
1.225 brouard 13297: for(j=1;j<=cptcoveff;j++)
1.227 brouard 13298: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 13299: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13300: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13301: }
1.208 brouard 13302: fprintf(ficresvij,"******\n");
13303:
13304: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13305: oldm=oldms;savm=savms;
1.235 brouard 13306: printf(" cvevsij ");
13307: fprintf(ficlog, " cvevsij ");
13308: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 13309: printf(" end cvevsij \n ");
13310: fprintf(ficlog, " end cvevsij \n ");
13311:
13312: /*
13313: */
13314: /* goto endfree; */
13315:
13316: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13317: pstamp(ficrest);
13318:
1.269 brouard 13319: epj=vector(1,nlstate+1);
1.208 brouard 13320: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 13321: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
13322: cptcod= 0; /* To be deleted */
13323: printf("varevsij vpopbased=%d \n",vpopbased);
13324: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 13325: 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 13326: 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 ");
13327: if(vpopbased==1)
13328: 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);
13329: else
1.288 brouard 13330: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13331: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
13332: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
13333: fprintf(ficrest,"\n");
13334: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 13335: printf("Computing age specific forward period (stable) prevalences in each health state \n");
13336: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13337: for(age=bage; age <=fage ;age++){
1.235 brouard 13338: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 13339: if (vpopbased==1) {
13340: if(mobilav ==0){
13341: for(i=1; i<=nlstate;i++)
13342: prlim[i][i]=probs[(int)age][i][k];
13343: }else{ /* mobilav */
13344: for(i=1; i<=nlstate;i++)
13345: prlim[i][i]=mobaverage[(int)age][i][k];
13346: }
13347: }
1.219 brouard 13348:
1.227 brouard 13349: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
13350: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
13351: /* printf(" age %4.0f ",age); */
13352: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
13353: for(i=1, epj[j]=0.;i <=nlstate;i++) {
13354: epj[j] += prlim[i][i]*eij[i][j][(int)age];
13355: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
13356: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
13357: }
13358: epj[nlstate+1] +=epj[j];
13359: }
13360: /* printf(" age %4.0f \n",age); */
1.219 brouard 13361:
1.227 brouard 13362: for(i=1, vepp=0.;i <=nlstate;i++)
13363: for(j=1;j <=nlstate;j++)
13364: vepp += vareij[i][j][(int)age];
13365: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
13366: for(j=1;j <=nlstate;j++){
13367: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
13368: }
13369: fprintf(ficrest,"\n");
13370: }
1.208 brouard 13371: } /* End vpopbased */
1.269 brouard 13372: free_vector(epj,1,nlstate+1);
1.208 brouard 13373: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
13374: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 13375: printf("done selection\n");fflush(stdout);
13376: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 13377:
1.235 brouard 13378: } /* End k selection */
1.227 brouard 13379:
13380: printf("done State-specific expectancies\n");fflush(stdout);
13381: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
13382:
1.288 brouard 13383: /* variance-covariance of forward period prevalence*/
1.269 brouard 13384: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13385:
1.227 brouard 13386:
1.290 brouard 13387: free_vector(weight,firstobs,lastobs);
1.227 brouard 13388: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 13389: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
13390: free_matrix(anint,1,maxwav,firstobs,lastobs);
13391: free_matrix(mint,1,maxwav,firstobs,lastobs);
13392: free_ivector(cod,firstobs,lastobs);
1.227 brouard 13393: free_ivector(tab,1,NCOVMAX);
13394: fclose(ficresstdeij);
13395: fclose(ficrescveij);
13396: fclose(ficresvij);
13397: fclose(ficrest);
13398: fclose(ficpar);
13399:
13400:
1.126 brouard 13401: /*---------- End : free ----------------*/
1.219 brouard 13402: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 13403: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
13404: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 13405: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
13406: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 13407: } /* mle==-3 arrives here for freeing */
1.227 brouard 13408: /* endfree:*/
13409: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
13410: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
13411: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 13412: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
13413: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
13414: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
13415: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 13416: free_matrix(matcov,1,npar,1,npar);
13417: free_matrix(hess,1,npar,1,npar);
13418: /*free_vector(delti,1,npar);*/
13419: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13420: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 13421: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 13422: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13423:
13424: free_ivector(ncodemax,1,NCOVMAX);
13425: free_ivector(ncodemaxwundef,1,NCOVMAX);
13426: free_ivector(Dummy,-1,NCOVMAX);
13427: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 13428: free_ivector(DummyV,1,NCOVMAX);
13429: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 13430: free_ivector(Typevar,-1,NCOVMAX);
13431: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 13432: free_ivector(TvarsQ,1,NCOVMAX);
13433: free_ivector(TvarsQind,1,NCOVMAX);
13434: free_ivector(TvarsD,1,NCOVMAX);
13435: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 13436: free_ivector(TvarFD,1,NCOVMAX);
13437: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 13438: free_ivector(TvarF,1,NCOVMAX);
13439: free_ivector(TvarFind,1,NCOVMAX);
13440: free_ivector(TvarV,1,NCOVMAX);
13441: free_ivector(TvarVind,1,NCOVMAX);
13442: free_ivector(TvarA,1,NCOVMAX);
13443: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 13444: free_ivector(TvarFQ,1,NCOVMAX);
13445: free_ivector(TvarFQind,1,NCOVMAX);
13446: free_ivector(TvarVD,1,NCOVMAX);
13447: free_ivector(TvarVDind,1,NCOVMAX);
13448: free_ivector(TvarVQ,1,NCOVMAX);
13449: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 13450: free_ivector(Tvarsel,1,NCOVMAX);
13451: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 13452: free_ivector(Tposprod,1,NCOVMAX);
13453: free_ivector(Tprod,1,NCOVMAX);
13454: free_ivector(Tvaraff,1,NCOVMAX);
13455: free_ivector(invalidvarcomb,1,ncovcombmax);
13456: free_ivector(Tage,1,NCOVMAX);
13457: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 13458: free_ivector(TmodelInvind,1,NCOVMAX);
13459: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 13460:
13461: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
13462: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 13463: fflush(fichtm);
13464: fflush(ficgp);
13465:
1.227 brouard 13466:
1.126 brouard 13467: if((nberr >0) || (nbwarn>0)){
1.216 brouard 13468: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
13469: 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 13470: }else{
13471: printf("End of Imach\n");
13472: fprintf(ficlog,"End of Imach\n");
13473: }
13474: printf("See log file on %s\n",filelog);
13475: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 13476: /*(void) gettimeofday(&end_time,&tzp);*/
13477: rend_time = time(NULL);
13478: end_time = *localtime(&rend_time);
13479: /* tml = *localtime(&end_time.tm_sec); */
13480: strcpy(strtend,asctime(&end_time));
1.126 brouard 13481: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
13482: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 13483: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 13484:
1.157 brouard 13485: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
13486: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
13487: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 13488: /* printf("Total time was %d uSec.\n", total_usecs);*/
13489: /* if(fileappend(fichtm,optionfilehtm)){ */
13490: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13491: fclose(fichtm);
13492: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13493: fclose(fichtmcov);
13494: fclose(ficgp);
13495: fclose(ficlog);
13496: /*------ End -----------*/
1.227 brouard 13497:
1.281 brouard 13498:
13499: /* Executes gnuplot */
1.227 brouard 13500:
13501: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 13502: #ifdef WIN32
1.227 brouard 13503: if (_chdir(pathcd) != 0)
13504: printf("Can't move to directory %s!\n",path);
13505: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 13506: #else
1.227 brouard 13507: if(chdir(pathcd) != 0)
13508: printf("Can't move to directory %s!\n", path);
13509: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 13510: #endif
1.126 brouard 13511: printf("Current directory %s!\n",pathcd);
13512: /*strcat(plotcmd,CHARSEPARATOR);*/
13513: sprintf(plotcmd,"gnuplot");
1.157 brouard 13514: #ifdef _WIN32
1.126 brouard 13515: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
13516: #endif
13517: if(!stat(plotcmd,&info)){
1.158 brouard 13518: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13519: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 13520: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 13521: }else
13522: strcpy(pplotcmd,plotcmd);
1.157 brouard 13523: #ifdef __unix
1.126 brouard 13524: strcpy(plotcmd,GNUPLOTPROGRAM);
13525: if(!stat(plotcmd,&info)){
1.158 brouard 13526: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13527: }else
13528: strcpy(pplotcmd,plotcmd);
13529: #endif
13530: }else
13531: strcpy(pplotcmd,plotcmd);
13532:
13533: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 13534: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 13535: strcpy(pplotcmd,plotcmd);
1.227 brouard 13536:
1.126 brouard 13537: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 13538: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 13539: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 13540: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 13541: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 13542: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 13543: strcpy(plotcmd,pplotcmd);
13544: }
1.126 brouard 13545: }
1.158 brouard 13546: printf(" Successful, please wait...");
1.126 brouard 13547: while (z[0] != 'q') {
13548: /* chdir(path); */
1.154 brouard 13549: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 13550: scanf("%s",z);
13551: /* if (z[0] == 'c') system("./imach"); */
13552: if (z[0] == 'e') {
1.158 brouard 13553: #ifdef __APPLE__
1.152 brouard 13554: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 13555: #elif __linux
13556: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 13557: #else
1.152 brouard 13558: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13559: #endif
13560: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13561: system(pplotcmd);
1.126 brouard 13562: }
13563: else if (z[0] == 'g') system(plotcmd);
13564: else if (z[0] == 'q') exit(0);
13565: }
1.227 brouard 13566: end:
1.126 brouard 13567: while (z[0] != 'q') {
1.195 brouard 13568: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13569: scanf("%s",z);
13570: }
1.283 brouard 13571: printf("End\n");
1.282 brouard 13572: exit(0);
1.126 brouard 13573: }
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